England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT
The Hon Mr Justice Davis
C08C00806

		Royal Courts of Justice Strand, London, WC2A 2LL
		5th May 2010

Lord Justice Patten :

Introduction

1. The first defendant, bmibaby Limited ("BMIB"), is the well-known airline which operates low cost carrier services to various European destinations. It is a subsidiary of the second defendant, British Midland Airways Limited ("BM"). The claimant, Durham Tees Valley Airport Limited ("DTVAL"), is, as its name suggests, the owner of Durham Tees Valley Airport ("the Airport") (formerly known as Teesside International Airport) near Darlington. Until it changed its name in line with the change of name of the Airport the company was called Teesside International Airport Limited ("TIAL").
2. On 23rd April 2003 TIAL entered into what has become known as the Base Agreement. The other party was British Midland Regional Limited ("BMRL") which traded as bmibaby until the subsequent incorporation of BMIB. The Base Agreement was the culmination of discussions and subsequently negotiations which took place throughout 2002. When they began TIAL was owned by local authority shareholders who were keen to re-generate and develop the Airport in order to boost the local economy. As of 2000 the Airport carried about 700,000 passengers per annum but had capacity for at least 1-2 million passengers. The local authorities were, however, limited in what they could afford to invest both by the financial constraints imposed upon them and by the difficulties involved in obtaining commercial loans. To overcome these problems they decided to seek new investment by offering for sale a controlling 75% interest in TIAL. Offers were invited and, after protracted negotiations, the successful bidder was Peel Airports Limited ("Peel").
3. Peel was the owner and operator of two other regional airports; Liverpool (John Lennon) Airport and Robin Hood Airport near Doncaster. In both cases it had successfully increased the turnover and profitability of the airports by increasing passenger volumes. This was due in large measure to being able to attract low cost carriers to use the airports. The evidence was that Peel's approach to airport operation was what was described as long-term and that it had been successful in persuading easyJet to enter into a 20 year contract to use Liverpool Airport.
4. Davis J (in a typically careful and detailed judgment [2009] EWHC 852 (Ch)) has set out the history of the negotiations which led ultimately to Peel entering into a Subscription and Shareholders' Agreement with the local authority shareholders on 1^st April 2003 and then to the Base Agreement between TIAL and BMRL. For the purposes of this appeal I need do no more than outline the judge's principal findings.
5. BMRL first established a base for bmibaby at East Midlands Airport in February 2002. It was then looking to establish further bases at other regional airports in order to expand its operations. At the same time TIAL and Peel were keen to attract a low cost carrier with a based operation at the Airport. A based operation means one under which the airline keeps its aircraft at the airport overnight at the conclusion of its daytime operations. This enables the airport to present the airline as based at the airport and to enjoy the economic benefits of increased passenger throughput and income which the flight patterns (or "rotations") of based aircraft bring.
6. TIAL was invited by BMRL to make a presentation to it on 1^st May 2002 with a view to the Airport becoming a base for bmibaby. It was asked to proceed on various assumptions including "up to two units for winter 2002" and "a minimum of three months for summer 2003". In the judge's words BMRL held all the cards in relation to the negotiations. TIAL was extremely anxious to secure the Airport as a base for bmibaby and faced competition from other airports including Newcastle. The Airport did, however, have some advantages for bmibaby in its capacity for expansion and the lack of competition from any rival carriers.
7. Although TIAL was invited to make a further presentation on 9^th September, Manchester was in fact selected as the third base for bmibaby's operations. But BMRL remained interested in the Airport as a potential base in the North East. Its internal financial evaluation recommended the Airport in preference to Newcastle as being potentially more profitable. On the assumption that two aircraft would be based there, a net operating profit of £378,000 was projected for 2003; £2.145 million for 2004 and £1.792 million for 2005.
8. By this time Peel had been identified by TIAL as a potential investor and was promoted as part of its negotiations with BMRL. At a meeting with representatives of BMRL in October 2002 TIAL were told that the airline required a commitment to re-brand the Airport and for additional funding to be made available for the development of the terminal. They were also looking for £750,000 in marketing support and for funding for training support. TIAL said that, in return, they were looking for what they described as a long-term deal.
9. By November 2002 all parties were proceeding on the assumption that Peel would take control of TIAL as and when an agreement was concluded with BMRL. On 28^th March 2003 Peel sent to BMRL a letter which was intended to form a collateral agreement to its acquisition of a majority stake in TIAL. So far as material it stated:-

"In the event that Peel Airports Limited ("Peel") acquires a shareholding in Teesside International Airport Limited ("TIAL") or otherwise acquires operational control of Teesside International Airport ("the Airport"), then in consideration of British Midland Regional Limited trading as bmibaby ("bmibaby") entering into an agreement with TIAL for the operation of services by bmibaby from the Airport no later than 31 October 2003 in the form annexed to this letter (or with such further amendments as are agreed by us), Peel agrees and undertakes as follows:-

1. Peel will pay to bmibaby the sum of £500,000 in the form of 9 monthly instalments to commence on or before 7 April 2003, or if later, within 7 days of acquisition of such shareholding or control as a contribution to route support and costs associated with bmibaby setting up its operation at the Airport. The first instalment will be £100,000 followed by 8 monthly payments of £50,000 over the period from May to December 2003.

In the event, however, that bmibaby fails to commence based operations in accordance with the contract with TIAL or alternatively, withdraws services from the Airport within twelve months of the date of commencement of services, then in either scenario, Peel shall be entitled to recover such monies paid to bmibaby under this letter (as shall not have been spent on marketing and launch costs by bmibaby) should bmibaby fail to prove the use of reasonable commercial endeavours in the development of the base operation at the Airport. Additionally, if such withdrawal occurs prior to all payments having been made, then for the avoidance of doubt, no further payments remain due and owing from Peel to bmibaby."

10. On 1^st April 2003 Peel entered into the Subscription and Shareholders' Agreement and thereby acquired a controlling interest in TIAL. On 23^rd April the Base Agreement was signed. The judge set out the material parts of the Agreement and his construction of certain key terms at paragraphs 24-31 of his judgment:-

"24. The Base Agreement, self-evidently, is not what one might style a lawyers' agreement (I was told that it was drafted by the business people concerned, albeit reviewed by in-house legal departments). It is nevertheless detailed and, on its face, a formal contract. The named parties are TIAL and British Midland Regional Limited. Somewhat inconveniently it has no numbered clauses.

25. The opening provisions are these:-

"Date: 23 April 2003

Airline: British Midland Regional Limited trading as bmibaby

Operation: Initial 'lead-in' flying programme (to an agreed number of destinations) to commence no later than 31 October 2003 to support the establishment of a minimum x 2 based aircraft operation (initially B737) operating exclusively from TIAL by Summer 2004.

Capacity: Capacity will be determined by the relevant route network delivered in consultation with TIAL.

Duration: 10 years from the establishment of the based aircraft operation as defined above. …"

26. There are then further provisions, including some quite technical details on charging structure. The charging structure, geared to load factor, involved low passenger charges in the early years, rising thereafter steadily up to year 7, with increases thereafter to year 10 to reflect the Retail Price Index. Under the heading "Marketing: Base Establishment – Initial 2 based aircraft operation" this was provided:-

"TIAL will pay to bmibaby the sum of £850k in the form of marketing support on the condition that bmibaby establish a 2 based aircraft operation at TIAL by Summer 2004. Delivery of this funding will be made via the Airport as a payment to bmibaby in the form of a contribution towards an approved Marketing Plan supplied by the airline, designed to promote the region. For a winter 2003 launch of services by bmibaby, this funding will be delivered 50% by 14 May 2003 and 50% by 31 July 2003.

In the event that bmibaby fail to commence operations or withdraw from TIAL within 12 months of launch, then TIAL reserves the right to clawback all such funding as shall not have been spent on marketing and launch costs by bmibaby should bmibaby fail to prove the use of best endeavours in the development of the base operation.

In addition to the above sum in the event that Peel Airports Limited or another investor acquire a shareholding in TIAL or control of the airport, £500k will be paid to bmibaby by Peel (as a contribution towards route support and costs associated with bmibaby setting up its operations at the airport) or such other investor in the form of 9 monthly instalments from, on or before 30 April 2003 to 1 December 2003 inclusive.

In the event, however, that bmibaby fails to commence based operations in accordance with the contract with TIAL or alternatively, withdraws services from the Airport within twelve months of the date of commencement of services, then in either scenario, Peel shall be entitled to recover such monies paid to bmibaby (as shall not have been spent on marketing and launch costs by bmibaby) should bmibaby fail to prove the use of reasonable commercial endeavours in the development of the base operation at the Airport. Additionally, if such withdrawal occurs prior to all payments having been made, then for the avoidance of doubt, no further payments remain due and owing from Peel to bmibaby. …"

It is not disputed that the sums payable by TIAL would primarily derive from external local government funding.

27. There were then various other provisions. By way of Tourist Board Funding, TIAL agreed to pay to bmibaby the sum of £100,000 on the terms specified, including payment in two tranches for "a winter 2003 launch of services". Route Development support was agreed to be provided in the form of a cash payment by TIAL per destination as follows:- Year 1 £50,000, Year 2 £25,000, on certain expressed assumptions, including an assured minimum of five frequencies a week to any destination and full consultation with TIAL on the route planning process. It was also provided that "this structure of route development support will be paid to bmibaby for each and every new destination that the airline introduces from TIAL during the term of the deal."

28. Under the heading "Training" it was provided that TIAL would pay to bmibaby the sum of £5,000 at the commencement of each bmibaby job created; and, prior to commencement of a base operation, a further £200,000 of training funds. Other provisions related to payment for aircraft parking; a fuel rebate "during the term of this deal"; a provision whereby TIAL would provide bmibaby with a package of advertising sites at the airport to a current market value of "no less that £100,000 in year 1" with contributions to advertising requirements "from year 2 onwards at a rate to be agreed"; free staff car parking and security passes (to a maximum number) for bmibaby staff "for a period of two years from the establishment of a base operation".

29. After various other provisions, this was provided under the heading "Development":-

"bmibaby has identified to TIAL certain potential destinations and cities it may wish to serve during the term of this agreement (see list attached at Schedule 1). TIAL confirms that it will not actively pursue further development of any specific scheduled destination, (as defined in Schedule 1), currently being operated, to any bmibaby base during the term of this deal.

From time to time, should TIAL be approached by another airline to operate to any destination within the existing bmibaby network at that time, then TIAL will discuss with bmibaby whether it has any plans to operate the route from TIAL within a 12 month period from that date. If declined then TIAL may proceed with negotiations outwith the terms of this agreement, but not on more favourable terms than those offered to bmibaby.

Similarly, (but save where TIAL is already bound by existing agreements with incumbent operators), TIAL will not offer more favourable discounts or incentives to another carrier on those routes operated by bmibaby from TIAL, nor offer a more favourable charging structure to any other carrier at the Airport.

For the avoidance of doubt, all existing route networks with incumbent operators at TIAL at the date of this agreement are specifically excluded and TIAL reserves the right to agree any such future terms as deemed appropriate with these operators on such existing routes from time to time."

The appended schedule contained a lengthy list of destinations: it was common ground before me that, by aviation industry standards, it was an unusually lengthy list for such a purpose and whereby DTVA was (if the routes were operated by bmibaby) restricted in its dealings with other airlines.

30. I should say something about the word "exclusively" as used in the "Operation" designation in the Base Agreement (and, subsequently in Clause 2 of the Novation and Variation Agreement). This, as the expert witnesses agreed, was a well understood meaning in the aviation industry. It connotes that a based aircraft will always fly from and return to its base. It is possible and quite common (in the absence of agreement to the contrary) for an aircraft, for example, to fly, taking the airport base as A, a route pattern of A ? B ? C ? B ? A (called a "W" pattern in the industry). But an exclusive pattern generally is A ? B ? A ? B ? A: all flights are thus either into or out of the airport. That of course increases the number of flights ("rotations") into and out of the base airport: and so has attractions for an airport. In addition, as I was told, it has also efficiency and financial advantages for the airline, if volume is there, because it exposes the airport to more passengers.

31. A different concept is "exclusivity". Under aviation law, putting it very generally, an airport cannot grant a monopoly on any route to an airline: another licensed airline cannot, generally speaking, be prevented, assuming slots are available, from operating such a route from that airport. Exclusivity, which commonly includes an agreement by an airport not to solicit other airlines thus connotes that a particular airline has a favoured deal in terms of passenger rates charged: another airline seeking to fly that route may be charged the full published rates by the airport. That concept is reflected in the terms of the "Development" clause in the Base Agreement and, in due course, in Clause 5, headed "Development and Exclusivity", of the Novation and Variation Agreement."

11. Both sides accept that the Base Agreement created a binding contract but they differ on how it should be construed. The principal issue before the judge was whether it imposed on BMRL a legal obligation to base and fly two aircraft from the Airport for a period of 10 years from Summer 2004 or whether (as the defendants contend) the agreement merely gave them a right but no obligation to do so. The issue is, however, complicated by the events which followed culminating in the execution of a Novation and Variation Agreement ("NVA") on 23^rd December 2005 under which BMIB took over the outstanding obligations of BMRL under the Base Agreement and revised terms were agreed in relation to the establishment of a two aircraft based operation at the Airport.
12. As provided by the Base Agreement, marketing steps were taken to publicise the BMIB operation and the agreed tranches of the £750,000 which was to be provided by TIAL and the £500,000 to be provided by Peel were duly paid. The name of the Airport was also changed and TIAL became DTVAL.
13. BMIB also commenced the lead in operation referred to in the Base Agreement using an aircraft based elsewhere but in October 2003 DTVAL were informed that, due to crewing difficulties and other problems, only one aircraft would be based at the Airport by March 2004. This led to what the judge described as protracted negotiations between the parties and expressions of surprise and disappointment on the part of DTVAL that, notwithstanding the re-branding of the airport and some £925,000 of financial support provided to BMIB for the operation, only one rather than two aircraft would be based at the Airport.
14. In the event BMIB commenced a based operation in April 2004 using a single aircraft flying a variety of routes. By November 2004 the airline had decided that it would not base a second aircraft at the Airport in 2005 but would proceed to establish a base operation at Birmingham Airport and to increase operations from its other bases including Manchester.
15. By then DTVAL had begun to accuse BMIB of reneging on its contractual commitments. BMIB's response was to ask for time. Its internal projections then showed that it predicted a loss of £2.8 million for its operations from the Airport in 2005 using two aircraft. But at the same time it continued to tell DTVAL in correspondence that it remained committed to meeting its contractual obligation "to base and operate a second aircraft" at the Airport.
16. Various proposals emerged but it is not suggested that these or, for that matter, the terms of the earlier correspondence provide a reliable or even admissible guide to the construction of the NVA which was signed on 23^rd December 2005. As the judge observed, this agreement shows every sign of having being professionally prepared. The parties were DTVAL, BMRL, BMIB and BM. The recitals recorded that the agreement was supplemental to the Base Agreement and that the parties had agreed to novate the original agreement to BMIB which had now taken over the operation of bmibaby from BMRL. BM was to act as the guarantor of its obligations. Recital (4) recorded that certain terms of the original agreement were also to be varied.
17. So far as material, the NVA provided as follows:-

"1. Novation

1.1 In consideration of the covenants and conditions contained in this Novation and Variation Agreement and with effect from the date of this Novation and Variation Agreement ("the Novation Date"):

(a) bmibaby agrees with DTVAL to assume all the obligations of BMRL to DTVAL under or arising from the original Agreement from the Novation Date except to the extent that such obligations have been fully and properly discharged prior to the Novation Date;

(b) DTVAL releases and discharges BMRL from all its obligations and liability to DTVAL under the original Agreement and from all (if any) liability to DTVAL (whether in contract or in tort or otherwise) arising from any act or omission of BMRL in the discharge or purported discharge of all or any of BMRL's obligations to DTVAL under the original Agreement; and

(c) DTVAL agrees with bmibaby to be bound by its obligations under the original Agreement (to the extent that they have not been fully and properly discharged prior to the Novation Date) in every way as if bmibaby had been a party to the original Agreement in place of BMRL and agrees that bmibaby shall enjoy the rights of BMRL in respect of the original Agreement in every way as if bmibaby had been a party to the original Agreement in place of BMRL.

1.2 The parties acknowledge and agree that:

(a) any obligations of BMRL owed to DTVAL pursuant to the terms of the original Agreement which remain undischarged at the Novation Date shall not be deemed to be waived by DTVAL by virtue of the novation of the original Agreement pursuant to the terms of this Novation and Variation Agreement, but shall be performed by bmibaby in accordance with clause 1.1(a);

(b) any obligations of DTVAL pursuant to the terms of the original Agreement which remain undischarged at the Novation Date shall not be deemed to be waived by bmibaby by virtue of the novation of the original Agreement pursuant to the terms of this Novation and Variation Agreement, but shall be performed by DTVAL in accordance with clause 1.1(c); and

(c) the novation of the original Agreement pursuant to the terms of this Novation and Variation Agreement shall be without prejudice to any rights accrued prior to the date of this Novation and Variation Agreement with the intent that such rights shall be enforceable by or against (as the case may be) bmibaby in place of BMRL.

(d) each party shall (at its own cost) execute such further deeds and documents and do all such things as the other parties reasonably require to give effect to the novation contained in this Novation and Variation Agreement.

2. Operation

bmibaby agrees to the establishment of a second based aircraft operation (initially B737) operating exclusively from Durham Tees Valley Airport ("DTVA") by Spring 2006 (to commence no later than 30 April 2006). For the avoidance of doubt, therefore, from Spring 2006 bmibaby will support a x2 based aircraft operation operating exclusively from DTVA.

3. Commercial

For the avoidance of doubt:

(a) the terms and charging structure will remain as set out in the original Agreement; and

(b) the term of the original Agreement shall expire on 30 April 2014.

4. Marketing

4.1 Base establishment/Tourist Board funding of £475,000 will be delivered from DTVAL to bmibaby within 7 days following the launch of the second based aircraft. Delivery of this funding will be made as a payment to bmibaby in the form of a contribution towards an approved Marketing Plan supplied by bmibaby, designed to promote the region. The clawback provision referred to in the original Agreement will continue to apply in relation to this funding (but for avoidance of doubt clawback does not apply to any funding paid in relation to the first based aircraft).

4.2 DTVAL will procure that Peel Airports Limited will contribute the sum of £100,000 to bmibaby's development funding such payment to be made within 7 days following the establishment of the second based aircraft. Again, the clawback provision will continue to apply in relation to this funding provision in respect of the second based aircraft only.

4.3 Route Development Support being £50,000 in year 1 and £25,000 in year 2 for each new destination shall continue to apply as set out in the original Agreement.

4.4 DTVAL will pay to bmibaby the sum of £5,000 at commencement of each bmibaby direct job created and the parties will continue to co-operate on facilitating any application by DTVAL to recover some or all of such payments from third party sources of funding.

4.5 DTVAL will use all reasonable endeavours to obtain and pay to bmibaby a further £100,000 for training funds from the Learning and Skills Council or such other organisation as may provide such funding on behalf of bmibaby such provision to replace the existing provision in the original agreement. However the parties acknowledge that they must liaise closely to ensure that any application criteria are strictly adhered to in order to seek to obtain such funding.

5. Development and Exclusivity

5.1 In relation to the destinations referred to at Schedule 1 of the original Agreement bmibaby only now accepts exclusivity in relation to those destination airports operated by it as specified below and not any of the other destinations referred to in that schedule.

Malaga, Alicante, Palma, London Gatwick, Paris-Charles de Gaulle, Jersey, Cork, Newquay, Knock, Bordeaux

For the avoidance of doubt such route exclusivity is agreed based on a minimum of 3 rotations per week to such defined destination airports.

5.2 The parties agree to liaise and consult diligently in order for them to agree the summer 2006 flying programmes to be operated by bmibaby from DTVA to reach agreement on the extent of destinations to be served by bmibaby for this period by the end of November 2005.

5.3 Excluding the destinations as defined in 4.1 [sic] DTVAL is entitled to pursue all other route development opportunities with other carriers as it considers appropriate.

5.4 For the avoidance of doubt the parties confirm that all terms of the original Agreement remain in full force and effect save as varied by the terms set out in this Novation and Variation Agreement."

18. At the time of the agreement BMIB's financial projections indicated that an operation based on two aircraft would in 2006 make a net operating loss in excess of £2 million. But, notwithstanding this, a second aircraft (a Boeing 737) was based at the Airport and was in operation on 26^th April 2006. The sums referred to in clause 4 were also duly paid by DTVAL and Peel. Thereafter things did not improve. The new managing director of BMIB appointed in May 2006 conducted a review of the airline's operations that was unfavourable to the Airport. His early preference was to close the operation during the winter and then to redeploy the aircraft in 2007. But he later concluded that to operate from the Airport at all was uncommercial and that it should cease to be a BMIB base. Losses were projected of up to £3.2 million in 2006 and £2.5 million in 2007. The decision to withdraw was taken in August 2006 and made public in September. The operations ceased by 6^th November 2006 and the aircraft were removed. The defendants calculate that, after giving credit for the funding provided by DTVAL and Peel (£2.177 million), the operation incurred a total loss for them of some £7.2 million between 2003 and 2006.

The claim

19. The proceedings were issued on 12^th March 2008. The original claimants were DTVAL and Peel and the action included a claim for £475,000 and £100,000 respectively under the clawback provisions contained in clauses 4.1 and 4.2 of the NVA. The claim form was subsequently amended, as a result of which these claims were deleted and Peel ceased to be a party. The action continued to trial as a claim by DTVAL alone for damages for breach of contract.
20. The pleaded case was that the reference to "Summer 2004" in the Base Agreement meant the six month period commencing on the last weekend in March 2004. The phrase "including B737" meant a Boeing 737 aircraft or its equivalent in size and capacity. On this basis BMIB (having taken over the liabilities and obligations of BMRL) was liable in damages for not having commenced the operation of the first aircraft until the end of April 2004; for having operated only one aircraft between May 2004 and March 2006; and for having no aircraft operating at all after November 2006. The only period for which no claim is made is between March and November 2006 when two aircraft were in operation.
21. The damages claimed are for loss of income from two sources. The first part of the claim was in respect of the charges payable per departing passenger under the formula contained in the Base Agreement; the second is the loss of income from duty free and other sales in the airport terminal and from catering and parking receipts. The pleading asserts that both types of income fall to be assessed by reference to the number of passengers who would have used the Airport but for the breaches of contract alleged.
22. DTVAL's primary case was that the provisions of the Base Agreement and clause 2 of the NVA imposed upon BMIB an obligation:-

"… to use its best endeavours to ensure that its operations from the Airport in respect of the 2 aircraft based there exclusively were performed to the maximum capacity in respect of the number of flights and the number of passengers per flight and that the maximum number of flights capable of being achieved were achieved and that the maximum number of passengers capable of being secured to fly on its aircraft and use of its services were secured."

This was pleaded both as an express and as an implied term.

23. In the alternative, it was alleged by way of amendment in paragraph 13.4 that the two agreements were subject to an implied term that BMIB (and previously BMRL) would operate the two aircraft based at the Airport in a manner which was reasonable in all the circumstances.
24. Correspondingly it is pleaded in paragraph 21.2 of the amended particulars of claim that damages should be assessed on the basis that the contract would have been performed in accordance with one or other of these terms.
25. As mentioned earlier, the defendants' position has always been (and remains) that the two agreements both individually and cumulatively were permissive rather than mandatory in relation to the operation of the aircraft. In the case of the Base Agreement, they contend that the permissive nature of the contract extends even to the establishment of the operation itself but they have to accept in the light of the wording of clause 2 of the NVA that they were under an obligation to establish a "x 2 based aircraft operation" at the Airport by 30^th April 2006. This did not extend, they say, to actually flying the aircraft.
26. In addition to these issues of construction they also took a number of points which go to the enforceability of one or both of the agreements. If the cumulative effect of the two agreements was to impose upon BMIB and BMRL a positive obligation to fly the planes commercially then the contract, they say, was too vague and uncertain to enforce. A promise to fly is insufficient in itself to enable the court to determine the measure of the obligation. It raises but does not answer the question of how many flights there should be. Neither agreement specified any minimum performance obligation or any other objective criteria which would allow the court to determine the level of operation necessary to constitute performance of the contract as opposed to a breach of it.
27. The second line of defence was that the NVA amounted to an unenforceable agreement to agree. This is based on clause 5.2 which requires the parties to liaise and consult in order to agree the flying programme for Summer 2006 by November 2005. Although this date had passed by the time that the NVA was executed, the defendants contended that the clause extends beyond 2006 to all future years of the contract.
28. The amendment to the particulars of claim and the reliance by DTVAL at trial on the implied term pleaded in paragraph 13.4 was really a response to the defendants' attack on the contractual arrangements on grounds of uncertainty. To some extent it played into their hands by appearing to recognise the difficulties which an unqualified obligation to operate might create. As the judge observed and I think Mr Brealey recognised, the reliance on the terms set out in paragraph 10 of the amended particulars of claim (see paragraph 22 above) as an express term of the contracts was negated by a simple reading of the two documents and it was not pursued either before the judge or before us as an implied term. What was pursued at the trial was the new and alternative case based on an alleged contractual duty to operate the two aircraft in a reasonable manner.

The judgment

(A) Construction

29. Davis J was not assisted by much of the evidence directed to what should be taken to be the commercial sense of the words used in the two agreements. So far as relevant, he had evidence that long-term contracts such as the one involving easyJet at Liverpool did occur in the market and were not therefore commercially unthinkable as contended by the defendants. But, perhaps not surprisingly, this led him to conclude that there was no standard or set form of agreement in this context and that each contract ultimately depended upon the terms agreed in the individual case. Neither side's case on construction could be said to be lacking in what Lord Diplock described in Antaios Cia Naviera SA v Salen Rederierna AB [1985] AC 191 as business common sense.
30. The judged therefore based himself upon what he took to be the ordinary and reasonable meaning of the words used subject to some input from the experts as to terms such as "based aircraft" and "operating exclusively" which it was common ground have particular meanings in the aviation industry. For the same reason, Mr Crane QC has also concentrated on the language of the documents and we have therefore been spared an unnecessary trawl through any background material beyond the facts found by the judge none of which is under challenge on this appeal.
31. The judge held that the Base Agreement did impose on BMRL (and therefore on BMIB) a positive obligation both to establish a two aircraft operation at the Airport and to operate the aircraft in the sense of flying them commercially. He accepted DTVAL's case that operate meant fly; that "initially B737" meant a Boeing 737 aircraft or its equivalent; and that "by Summer 2004" meant by (i.e. before) the period commencing on the last Sunday in March and ending on the last Saturday in October. The operation had therefore to be in place by 1^st April 2004.
32. Mr Crane accepted that operate means fly and there was no real challenge on this appeal to the judge's construction of the other two terms I have referred to. It seems to me that he was obviously right about the phrase "initially B737" and that, in the absence of any indications to the contrary, the period of Summer should be taken to begin with the start of British Summer Time. The real issue between the parties in relation to the Base Agreement is whether the judge was right about the mandatory nature of the provisions about the "Operation".
33. The judge's view was that the requirement to commence a lead-in flying programme by no later than 31^st October 2003 to support the establishment of the two aircraft base operation "operating exclusively from TIAL by Summer 2004" constituted, as he put it, the language of obligation. The treatment of these provisions as permissive was also inconsistent in his view with a continuing obligation on the part of TIAL to offer the preferential rates of charge set out in the agreement. It seems to me that the judge was right about this. What BMRL was required to do by the dates specified was to establish "a minimum x 2 based aircraft operation … operating exclusively from TIAL". The contract does not therefore distinguish between the basing of the two aircraft at the Airport and their operation and any ambiguity is resolved by the stipulation that the aircraft should be "operating exclusively" from the Airport by Summer 2004. On the judge's construction of the word "operate", which is accepted, this points more naturally to an obligation rather than a permission to fly.
34. Mr Crane's construction of the Base Agreement treats these provisions as amounting to an obligation to pay at the specified rates if and when the aircraft fly. Consistently with that, he contends that the agreement did not even impose on his clients an obligation to base the aircraft at the Airport. That seems to me to be at odds with the time limits specified in the agreement. If the two aircraft operation did not have to be up and running at all by Summer 2004 it would have been sufficient merely to have stated a 10 year period within which BMRL would be able (if it wished) to make use of the Airport as a base for two of its aircraft. What the parties in fact agreed was that the 10 year period should commence from the establishment of the based aircraft operation "as defined above" which is a reference back to the time limit of the start of Summer 2004.
35. More promising for the purposes of Mr Crane's argument are the clawback provisions which permitted TIAL and Peel to recover any of the monies provided for marketing which remained unspent at the time when BMRL either failed to "commence operations" or withdrew from the Airport within 12 months of the launch. It is said that these rights to recovery are not conditional in terms on the prescribed events amounting to a breach of the agreement. They therefore indicate that BMRL was free either not to commence operations or to withdraw its operations from the Airport and, consistently with that, give to TIAL and Peel a contractual right to recoup monies which would have been recoverable as damages had the failure to commence the operation or its subsequent discontinuance amounted to a breach of contract.
36. This construction of the clawback provisions is, as the judge observed, inconsistent with the reference to BMRL failing to commence based operations "in accordance with the contract with TIAL". But, more generally, the presence of specific clawback provisions dealing with the return of unspent funding is not, in my view, necessarily inconsistent with the existence of a contractual obligation to operate the two aircraft, a breach of which sounds in damages. The clawback provisions are limited and specific to the recovery of initial expenditure, some of which was provided by a company (Peel) which was not a party to the Base Agreement. As such, they operate in addition to any remedies which the law provides in favour of DTVAL. In the context of an agreement drawn up by businessmen rather than lawyers, they are not, in my view, sufficient to convert the otherwise clear language of the provisions about the base operation into the type of licence for which Mr Crane contends.
37. As mentioned earlier, the defendants accept that clause 2 of the NVA does impose on them an obligation at least to base two aircraft at the Airport from 30^th April 2006. But this understates what the contract in terms provides. As in the case of the Base Agreement, what BMIB agreed to do was to establish at the Airport "a second based aircraft operation". That is a composite phrase which includes not only the location of the two aircraft but also their operation. Like the judge, I cannot see how one can sensibly attribute to the parties an intention to require two passenger aircraft to be kept at the Airport but with no corresponding obligation to use them. That makes no commercial sense at all. The concluding words of clause 2 seem to me to put the matter beyond doubt.
38. The same goes for the provisions of clauses 5.1 and 5.3. In the Base Agreement TIAL agreed that it would not actively pursue the development of the routes to the destinations specified in Schedule 1. These were destinations within the existing bmibaby network at the time. As Mr Crane pointed out, TIAL could not in fact grant BMRL a monopoly of use of such routes from the Airport to the exclusion of all other carriers because to do so would be anti-competitive and involve a breach of the company's obligations under the 2000 Air Navigation Order (SI 2000 No. 1562). But to the extent that TIAL was able to promise not actively to encourage other carriers to compete on the same routes, it did so. Again for an airport operator anxious to develop the Airport this makes no sense if, in return, it was to receive no guarantee that BMRL or later BMIB would operate there at all.
39. It is convenient at this stage to deal with a subsidiary point about the effect of the NVA in relation to the obligation to commence the use of two aircraft. Under the Base Agreement that obligation arose with effect from 1^st April 2004 and the claim for damages includes the use of one rather than two aircraft between May 2004 and March 2006. This claim is maintained notwithstanding the provisions of clause 2 of the NVA under which BMIB agreed to operate a second aircraft "no later than 30^th April 2006". What is said is that this did not release the existing claim for breach of the Base Agreement which, by the date of the NVA, had subsisted for some 21 months. Mr Brealey relied on clause 1.2(c) of the NVA which expressly preserves "any rights accrued" prior to the date of the NVA and makes them enforceable against BMIB in place of BMRL.
40. Although the judge was of a different view, I consider that the breach of contract consisting of a failure to operate a second aircraft from the Airport by 1^st April 2004 was waived or released by the NVA as part of the variation of the terms of the Base Agreement to impose an obligation to base and operate two aircraft there no later than 30^th April 2006. DTVAL's case seems to contemplate that BMIB (by process of novation for BMRL) would remain in breach of contract under the Base Agreement notwithstanding that its only contractual obligation after 23^rd December 2005 was to base a second aircraft at the Airport by April 2006. Not only does this seem to me to be at odds with the evident intention of the parties to establish a new timetable for the two aircraft operation; it is also, I believe, inconsistent with the provisions of clause 1 of the agreement.
41. Under clause 1.1 BMIB agreed to assume all the obligations of BMRL which had not as of then been fully and properly discharged. In return, DTVAL released BMRL from "all its obligations and liability" to DTVAL under the Base Agreement. On one view the obligation to base two aircraft at the Airport by 1^st April 2004 (being time limited) did not remain to be discharged. The date had passed and BMRL was in breach of contract. Its only obligation to DTVAL was one to pay damages. If this is right then that liability was released and discharged by clause 1.1(b) and replaced by a new contractual obligation under clause 2 to base and operate two aircraft by 30^th April 2006. Clause 1.2(c) deals not with the liability of BMRL in damages for prior breaches of contract but with accrued contractual rights which, as a result of the novation, become enforceable against BMIB.
42. But if this is the wrong analysis and the contractual obligation to base and operate two aircraft by 1^st April 2004 remained intact and therefore constituted an obligation within the meaning of clauses 1.1(a) and 1.2(a), the burden of which passed to BMIB, it must follow that it was varied by clause 2 in the manner indicated. It is not possible, in my view, to regard the NVA as imposing as of 3^rd December 2005 two inconsistent contractual obligations and clause 5.4 of the NVA provided in terms that the Base Agreement remained in force save as varied by the terms set out in the NVA. What clause 2 did was in effect to extend the time for establishing the two aircraft operation in place of the original 2004 deadline. For these reasons, any claim for damages in respect of the failure to base a second aircraft at the Airport is limited, in my view, to the period after November 2006.

(B) Certainty

43. Having dealt with these questions of construction, the judge summarised his conclusions on that issue in these terms:-

"82. Thus I conclude that on the terms of the Novation and Variation Agreement bmibaby was obliged to establish by 30 April 2006, and thereafter to continue to base and to operate two aircraft: and this obligation was for the term of the contract, defined by Clause 3(b) as expiring on 30 April 2014. Such a conclusion is, in my view, consistent moreover with the factual matrix and also gives rise to (objectively viewed) a commercially sensible result."

44. He then turned to the questions of enforceability raised by the defendants. I can deal first with the argument that the NVA amounted to an unenforceable agreement to agree. This was based on clause 5.2 of the NVA which required the parties to liaise and consult in order "to agree the Summer 2006 flying programme to be operated by bmibaby from DTVA to reach agreement on the extent of destinations to be served by bmibaby for this period by the end of November 2005".
45. Insofar as the argument was based on clause 5.2, the judge was right to reject it. Agreement on the 2005 flight programme was not a pre-condition to the NVA as a whole and has, I think, to be read very much in the context of clause 5 which is concerned with determining route exclusivity. By the time of the NVA the date for agreement had, in any event, passed and there is nothing in clause 5.2 which imposes any similar obligation to agree a flying programme for the subsequent years covered by the agreement. The clause 5.2 point does not therefore affect either the enforceability of the NVA or its construction. But it does leave one with a contract under which the parties have not in terms specified how many flights are to be undertaken as part of the operation which BMIB has to conduct or prescribes a formula by reference to which that matter can either be ascertained or agreed.
46. This therefore leads to the defendants' alternative case on enforceability which is that the contract was too uncertain absent some stated or objectively ascertainable measure of performance which governed the number of flights which BMIB was required to operate in order to comply with the contract. The judge did not begin his analysis of this issue by asking himself whether the terms of the contract as construed by him and set out in paragraph 82 of his judgment represented a contract which the court could enforce. Instead he embarked on an analysis of whether it was possible to imply into the NVA the terms pleaded by DTVAL. On one view the appellant's reliance in its pleading on an implied term about the mode of operation of the aircraft could be regarded (and, in reality I think, was regarded by the judge) as an admission that the express terms as construed by him did not amount to an enforceable contract. But Mr Brealey disavowed this both before the judge and on this appeal. He submitted that once the judge had construed the contract and found it to have been breached by the withdrawal of both aircraft after November 2006 he should simply have directed an inquiry as to damages. He further contended that, given that there had been a complete repudiation of the NVA, the court did not have to determine what BMIB's minimum obligations were. That and other issues might arise in relation to the quantification of damages but that was not an answer to liability. He accepted, however, that as part of that inquiry the court would have to determine how many flights there would have been in order to assess passenger numbers and thereby determine the loss of income suffered by the appellant. And in his skeleton argument on this appeal he recognises (as the judge did) that the implication of a suitable term which incorporates an ascertainable measure of performance would largely remove the problems of enforceability and quantification which would otherwise arise.
47. The judge concluded that it was not possible to imply into the NVA the term pleaded in paragraph 13.4 of the amended particulars of claim which, by the end of the trial, was the only implied term relied upon. His principal reason was that an implied obligation to operate the two aircraft in the manner which was reasonable in all the circumstances did not resolve the question of how many flights BMIB was required to operate in order to comply with the agreement. Although a criterion of reasonableness may remove uncertainty in some cases (e.g. a tenancy under which the tenant agrees to pay a reasonable rent), these are largely cases in which objective criteria exist to enable the court to determine what would be reasonable in terms of quantity or price. In the present case no such criteria exist. The assessment of damages would be impossible for the same reason. The court would be unable to specify what level of performance was necessary in order to constitute compliance with the contract. The judge therefore dismissed the action.
48. This line of reasoning amounts to a finding that the contract was void for uncertainty. Although the implied term is relied on as a solution to problems inherent in the assessment of damages, those problems are largely consequential on the court's inability to identify the contractual obligation which has to be performed. The judge does not say in clear terms that the consequence of his rejection of the implied term is a finding of uncertainty and Mr Brealey has questioned in his submissions what the judge did actually decide. But in paragraph 113 of his judgment the judge refers to the submissions made by Mr Shah on behalf of the respondents as follows:-

"I was referred also to the Privy Council case of Australian Blue Metal Limited v Hughes [1963] AC 74. That was a case concerning a permission to extract minerals from an area of land. The case was, of course, one decided on its own circumstances. But it is relevant that, in delivering the judgment of the Board, Lord Devlin also said this at p.94:-

"The second feature is that no express obligation was imposed on the appellants to do any mining at all, and in their Lordships' opinion none can be implied. The only practicable way of framing such an obligation with sufficient precision to make it enforceable is to do what was done by the parties in the 1942 agreement and specify a minimum quantity of material that has to be won in a given period. Their Lordships were referred to Hillas & Co Ltd v Arcos Ltd, a case in which the House of Lords was able to use the implication of reasonableness to fill the gaps left by the parties. But in the present case there are no criteria which would enable a court of law to determine what would be a reasonable quantity. There would be too many uncertain factors to be taken into account, such as the profitability of mining in the future … ."

Mr Shah says that is likewise so in the present case: even granted that here is an express obligation to base and fly aircraft (as I have concluded there was) the obligation is insufficiently imprecise, there is no requirement for a minimum number of flights or passengers, there are too many uncertain factors as between parties with competing interests and the defect cannot be met by an implication of reasonableness. There is, in my view, obvious force in these submissions."

49. Similarly he rejected Mr Brealey's argument that, regardless of the implied term, liability could be established and an assessment ordered on the basis that by November 2006 BMIB had repudiated the NVA:-

"90. Mr Brealey's starting point was beguilingly simple. The express obligation on bmibaby, contained in the contracts, was to establish and operate two based aircraft at DTVA and to continue to do so for the 10 year period. In November 2006 bmibaby conclusively breached the contracts by withdrawing the two based aircraft from DTVA. Therefore, he says, liability is established: and the matter simply becomes one of quantifying the damages: albeit he accepted that "difficult issues", as he put it, of quantification could arise.

91. But such an approach only operates to defer to another forum a decision which has to be made in any event and is appropriately made at this stage. For the question still has to be asked: what is it that, contractually, bmibaby was obliged to do? Until that is established, an assessment of whether there is a breach causative of loss cannot be made. And to assert that bmibaby was obliged to base and operate two aircraft from DTVA over 10 years immediately raises the question of the nature and extent of its obligations: how many flights? how many passengers? and so on.

92. Mr Brealey accepted that this had to be decided. Indeed the point was pleaded at the outset fully in the Particulars of Claim, no doubt just because it was appreciated that it was an essential element of the claim….".

50. Although it could perhaps have been more clearly stated, it is, I think, implicit in these passages that the judge did accept the defendants' case on uncertainty and concluded that the contract was unenforceable. DTVAL therefore appeals on two grounds. The first is that even without the implied term the contract, as construed by the judge, was enforceable and damages could therefore be assessed. The second is that the judge was wrong to find that it was legally impermissible to imply the term contended for.
51. The judge was, in my view, clearly right in the passages I have quoted to regard the first question as being whether the contract was or was not enforceable with or without the implied term.
52. He was also right, I think, to consider questions of enforceability both in relation to what the contract requires the parties to do and in relation to whether the court can apply the rules for assessing damages in relation to an agreement in those terms. For a contract to be enforceable the court has to be able to say whether any particular standard of performance is or is not a breach of contact and, if a breach exists, to determine what measure of damages is appropriate. The latter issue raises an interesting question of whether damages for breach of contract fall to be assessed by reference to the minimum level of performance permitted under the contract. If so, it is obviously necessary to be able to specify and determine what that is. But, regardless of the outcome of that question, I accept Mr Crane's submission that the court cannot sidestep the issue of enforceability by reference to what is necessary for purposes of assessing damages. It still has to ask itself the prior question of whether the terms of the contract are too vague. It cannot order an assessment of damages in respect of an unenforceable contract.
53. This brings me therefore to the first ground of appeal. The judge's assessment of whether the contract was enforceable absent the implied term is largely confined to the passage in paragraphs 90-92 of his judgment quoted above. This is largely explicable in the light of Mr Brealey's reliance on an implied term in his client's pleading and his acceptance in argument that it was necessary for the court to decide how many flights and passengers there should have been. The judge therefore embarked on a detailed examination of the law on implied terms and concluded that the term contended for could not be implied and did not, in any event, resolve the problem of uncertainty created by the absence of any minimum standards of performance. Having reached that conclusion he treated the issue of certainty as resolved in favour of the respondents.
54. The first issue therefore on this appeal is whether he was right to regard the express terms of the NVA as too uncertain to be capable of enforcement by the court. If that issue is resolved in favour of DTVAL then reliance on any implied term becomes unnecessary. For a contract to be binding its terms must be "sufficiently definite to enable the court to give it a practical meaning. Its terms must be so definite, or capable of being made definite without further agreement of the parties, that the promises and performances to be rendered by each party are reasonably certain": see G. Scammell and Nephew Limited v H C and J G Ouston [1941] AC 251 at p. 268. But in applying this test the court has an established reluctance to strike down what were obviously intended to be legally enforceable commercial agreements. In Hillas & Co Ltd v Arcos Ltd (1932) 147 LT 503 at page 504 Lord Wright said:-

"The document of 21 May 1930, cannot be regarded as other than inartistic, and may appear repellent to the trained sense of an equity draftsman. But it is clear that the parties both intended to make a contract and thought they had done so. Businessmen often record the most important agreements in crude and summary fashion; modes of expression sufficient and clear to them in the course of their business may appear to those unfamiliar with the business far from complete or precise. It is accordingly the duty of the court to construe such documents fairly and broadly, without being too astute or subtle in finding defects, but, on the contrary, the court should seek to apply the old maxim of English law verba ita sunt intelligenda ut res magis valeat quam pereat. That maxim, however, does not mean that the court is to make a contract for the parties, or to go outside the words they have used, except in so far as there are appropriate implications of law, as, for instance, the implication of what is just and reasonable to be ascertained by the court as a matter of machinery where the contractual intention is clear but the contract is silent on some detail. Thus, in contracts for future performance over a period the parties may neither be able nor desire to specify many matters of detail, but leave them to be adjusted in the working out of the contract. Save for the legal implication I have mentioned, such contracts might well be incomplete or uncertain; with that implication in reserve they are neither incomplete nor uncertain . . . ."

55. More recently the same point has been made by Rix LJ in Scammell v Dicker [2005] EWCA Civ 405 at paragraph 30:-

"In my judgment, however, these conclusions were erroneous. In the first place, as a matter of fact, it was not impossible to implement the parties' agreement. The surveyors, Mr West and Mr Pollard, in fact did so implement it. Although I am content to assume that the surveyors' agreement on the pegging out of the boundary line did not bind the parties (although I can see a strong argument for saying that there was a collateral agreement between the parties that it should do so) and that therefore the parties were entitled to dispute the surveyors' solution, nevertheless it is simply a non sequitur to argue from a disagreement about the meaning and effect of a contract to its legal uncertainty. Parties are always disagreeing about the contracts which they make. They take those arguments, if necessary, to the courts, or to arbitration, for their resolution: and sometimes the resolution is very difficult indeed to arrive at. That is equally true of disputes as to the meaning of contracts and of disputes as to the application of contracts to the facts and of disputes as to the proper understanding of the facts. None of that makes a contract uncertain. For that to occur – and it very rarely occurs – it has to be legally or practically impossible to give to the parties' agreement any sensible content."

56. Many of the cases in which this issue has arisen involve contracts for the supply of goods where either quantity, quality or price is expressed in non-specific terms. In Hillas & Co Ltd v Arcos Ltd itself the timber sold was to be "of fair specification". The court's attitude has been to attempt to give effect to the contract by resorting to market rates and prices or to trade standards which provide objective criteria for the determination of what the parties agreed. Where such criteria exist and are ascertainable then a term of reasonableness is often implied in order to incorporate them as the measure of performance required under the contract. This is no more than part of the process of construction which the court is required to undertake.
57. But the first question which arises on this appeal is whether any of that is necessary in the present case. The NVA is not a contract for the supply of goods or services or one which raises any difficulties about price. It preserves the detailed schedule of fees and charges set out in the Base Agreement which are readily calculable based on the number of departing passengers. The focus of the alleged uncertainty is the contractual obligation imposed on BMIB to operate two aircraft from the Airport by 30^th April 2006 at the latest. As already indicated, the judge held that this meant fly. Fly, in the context of the agreement, must mean fly commercially in the sense of providing a passenger service. The contract also identifies the type of aircraft to be used and the pattern of rotations ("exclusively"). What the NVA does not do is to specify the minimum number of flights required and it is this which is said to be fatal to its enforceability.
58. The absence of a minimum performance clause indicates that any decisions on the detailed flying programme were to be a matter for BMIB. This was Mr Brealey's submission to the judge who seems to have accepted it. As I have indicated, Mr Shah's submissions that these terms were too uncertain to be enforceable without the addition of some specification of minimum numbers is scarcely touched on by the judge and the most relevant passage is his apparent acceptance of Mr Shah's argument to this effect at paragraph 113 of his judgment (quoted in paragraph 48 above).
59. In my view the judge was wrong to regard the addition of a term as to the minimum number of flights as being necessary for the enforceability of the NVA. He resolved the question of whether the aircraft had to be flown commercially by his construction of the word "operating". He also identified the requirement that they should on any day of operation leave from and return to the Airport. Were it necessary for the court to determine whether the number of flights undertaken during any given period amounted to operating the aircraft, the contract contains sufficient terms to allow that to be done. The question for the judge faced with an allegation of breach of contract would be whether BMIB was, in a real and genuine sense, flying its aircraft. Token flights or a complete absence of any flights (which is this case) clearly would not amount to operating the aircraft. Subject to this, the question of how many times and to where are matters for the airline's discretion.
60. Mr Crane raised the problems which might occur in relation to aircraft which were grounded for technical reasons. I can see no difficulty about that. DTVAL accepts that the NVA confers a discretion on BMIB as to the flight destinations it operates to and the number of flights it makes to such destinations in every week. If an aircraft lies unused on any particular day due to technical problems a decision to cancel its flying programme for that day is within the discretion conferred by the contract. BMIB is not required to do the impossible.
61. This makes it unnecessary in my judgment for DTVAL to rely upon an implied term that BMIB would operate the aircraft in a way that was reasonable in all the circumstances. The NVA includes sufficient terms to enable the court to determine whether BMIB's obligations have been broken. I can therefore turn to consider whether any additional complications arise in relation to the assessment of damages.

Damages

62. My decision on the construction of the NVA disposes of any difficulties for the assessment based on the alleged lack of certainty in its terms. But the issue that has been raised as to whether the court is required to assume that the party in breach would not have operated above the minimum level of performance required under the contract, although not decisive of this appeal, remains relevant to the computation of damages on the assessment. The judge was, of course, only concerned to determine the question of liability but, because we have heard argument on this point, it is convenient to set out shortly my own views on this issue.
63. The established measure of damages in the case of a breach of contract is the sum necessary to put the injured party in the same position as he would have been in had he not sustained the wrong: see Livingston v Rawyards Coal Company (1880) 5 App.Cas.25 at page 39. The inquiry is therefore directed to what the party in breach is to have been taken to have done had he in fact performed the contract. In all cases of repudiation (and possibly even a breach) this will be a counter-factual assessment which will involve the court in stipulating the manner of performance which is to be assumed.
64. The principle that the defaulting party should be taken for the purposes of this exercise not to have done more than the contract requires of him is one way of preventing the process of assessment from becoming arbitrary as well as fictional. It also, as Mr Crane submits, prevents the claimant from recovering damages in respect of something which the defendant was not obliged to do. Most of the cases are concerned with defining the limits of this principle. It is one which is capable of operating or being satisfied at a number of different levels. Most obviously it would exclude an assessment of the value of the contract to the innocent party which was conducted by reference to benefits which were themselves extra-contractual. An example of this would be the payment of discretionary bonuses to an employee or contractor where the contract makes no provision for them. This was considered by the Court of Appeal in Lavarack v Woods of Colchester Ltd [1967] 1 QB 278 which I shall come to shortly. But the principle could also be engaged to require the court to assume that the defaulting party would, as between two or more alternative methods of contractual performance, have opted for the one least onerous to himself. Similarly it might be said that, where the contract does not stipulate alternative methods of performance but requires the defaulting party (for example) to purchase goods above a stated minimum quantity, damages should be assessed by reference to the minimum level of lawful performance thereby disregarding the possibility (even, in some cases, the probability) that had the contract in fact been performed, a larger number of the goods would have been purchased. The question in cases of this kind is whether the argument or principle that a claimant should not receive damages for something which the defendant was not, strictly speaking, required to do should displace or limit the court's factual inquiry as to what, in the circumstances (bar the repudiation), the defendant would in fact have done.
65. The question of what is the correct test to apply in relation to this legal hypothesis has been considered in a series of decisions of the Court of Appeal which we are required to apply. A convenient starting point is the decision in Abrahams v Herbert Reiach Ltd [1922] 1 KB 477 where the contract was one to publish articles in the form of a book on payment to the authors of a royalty. The contract did not specify the number of copies that were to be printed or the price of the book. Sankey J held that this was an enforceable contract which required the publication of at least one book. The issue for the Court of Appeal was how the publisher was to be assumed to have performed the contract had he not repudiated it. The case has therefore a number of obvious similarities to the present one.
66. The defendants submitted that they had a discretion as to the number of copies to be printed and that it should not therefore be assumed that they would have done more than to publish the smallest number allowed. All three members of the Court of Appeal held that the plaintiff was entitled to £100 in damages but their approach to the issue of principle was not identical. Bankes LJ said (at page 480) that:-

"This contract only imposes one obligation upon the appellants namely, to publish. The question is what will satisfy that obligation? The appellants have a wide discretion; the time of publication, the number of copies to be printed, the price at which they are to be offered, and the form the book is to take are all left to their judgment. That however does not dispose of the case, because they have repudiated their obligation altogether, and the difficult question we have to decide is in what position the respondents would have stood if the appellants had performed their obligation. To answer this question the Court must come to some conclusion on matters on which there is no evidence; how the appellants would have exercised their discretion; what number of copies they would have published; how many editions would be reasonable."

67. Scrutton LJ (at page 481) took a rather narrow view:-

"Mr. Jowitt contended that if one copy was published that would satisfy the appellant's obligation. That cannot be laid down as a proposition of law. I think the appellants were bound to make such a publication as could be considered reasonable in the circumstances. Having done that they are not bound to do anything further. Provided that they make a reasonable publication the number and price of the copies are left to them. They broke their contract, and now what is the measure of damages? There are two principles which may seem to clash. One of these is stated by Lord Selborne in Wilson v. Northampton and Banbury Junction Ry. Co. (1) "In the case of damages, as it appears to me, the plaintiff will be entitled to the benefit of such presumptions as, according to the rules of law, are made in Courts both of law and equity against persons who are wrongdoers in the sense of refusing to perform, and not performing, their agreements. We know it to be an established maxim, that in assessing damages every reasonable presumption may be made as to the benefit which the other parties might have obtained by the bona fide performance of the agreement." I am not inclined to be strict in limiting the damages recoverable against wrongdoers, but if their obligation is left so much to their discretion that there are several ways of performing it, I have always understood that the Court assesses damages on the basis that "if the contract could have been performed by the performance of the alternative least beneficial to the plaintiff, the measure of damages would be regulated by the loss occasioned by non-performance of that alternative": Deverill v. Burnell, per Bovill C.J. The simple reason for this is that a defendant is not liable in damages for not doing that which he is not bound to do. In assessing the damages in this case I try first to ascertain what edition of the book would have been a performance of this contract; I do not forget that the respondents cannot recover more than they would have suffered if they themselves acted reasonably; I bear in mind also that the appellants are wrongdoers; and acting on these lines I do not dissent from an assessment of 100l., though I think myself the plaintiffs might have got considerably less."

67. Atkin LJ said that he was unable to go all the way with Scrutton LJ. At page 483 he said that:-

"If a merchant makes a contract to deliver goods to a shipowner to be carried by him for reward, and the merchant fails to provide the goods, the Court must first find what is the contract which has been broken; and if it was to carry the goods to one of two alternative ports at different distances from the port of loading at rates of freight differing according to the distance, the only contract on which the shipowner can sue is a contract for carriage to the nearer port. The plaintiff cannot prove a contract for performance of the more onerous obligation. This explains why in cases of this kind the Court regards only the lesser of two alternative obligations. But in the present case there are no alternatives, and to adjust the rights of the parties the only method is to form a reasonable estimate of the amount the respondents would be in pocket if the appellant had kept his promise. Everything likely to affect the amount of the profit must be considered; the nature and popularity of the subject matter, the reputation of the authors, the cost of producing a book on that subject, the price at which it would command a sale, the business capacity of the publishers and the chances of earning a profit by the sale of the book. On the other hand the publishers are not bound to run risks contrary to their judgment; they would naturally and properly allow for fluctuation in the public taste for literature of this kind. An analogous calculation has to be made when a man having engaged to take another into his service for a time and to pay him a share in the profit of his business, refuses to employ him at all. In assessing the damages for the breach of this contract the question is not how the employer could carry on his business so as to make the least possible profit and so involve himself in the least possible obligation towards the plaintiff. Apart from his contract, he need not carry on business at all. The proper method of assessment is quite different; it is to make a reasonable computation of the amount the respondents would have received had the contract been fulfilled."

68. There was clearly a consensus between the members of the court that, in relation to alternative methods of performance, the claimant will be unable to rely upon the defendant performing the contract in the more onerous of the two or more ways permitted. But that was not the type of contract under consideration in Abrahams and it is not the type of contract which we have to deal with in this case. Where there is only a single obligation to be performed it is clear that the majority view was that an assessment of damages should not, as a matter of law, be limited strictly to what was the minimum level of performance permitted under the contract but should extend to a calculation of how the contract would have been performed at the relevant time had it not been repudiated. This will take into account the likely profitability of the contract and any other relevant facts that would have influenced the method of performance.
69. The next case to mention is Lavarack (supra). This was a claim for loss of earnings due to wrongful dismissal. The contract of employment entitled the plaintiff to a salary of not less than £4,000 per annum plus "such bonus (if any) as the directors … shall from time to time determine." The majority of the Court of Appeal held that the damages could not include an amount in respect of lost bonuses which the defendant employer was under no obligation to pay. Diplock LJ (at page 294) said this:-

"The general rule as stated by Scrutton L.J. in Abrahams v. Reiach (Herbert) Ltd., that in an action for breach of contract a defendant is not liable for not doing that which he is not bound to do, has been generally accepted as correct, and in my experience at the Bar and on the Bench has been repeatedly applied in subsequent cases. The law is concerned with legal obligations only and the law of contract only with legal obligations created by mutual agreement between contractors - not with the expectations, however reasonable, of one contractor that the other will do something that he has assumed no legal obligation to do. And so if the contract is broken or wrongly repudiated, the first task of the assessor of damages is to estimate as best he can what the plaintiff would have gained in money or money's worth if the defendant had fulfilled his legal obligations and had done no more.

Where there is an anticipatory breach by wrongful repudiation, this can at best be an estimate, whatever the date of the hearing. It involves assuming that what has not occurred and never will occur has occurred or will occur, i.e., that the defendant has since the breach performed his legal obligations under the contract, and if the estimate is made before the contract would otherwise have come to an end, that he will continue to perform his legal obligations thereunder until the due date of its termination. But the assumption to be made is that the defendant has performed or will perform his legal obligations under his contract with the plaintiff and nothing more. What these legal obligations are and what is their value to the plaintiff may depend upon the occurrence of events extraneous to the contract itself and, where this is so, the probability of their occurrence is relevant to the estimate.

…

The events extraneous to the contract, upon the occurrence of which the legal obligations of the defendant to the plaintiff thereunder are dependent, may include events which are within the control of the defendant: for instance, his continuing to carry on business even though he has not assumed by his contract a direct legal obligation to the plaintiff to do so. Where this is so, one must not assume that he will cut off his nose to spite his face and so control these events as to reduce his legal obligations to the plaintiff by incurring greater loss in other respects. That would not be the mode of performing the contract which is "the least burthensome to the defendant."

70. It is important, I think, to read the references to the defendant being assumed to have fulfilled his legal obligations and nothing more in the context of what the court had to decide in that case. The issue to which these observations are directed was whether the damages awarded for the wrongful repudiation of the contract should include compensation for the loss of bonuses which the employer was never under any form of contractual obligation to pay. Diplock LJ was not, I think, referring to the question decided in Abrahams which was how to assess damages in relation to benefits which were undoubtedly contractual. This is made clear in the judgment of Russell LJ (at page 298) where he says that:-

"A plaintiff in an action for damages for wrongful dismissal can rely only on the fact that the defendant was obliged to carry out the contract sued upon. His prospects in terms of money or money's worth resulting from the carrying out of the contract may be conditioned by the estimated impact of external events on the results of the carrying out. But it has never been held that the plaintiff can claim any sum on the ground that the defendant might after the repudiation date have voluntarily subjected himself to an additional contractual obligation in favour of the plaintiff."

71. Lavarack can therefore be contrasted with the decision of the Court of Appeal in Cantor Fitzgerald International v Horkulak [2004] EWCA Civ 1287 where the employee was contractually entitled to receive a bonus but the amount of the bonus was left to the employer's discretion. The Court of Appeal rejected a submission that damages should be assessed on the basis that the bonus was fixed at a nil or nominal amount. That exercise of the discretion would have involved, they said, a breach of an implied obligation to exercise it in good faith and not either capriciously or arbitrarily. An paragraph 48 of the judgment the Court said:-

"[48] The judge was correct to find that application of the principle in Lavarack v Woods provided no rule of thumb applicable to discretionary bonus cases for reasons which I have already made clear. In that case the claimant was never party to the putative agreement in respect of which he claimed damages and the court reserved its position in respect of the outcome if the claim had been made for loss of bonus under the scheme applicable to the claimant during his employment, had it continued (see paras 31-35 above). Nothing was said in Lavarack v Woods to suggest that, in respect of a claim for damages put upon the basis that the claimant would have received payments under a discretionary bonus scheme of which he was already a potential beneficiary, the court should assume that the employer's discretion would be exercised against him in a case where such a decision would be irrational or arbitrary or one which no reasonable employer would make. The broad principle that a defendant in an action for breach of contract is not liable for doing that which he is not bound to do will not be applicable willy-nilly in a case where the employer is contractually obliged to exercise his discretion rationally and in good faith in awarding or withholding a benefit provided for under the contract of employment. Where the employer fails to do so, the employee is entitled to be compensated in respect of such failure: cf the observations of Timothy Walker J in Clark v BET plc quoted at para 36 above."

72. Of course, the implication of a term not to act arbitrarily (which is not contended for in this case) allows the court to set what it considers to be a reasonable level of performance even applying the approach of Scrutton LJ in Abrahams. In such a case that is the minimum level of performance permitted by the contract. The same applies to the decision of Mustill J in Paula Lee Ltd v Robert Zehil Ltd [1983] 1 AER 390 where a contract to purchase not less than 16,000 garments each season was construed by the judge as subject to an implied term that the garments would be selected in a reasonable manner (i.e. not the cheapest garments from a single range). Performance even strictly in accordance with this obligation would, by definition, have to be reasonable.
73. These cases do not therefore really resolve the difference of opinion which existed between the members of the court in Abrahams. They can on one view be regarded as consistent with, or an application of, Scrutton LJ's test. This was certainly the view of the Court of Appeal in the "World Navigator" [1991] 2 Lloyd's Rep. 23. After quoting from the judgment of Mustill J in Paula Lee, Parker LJ (at page 28) said that:-

"From this it appears to me that whatever he may have said earlier he was in truth applying the ordinary principle and seeking to find what was the minimum obligation so that damages might be assessed by reference to that obligation.

He was not therefore considering what the defendants would probably have done but what would be the minimum which they could have been obliged to do had the contract continued."

74. Similarly at page 32 Staughton LJ said that:-

"Neither case is inconsistent with the general principle. In each the defendants were held to have, not an unfettered choice or option as to how they would perform their obligation, but a duty to perform it in a reasonable manner. Once the contracts were thus construed, the damages had to be calculated on the basis of such performance. Indeed the Paula Lee case reinforces the general principle instead of contradicting it, for Mr. Justice Mustill held that many different selections of garments might be reasonable, and that within that range the damage must be based on the selection which would have been least profitable for the plaintiffs.

The Abrahams and Paula Lee doctrine cannot be applied here. It is well established that a shipper of goods may use all the laytime in loading if he pleases, even though he could easily have loaded more rapidly. He is under no obligation to act reasonably in that respect: Margaronis Navigation Agency Ltd v Henry W. Peabody Ltd [1964] 1 Lloyd's Rep. 173; [1965] 1 QB 300, Nolisement (Owners) v Bunge & Born [1917] 1 KB 160."

75. The "World Navigator" was a claim by the buyers of a consignment of maize for damages against the sellers for wrongfully detaining the vessel at the port prior to the commencement of loading. They claimed that the sellers were liable for the additional demurrage incurred as a result of the delay in loading some 12,000 tonnes of the cargo. The claim could only be made good if the sellers (whilst having delayed the start of loading) could not take advantage of the fact that the contract allowed for up to 24 days of laytime on the basis of a contractual load rate of 500 tonnes per working day. The rate of loading had in fact exceeded that with the result that only 18 out of the maximum 24 days permissible had been used. If the sellers could rely on the full period of laytime available (although not in fact used) this would exclude any loss attributable to the delay in the start of loading for which they were contractually responsible.
76. The Court of Appeal accepted the sellers' argument that they were entitled to take into account the full period of laytime permitted under the contract in determining whether the loading had been completed in time. They rejected the approach of the judge which had been to look at the sellers' breach of contract as a breach of a discrete obligation and to assess damages by reference to the time actually taken to load compared with the time that would have been taken had the vessel berthed on time and loading had taken place at the same rate. The reasoning of Parker LJ on this point is set out in his judgment at page 29 as follows:-

"To accept the buyers' contention would in my judgment be making the sellers pay damages for failing to do that which they were not obliged to do. For the assumed breach damages would no doubt have been recoverable if loading had taken longer than the end of the period of laytime allowed by the contract if the vessel had berthed on June 26, but it did not.

The truth of the matter is that by their assumed breach the sellers had in effect deprived themselves of the ability, if they were to escape a liability to pay damages to load only at the minimum rate prescribed, and put on themselves the necessity of loading at a higher rate in order to prevent such a liability occurring.

Before leaving this aspect of the case I should mention that although I agree that the assumed obligation and breach was single the sellers were left, in effect, with the option of alternative modes of performing their obligation. They could load when berthed at the minimum rate prescribed by the formula, in which case damages would occur, or they could load at whatever minimum rate was necessary in order just to use up the laytime available under the formula but no more, or they could, as in fact they did, load at such a rate that the vessel was free before the time when they would have been obliged to make her available had there been no breach. In the second and third case no damage is suffered."

77. The minimum load rate terms were relied on by the sellers in that case to establish that no loss in fact resulted from the wrongful detention of the vessel. Insofar as the contract allowed them an alternative loading rate to the one they actually used, they were entitled to take advantage of it in response to the damages claimed. Looked at in this way the decision does not venture beyond what might be described as the alternative performance rule which was accepted and explained in Abrahams in the passages I have quoted. It does not in terms deal with the issue in this case which is how one assesses damages where no such alternatives exist.
78. None of the cases I have referred to has or could have questioned the principle laid down by the majority of the Court of Appeal in Abrahams which is set out most clearly in the judgment of Atkin LJ. The court, in my view, has to conduct a factual inquiry as to how the contract would have been performed had it not been repudiated. Its performance is the only counter-factual assumption in the exercise. On the basis of that premise, the court has to look at the relevant economic and other surrounding circumstances to decide on the level of performance which the defendant would have adopted. The judge conducting the assessment must assume that the defendant would not have acted outside the terms of the contract and would have performed it in his own interests having regard to the relevant factors prevailing at the time. But the court is not required to make assumptions that the defaulting party would have acted uncommercially merely in order to spite the claimant. To that extent, the parties are to be assumed to have acted in good faith although with their own commercial interests very much in mind.

Conclusion

79. For these reasons I would allow this appeal and direct an assessment of damages in the claim.

Lord Justice Toulson :

80. I agree.
81. The main issues debated in this appeal can be divided into four:

1. Did the parties intend the base agreement and the novation agreement to impose an obligation on the respondent (the airline) to carry on the operation referred to in the opening provisions of the base agreement, and in clause 2 of the novation agreement, for the duration of the agreement?

2. If so, was that obligation void for uncertainty?

3. If the respondent (the airport) is right on the first and second issues, how should the court approach the question of damages for breach of the obligation?

4. Ought the judge to have found that it was an implied term of each agreement that the airline would operate the two aircraft based at the airport in a manner which was reasonable in all the circumstances?

82. In the well presented submissions on both sides the issues were interwoven and the arguments overlapped.

Issue 1: Did the parties intend the base agreement and the novation agreement to impose an obligation on the airline to carry on the operation referred to in each agreement for the duration of the agreement?

83. This is a question of the proper construction of each agreement in its commercial context. I agree with Davis J and with Patten LJ that the question should be answered in the affirmative for the reasons given by them.
84. Mr Crane QC argued that each agreement was facultative in relation to the airline but obligatory in relation to the airport. In other words, the airline was free to decide whether to conduct the operation and, if so, for how long (up to the end of the period of the contract); if and for as long as it chose to do so, the airport was obliged to provide the facility and benefits identified in the agreement. I agree with the judge that the description of the operation in the base agreement is key to the whole agreement and that the language is more naturally read as language of bilateral obligation than permission coupled with unilateral obligation. The position is still more clear under the novation agreement, clause 2 of which stated that:

"…For the avoidance of doubt…from Spring 2006 bmibaby will support a x2 based operation operating exclusively from DTVA."

85. Each agreement should axiomatically be read as a whole, but like the judge I am not persuaded that the other provisions relied on by Mr Crane, such as the claw back provisions, lead properly to the conclusion that the airline undertook no obligation to establish or maintain a x2 aircraft based operation. It is not surprising, particularly where public money was involved, that the parties should want the terms of any claw back payments to be specific. I am also unpersuaded by the argument that to construe either agreement as requiring the airline to establish and maintain a x2 aircraft based operation would be so commercially unrealistic or unreasonable that the parties cannot have intended it.
86. I agree moreover with Patten LJ that although the parties contemplated reaching further agreement on operational matters, it is not right to construe either agreement as intended to contain no obligation to establish or maintain a x2 aircraft based operation unless and until a further agreement was reached (the "agreement to agree" argument).

Issue 2: Was the obligation void for uncertainty?

87. Where parties intend to create a contractual obligation, the court will try to give it legal effect. The court will only hold that the contract, or some part of it, is void for uncertainty, if it is legally or practically impossible to give to the agreement (or that part of it) any sensible content. (See Scammell v Dicker [2005] EWCA Civ 405, [2005] 3 All ER 838, para 30, per Rix LJ.)
88. In a forceful argument, Mr Crane submitted that "a x2 aircraft based operation" lacks the necessary content to be capable of being the subject of an obligation by the airline. The contract did not specify a minimum number of hours or rotations for which each aircraft was to be flown over any particular period in order to constitute the aircraft being operated, nor did it contain any mechanism by which the court could determine as at the date of the contract what the minimum requirement would be. I accept that the uncertainties of what might happen during a 10 year period were such that nobody could have determined at the outset what would be the minimum number of flights or rotations during the lifetime of the contract in order for an aircraft to continue to be operated under it. Wars, strikes, pandemics, terrorism, bad weather conditions and many other factors might affect what would be practicable.
89. Therefore, argued Mr Crane, the obligation was void for uncertainty, because the obligee (the airline) could not know in advance how many flights it had to operate per aircraft in order to comply with its contractual obligation. At this point I part company with him. To my mind the proper test is not whether a court would have been able to determine on, say, an application for specific performance what would be the minimum number of flights required in order to constitute a x2 aircraft based operation during all vicissitudes which might occur during the period of the contract, but whether the court would be able to determine on a set of facts which had occurred (including any unusual vicissitudes) whether the airline had, or had not, maintained a x2 based aircraft operation at the airport. In reaching that decision it would have to take into account the factual circumstances. Facts might arise where the question whether the airline had continued to maintain such an operation might be borderline, but that may be said of many contractual obligations. In this case, the airline had a wide measure of discretion in fixing any programme. As Mummery LJ observed during the argument, an agreement allowing wide scope to one party to determine its manner of performance, which may be subject to future imponderables, may not be capable of enforcement by an order for specific performance without it necessarily following that the agreement is too uncertain to have legal effect at all.
90. It is not uncommon in commercial life for parties to enter into a contract by which one party is to conduct an operation over which it has a large degree of discretion; but it will usually be possible for a court to determine without too much difficulty whether that party has ceased to perform the operation at all. The present contract may be unusual in its length and the degree of discretion given to the airline, but I do not see that the court would have an insuperable difficulty in deciding whether the airline had altogether ceased to conduct the operation.
91. There is a further problem with the airline's argument on the issue of voidness for uncertainty. In the course of his argument Mr Crane conceded at first that if the airline's case on voidness for uncertainty was right, this would not only apply to the airline's obligation to maintain a x2 aircraft based operation. It would logically also lead to the conclusion that the airport's obligation to provide the benefits promised to the airline on the basis of the airline maintaining a x2 aircraft based operation was void by reason of the same uncertainty about what the maintenance of a x2 aircraft based operation meant. This concession ran counter to Mr Crane's argument about the facultative/obligatory nature of the agreements and after further reflection he withdrew it, but I consider that it was rightly made.

Issue 3: How should the court approach the question of damages for breach of the obligation?

92. It could be argued that since it had been agreed that the trial should be confined to liability (and a discrete issue about whether certain indirect losses were sufficiently foreseeable to be claimable as damages), the question how damages should be measured did not need to be considered. The finding that the agreements had imposed on the airline an obligation which it had repudiated should have entitled the airport to judgment for at least nominal damages, and the question whether it could establish a claim for substantial damages was a matter for later enquiry. However, it would not be helpful for the court to adopt that position at this stage and in any event the implied term arguments were interwoven with arguments about the proper approach to damages. On the appeal Mr Brealey QC advanced his case on an implied term as a belt and braces argument. His primary submission was that, without any need for an implied term, the airport was entitled to an assessment of damages based on the net earnings which the airport would have been likely to have received if the airline had established and maintained a x2 aircraft based operation over the period of the contract. The fact that difficult questions may arise and that the airport's loss of profits may depend upon many speculative factors is not a sufficient reason for denying the airport an assessment. (See Simpson v London and North Western Railway Company (1876) 1 QB 274, 277, per Cockburn CJ, and Chaplin v Hicks [1911] 2 KB 786, 792, per Vaughan Williams LJ.)
93. Mr Crane submitted that any assessment would have to be based on the minimum number of flights which the airline was contractually bound to operate, and, since that was impossible to state as a matter of construction of the contract, any damages must be nil. (He also submitted that the implied term put forward by the airport provided no solution, but that is another point.)
94. The cases show that the question how damages should be assessed for repudiatory breach of a contract which allows a measure of choice to the party in breach (the defendant) may be difficult. Referring to the difficulties which can arise in this branch of the law, Mustill J observed in Paula Lee Ltd v Robert Zehil & Co Ltd [1983] 2All ER 390, 393 that the principles rarely have to be considered but are less clear than might be thought. It may be added that it is not easy to reconcile all the dicta in the various cases.
95. The cardinal principle of any assessment of damages for breach of contract is that the innocent party (the claimant) is entitled to be put in the same position as he would have been in if the defendant had not broken the contract. This requires a careful analysis of the contract. Subsidiary general rules have been developed for measuring damages in different types of case, although there may be a need for caution to see that they are not applied mechanistically in particular situations where to do so would defeat the cardinal principle. As the case law shows, there is a wide range of possible permutations which may affect the right way to assess damages. They include the following, although not every case falls neatly into one of them:

1. The contract requires the defendant to do X or Y.

2. The contract requires the defendant, if he has not done X, to do Y.

3. The contract requires D to do X and the claimant has a reasonable expectation that he will do Y.

4. The contract requires the defendant to do X and allows him a discretion how he performs the obligation.

96. In the first type of case (where the contract requires the defendant to do X or Y) the rule is straightforward if – and this is a big if – the construction of the contract is itself straightforward. The rule was stated by Lord Cranworth LJ in Robinson v Robinson (1851) 1 De G M & G 247, 256-258:

"…in ordinary cases, every wrongdoer is bound to put the party injured, so far as the nature of the case allows, in the same situation in which he would have stood if the wrong had not been done.

…

Where a man is bound by covenants to do one of two things, and does neither, there in an action by the covenantee, the measure of damage is in general the loss arising by reason of the covenantor having failed to do that which is least, not that which is most, beneficial to the convenantee: and the same principle may be applied by analogy to the case of a trustee failing to invest in either of the two modes equally lawful by the terms of the trust.

…

The trustee is answerable for not having done what he was bound to do, and the measure of his responsibility should be what the cestui que trust must have been entitled to, in whatever mode that duty was performed."

97. In that case the court had no difficulty in analysing what the trustee had to do. In other cases the task may be more problematic. This is illustrated by the cases of Thomas v Clarke and Todd (1818) 2 Stark 450, Capper v Forster (1837) 3 Bing N C 938 and Cockburn v Alexander (1848) 6 CB 791, which Mustill J discussed in Paula Lee.
98. The three cases had in common that they involved the charter of a vessel for an outward and return voyage, and that on the return leg the vessel was to be loaded with a cargo consisting of various possible forms of merchandise for which different freight rates were provided.
99. In Thomas v Clarke Abbott CJ (later Lord Tenterden) laid down a rule summarised in the head note as follows:

"In an action for not supplying a cargo under a charter party, according to the terms of which different articles of freight are to be paid for at different rates by weight, and the freighter is at liberty to supply which articles he pleases, an average value of freight, calculated upon the various rates of freight in the proportion of different articles usually carried on such a voyage, is the proper measure of damages."

100. The charterers argued unsuccessfully that they were not liable for more than the amount of the freight "supposing the vessel to have been laden with one of the articles specified, which would have yielded the lowest amount of freight". The owners gave evidence of the amount which the vessel had actually earned on former similar voyages. The report states that the judge "intimated that the proper course would be to estimate the freight by means of an average, so as to take neither the greatest possible freight nor the least", but it does not state his reasoning. In Paula Lee Mustill J said that he believed the reasoning to be that, whatever goods the charterer might choose to ship and in whatever proportions, he was obliged on the true construction of the charter party to pay freight calculated on the basis of an average of the enumerated articles. That is a possible explanation. Another possible explanation is that since there was no reason to suppose that it would have been any more onerous for the charterer to load the vessel with one article rather than another, and since there was no way of telling what cargo the charterer was most likely to have loaded, the fairest way of estimating the owners' loss was by taking an average.
101. The head note to Capper v Forster reads as follows:

"Where a ship is chartered to bring home a cargo of enumerated articles at rates of freight specified for each, which articles are not provided by the charterer, the freight must be paid upon average quantities of all of the articles, whether the ship return empty or laden with a cargo of articles different from those enumerated."

102. The charter party provided that the charterer, having discharged the outward cargo, should "reload a full and complete cargo of lawful merchandise". It enumerated various types of merchandise with different rates of freight. The charterer re-loaded with a small quantity of specified goods and a much larger quantity of goods for which there was no specified freight. The owners contended that on the proper construction of the charter party they were entitled to an amount of freight which would have been earned if the ship had brought home a cargo consisting of average quantities of all the enumerated articles. The charterer contended that the owners were only entitled to freight for the cargo brought home upon a quantum merit or to the freight due on a full cargo of any one of the enumerated articles at the charterer's option.
103. Tindall CJ said that if the ship had returned empty the measure of damages was settled by the decision in Thomas v Clarke, which he said "appears to us to lay down and establish a rule, which is at once just and reasonable, and may fairly be inferred to meet the intentions of the contracting parties". He further considered that there was no distinction to be made between the application of the rule to a case where the ship returned empty and one where she returned with a cargo of articles largely of a different nature and quality than those enumerated in the charter.
104. In Cockburn v Alexander the charter party provided that the charterer was to load "a full and complete cargo of wool, tallow, bark or other legal merchandise" and stipulated freight rates for the identified articles but not for "other legal merchandise". It also prescribed maximum quantities of bark, tallow and hides. The charterers were not able to load the vessel entirely with specified goods in the permitted quantities, because there was a shortage of wool. Instead they loaded the vessel with excessive quantities of bark, tallow and hides, with some wool, and with an assortment of other legal merchandise. The owners contended that they were entitled to be paid freight on the assumption that the cargo consisted of the specified maximum quantities of tallow, bark and hides, the residue of the ship being filled up with wool. As a matter of arithmetic, this involved applying the freight rates specified in the charter party in the way least favourable to themselves (and less favourable than if they had applied the average approach adopted in Thomas v Clarke and Capper v Forster). The charterers contended that, under the words "or other legal merchandise", they were entitled to ship goods other than those enumerated, to an extent consistent with the proper trim of the vessel; and that the freight payable for such goods, none being expressly stipulated for by the charter party, was to be at the current rate at the port of loading. This would have produced a lower figure than the sum claimed by owners.
105. The court (Wilde CJ, Maule and Vaughan Williams JJ) found the construction of the contract to be difficult, but concluded that the freight was to be calculated by reference to the rates specified for the articles enumerated. Wilde CJ considered that the matter was governed by the rule in Thomas v Clarke and Capper v Forster, which he said ought not to be disturbed, and he noted that the owners were not seeking to go as far as in Thomas v Clarke. Perplexingly, he concluded his judgment by saying that the principle contended for by the owners was correct and that damages were to be calculated as if the charterer had performed the contract in the way most favourable to himself, and least favourable to the ship owners – a comment difficult to reconcile with his affirmation of the rule in Thomas v Clarke and Capper v Forster on which he said that he rested his opinion. Vaughan Williams J agreed with Wilde CJ that "it would be inconvenient now to unsettle the law as established by Thomas v Clarke and Capper v Forster" and that the matter was governed by those cases.
106. Maule J stated that:

"Generally speaking, where there are several ways in which the contract might be performed, that mode is adopted which is the least profitable to the plaintiff, and the least burthensome to the defendant."

107. As Mustill J observed in Paula Lee, unless this was only an elaborate way of saying that the performance was to be assumed which would yield the least award, the two halves of the rule stated by Maule J are capable of leading to different answers.
108. Maule J went on to offer his own explanation of Thomas v Clarke. He observed that in cases of the kind under consideration, different amounts might be the proper measure of damages for breach of contract in different circumstances. If only one of the articles permitted to be loaded was available, then the charterer would not in truth have any option; he would be contractually bound to load the requisite quantity of the article which was available. That might have involved the loading of goods which would have been most profitable to the owner or the opposite. In either event, the damages would be governed by what was available. If no goods at all were available, he considered that the right way to ascertain the damages might be to take an average of all possible kinds of goods, and that this was the way that "Lord Tenterden arrived at the opinion expressed in Thomas v Clarke – viz that where there is no cargo at all to be had, the average is to be taken of all possible kinds of cargo", the rationale being that the obtaining of goods of any one kind could not a priori be considered as more probable than the obtaining of any of the others.
109. These three cases are capable of different interpretations, but they pre-dated Lord Cranworth LJ's statement of the general rule in Robinson v Robinson, which has since been followed. What they illustrate is the need for care in the construction of the contract and the application of the rule to the factual circumstances, because events extraneous to the contract may affect the legal obligations of the defendant.
110. The same lesson emerges from Deverill v Burnell (1873) LR 8 CP 475. The plaintiff entrusted certain drafts to the defendant for presentment on terms that the defendant should return the drafts, if not paid after acceptance, to the plaintiff or should pay the plaintiff the amount of them. The defendant did neither. The bills were found by the jury to be worthless. It was argued by the defendant that the plaintiff was entitled only to nominal damages on the basis that the defendant was entitled to have performed the contract by returning the bills and that by his failure to do so the plaintiff had lost nothing. The majority of the court found that the defendant was liable to pay the amount of the bills on the basis that the defendant, having not returned the bills, was obliged on the proper construction of the contract to pay the amount for which they were drawn. Bovill CJ dissented. The difference between him and the majority was not as to the principles applicable but as to the construction and effect of the contract. He stated the issue in this way:

"If the contract…is simply in the alternative, to do one of two things, it would be satisfied by the performance of either, and the damages would be the loss occasioned by non-performance of that alternative which would be least beneficial to the plaintiff. If the true construction be that of the two things to be done one depended upon the non-performance of the other, that is, if the defendant did not return the bills, then he should pay the amount of them, the damages would be the non-payment of that amount. The rule of law is clear, that, in the case of alternative contracts, the person who has to perform the contract has the right to elect which branch of the alternative he will perform. On the other hand, it is equally clear, if the contract is to do a thing, and if not to pay a sum of money, then the damages for not doing the thing are the sum of money."

111. In short, the case illustrates the distinction between the first and second of the categories referred to in paragraph 96.
112. In the third type of case (where the contract requires the defendant to do X and the claimant has a reasonable expectation that the defendant will also do Y) it is again critical to determine the bounds of the defendant's contractual obligations. This is illustrated by the contrasting decisions of the Court of Appeal in Lavarack v Woods of Colchester Limited [1967] 1 QB 278 and Cantor Fitzgerald International v Horkulak [2004] EWCA Civ 1287. Both involved the wrongful dismissal of an employee.
113. In Lavarack v Woods the plaintiff entered into a five year employment contract with effect from 1 April 1962. The contract provided that the defendants should pay him "as remuneration for his services a salary at the rate of not less than £4000 per annum…and such bonus (if any) as the directors…shall from time to time determine…" He was wrongfully dismissed in July 1964. From 31 March 1965 the defendants, for reasons of policy, abolished their bonus scheme, and it was held that if the plaintiff had remained in their employment such abolition would not have constituted a breach of his contract. However, when the bonus scheme was abolished, the defendants entered into negotiations with those employees of similar grade to the plaintiff who were still in their employment, as a result of which their salaries were increased. The plaintiff claimed that his damages ought to include a sum representing the amount by which his salary would probably have been increased if he had not been wrongfully dismissed.
114. Lord Denning MR, dissenting, held that this head of claim was valid. He rejected the argument that it was a case in which the dictum of Maule J in Cockburn v Alexander ("where there are several ways in which the contract might be performed, that mode is adopted which is the least profitable to the plaintiff and the least burthensome to the defendant") was applicable. He held that there were not two ways in which the contract could be preformed. There was only one way, namely by the plaintiff performing his service and the defendants paying him for it. In such a case the compensation was to be based on the probabilities of the case, which were that if he had not been wrongfully dismissed he would probably have received an increased salary in lieu of future bonuses.
115. Diplock and Russell LJJ analysed the relationship between the parties differently. Diplock LJ began with the proposition that "a defendant is not liable in damages for not doing that which he is not bound to do" and he referred to Maule J's dictum in Cockburn v Alexander as one of the most firmly established applications of that general rule. He stated at 294:

"The law is concerned with legal obligations only and the law of contract only with legal obligations created by mutual agreement between contractors - not with the expectations, however reasonable, of one contractor that the other will do something that he has assumed no legal obligation to do."

116. He added the qualification that events extraneous to the contract, upon the occurrence of which the legal obligations of the defendant to the plaintiff are dependant, may include events which are within the control of the defendant: for example, continuing to carry on business even though he has not assumed by his contract a direct legal obligation to the plaintiff to do so. Where this is so, he said that the court will not assume that he will "cut off his nose to spite his face and so control these events as to reduce his legal obligations to the plaintiff by incurring greater loss in other respects". That would not be the mode of performing the contract which is "the least burthensome to the defendant".
117. Applying that general principle to the facts of the case, Diplock LJ observed that if the defendants had continued their bonus scheme, it might well be that on the true construction of the contract the plaintiff would have been entitled to compensation for loss of bonus; but that since they were under no contractual obligation to continue the scheme and had in fact discontinued it, he had no such entitlement. It was true that the employers might have entered into a fresh salary agreement with him, but Diplock LJ said of that possibility at 297:

"I know of no principle on which he could claim as damages for breach of one service agreement compensation for remuneration which might have become due under some imaginary future agreement which the defendants did not make with him but might have done if they wished."

118. In short, the plaintiff's claim for future increased salary under a new agreement was a claim for an extra-contractual benefit. Russell LJ agreed with the approach of Diplock LJ.
119. Cantor v Horkulak is an example of the hypothetical case which Diplock LJ distinguished in Lavarack v Woods. The claimant was employed as a senior manager working on interest rate swaps for a bond and money-broking house which traded in bonds, equities and interest rate derivatives. His contract provided for a basic salary and an annual discretionary bonus, the amount of which was to be mutually agreed between himself, the chief executive of the company and the president of its parent, and ultimately determined by the president. Giving the judgment of the court, Potter LJ observed (at 35) that Lavarack v Woods was not a case about the true construction of a discretionary bonus clause. He stated at 48:

"The broad principle that a defendant in an action for breach of contract is not liable for doing that which he is not bound to do will not be applicable willy-nilly in a case where the employer is contractually obliged to exercise his discretion rationally and in good faith in awarding or withholding a benefit provided for under the contract of employment. Where the employer fails to do so, the employee is entitled to be compensated in respect of such failure."

120. The leading authority in the fourth type of case (where the contract requires the defendant to do X and allows the defendant a discretion how he performs the obligation) is Abrahams v Reiach (Herbert) Limited [1922] 1 KB 477, a case further distinguished by the composition of the court (Bankes, Scrutton and Atkin LJJ). A firm of publishers contracted with the authors of a series of articles to publish them in a book and to pay a royalty of 4d for every copy sold. The form and price of the book, the number of copies to be printed and the date of publication were left to the publishers' discretion. They failed to publish the book at all. In an action for damages the defendants paid into court £25, representing the royalties on 1,500 copies. The trial judge awarded the plaintiffs £500, which represented 30,000 copies.
121. Counsel for the publishers argued that they would have performed their obligation by publishing the smallest number of copies that could be called an edition and that no one could dispute that 1,500 copies would constitute a publication within the meaning of the contract. If the defendant could perform his obligations in either of two ways, the plaintiff could only claim for a breach of that stipulation which was less profitable to himself.
122. Counsel for the authors argued that the contract did not impose alternative obligations on the appellants but a single obligation, namely to publish the book. It might be difficult to define the obligation, but it was still an obligation, although its bounds might be vague or obscure. Granted that an issue of 1,500 copies would be a publication within the meaning of the contract, so also would be an issue of 30,000 copies; but that did not convert the contract into an obligation to issue 1,500 or 30,000 copies at the option of the publishers. The court had to determine how much the authors lost by the publishers' failure to make such a publication as would have been reasonable in all the circumstances.
123. The Court of Appeal reduced the award from £500 to £100 but their reasons differed.
124. Bankes LJ rejected the appellants' argument that the case was analogous to one where the contract provided alternative obligations. The contract imposed a single obligation on the appellants, namely to publish. It was true that the appellants had a wide discretion, but since they failed to publish at all, the court had to come to some factual conclusions on which there was no evidence; how the appellants would have exercised their discretion; what number of copies they would have published; how many editions would be reasonable. He accepted that there were a number of points which would induce a reasonable expectation of a large sale, including the low cost of production, the reputation of the authors and the opportune occasion for publishing the book. But there were other factors to be taken into account and he concluded that the judge's overall estimate was too favourable to the authors.
125. Scrutton LJ took a different approach. He accepted the applicability of the principle that if there was more than one way of performing a contract the court would measure damages by the performance least beneficial to the plaintiff. He also held that the publishers were bound to make such a publication as could be considered reasonable in the circumstances. In assessing the damages he said that he tried first to ascertain what edition of the book would have been a performance of the contract. Without revealing his answer to that question, he said that he did not dissent from an assessment of £100, although he personally thought that the plaintiffs might have got considerably less.
126. Atkin LJ agreed in the result but not with Scrutton LJ's reasoning. Like Bankes LJ, he rejected the applicability of the rule that where a defendant breaks a contract containing alternative obligations, damages are to be assessed by reference to performance of the less onerous obligation. He said that in the present case there were no alternatives, and that to adjust the rights of the parties the only method was to form a reasonable estimate of the amount the authors would have been in pocket if the publishers had kept their promise. As to that he said at 483:

"Everything likely to affect the amount of the profit must be considered; the nature and popularity of the subject matter, the reputation of the authors, the cost of producing a book on that subject, the price at which it would command a sale, the business capacity of the publishers and the chances of earning a profit by the sale of the book. On the other hand the publishers are not bound to run risks contrary to their judgment; they would naturally and properly allow for fluctuation in the public taste for literature of this kind."

127. He further stated that in assessing the damages the question was not how the defendants could carry on their business so as to make the least possible profit and so involve themselves in the least possible obligation towards the plaintiffs. The proper method of assessment was quite different; it was to make a reasonable computation of the amount the authors would have received had the contract been fulfilled. This was the method pursued by the trial judge, but in his view the actual result was too favourable to the authors. To produce £500 damages there must have been a sale of 30,000 copies of the book, and that was too large an estimate, although the evidence warranted the inference that there would have been a substantial sale.
128. It is noticeable that Atkin LJ did not say that the contract contained an implied obligation to publish a reasonable amount. On his approach, this was unnecessary. The assumption in assessing damages was that the publishers would have taken into account all that would have maximised the number of sales, such as the nature and popularity of the subject matter, but that they would also not have been expected to run risks contrary to their commercial judgment.
129. Reading the judgment of Bankes LJ as a whole, I do not see any difference in substance between him and Atkin LJ.
130. Although in Lavarack v Woods Lord Denning was in a minority in his analysis of the contractual position between the parties, he was in my respectful opinion right in his citation of Abrahams v Reiach as authority for the proposition that where a contract imposes a single obligation, rather than alternative obligations, compensation is to be based on the probabilities of the case – on the remuneration which the claimant might reasonably be expected to receive – and not on the bare minimum necessary to have amounted to performance of the contract.
131. There is good practical reason for this. Where a contract imposes alternative obligations the contract itself will identify them. But where there is a single obligation expressed in broad terms, it may be conceptually very difficult to identify as a theoretical exercise what would have been a minimum performance level, as Abrahams v Reiach demonstrated. In that case the damages of £100 which the Court of Appeal considered appropriate would have been equivalent to the royalties on 6,000 copies. Would the printing of 6,000 copies have been a minimum contractual performance? If so, why? Why not 5,500 or 5,000? The questions are impossible to answer and it is notable, as I have said, that although Scrutton LJ stated that he considered what was the minimum number which would constitute a contractual performance, he did not state his conclusion or reasoning. It would be more possible, as the majority did, to make a broad brush assessment of the number of copies which the publishers would have been likely to print having regard to the potential saleability of the book. For that reason, the approach of Atkin and Bankes LJ affords a more practical and realistic way of assessing the true loss suffered by the breach. Indeed, the logic of the publishers' argument, as their counsel submitted, was that the authors should have recovered only nominal damages. This would not have done justice.
132. In Paula Lee the plaintiff dress manufacturers appointed the defendants as their sole distributors for the sale of a range of garments in various Middle Eastern countries. Under the agreement the defendants undertook to purchase not less than 16,000 garments each season and it was agreed that defendants were to have complete discretion in the marketing and selling policy to be adopted in the territory. The defendants wrongly terminated the contract when it still had two seasons to run. The defendants contended that damages were to be assessed by choosing 32,000 of the cheapest garments in the range. The plaintiffs contended that damages were to be assessed by the average price of all the garments in the range, since the likelihood was that the defendants would have chosen garments of different styles and quality throughout the range.
133. Mustill J said that there was a single obligation, to purchase 16,000 dresses each season, and that the problem was to determine the content of the obligation in the absence of any express provision as to the ratios in which the dresses were to be ordered. As to the proper answer, Mustill J rejected both parties' contentions.
134. He noted, at 394, that one possible view was that the court should try to forecast how the defendant would have performed the contract but for the repudiation, but he rejected that approach as inconsistent with principle, since that would potentially involve penalising the defendant for failing to do something which he was not obliged to do. He also rejected the argument that the cheapest dresses must form the yardstick for the entire computation. Rather, he concluded that the defendant was under an implied obligation to have made a reasonable selection, and that the task of the court was to look at the range of reasonable methods and select that one which was least unfavourable to the defendant, bearing in mind that in deciding what methods qualified as reasonable the question must be approached with the interest of both parties in mind. He accepted that this might be a very difficult exercise to perform, but he said that he believed that this approach accounted for the decision in Abrahams v Reiach.
135. I respectively differ from Mustill J's analysis of the reasoning of the majority in of Abrahams v Reiach. As I read the case, Atkin LJ was explicit that the question was not how the defendants might have carried on their business in a way that would involve the least obligation towards the plaintiff. The proper method of assessment was quite different, namely to make a reasonable computation of the amount which the plaintiffs would have received, taking into account everything that was likely to have affected the size of the publication. As with Abrahams v Reiach so with Paul Lee, the difficulty with the approach favoured by Mustill J would be in determining how a judgment was to be made as to the minimum variety and price of garments which the defendants might reasonably have chosen. He said that he was very conscious of the problem which would face the court in establishing the boundaries of the range, but the task was left to a master with no further guidance. By contrast, it would have been conceptionally and practically simpler to carry out the kind of assessment contemplated by Atkin LJ, for which the court would have had the evidence of past performance.
136. The potential application of Abrahams v Reiach was considered by the Court of Appeal, on a rather unusual set of facts, in The World Navigator [1991] 2 Lloyds Rep 23. The dispute arose from a f.o.b. contract for the sale of maize. The buyers were required to tender the vessel in readiness to load at the port of loading between specified dates. There was an implied term requiring the sellers to co-operate in providing appropriate loading documentation and an express term stipulating the minimum rate at which the sellers were to load the goods.
137. The vessel gave notice of readiness to load on 13 June 1985. If the necessary documentation for loading had been in order, the vessel would have berthed on 25 June 1985. However, in breach of the implied term, the sellers failed to provide the necessary loading documentation on time, with the result that the vessel was unable to berth until 18 July 1985. Had she berthed on 25 June, the sellers would have had 48 days in which to load the entire cargo, i.e. until 12 August. Although loading did not begin until 18 July, the sellers completed the loading by 22 July, i.e. well within the permitted time if the commencement of the loading had not been delayed by the sellers' breach of the implied term.
138. The buyers claimed damages on the basis that the implied term was a discrete requirement, which allowed the sellers no option, and that in calculating damages the arbitrators should assess as a matter of probability what would have happened if the commencement of the loading had not been delayed.
139. This argument was upheld by the arbitrators, rejected by the arbitral board of appeal, accepted by Phillips J but rejected by the Court of Appeal. It is unsurprising that there was a divergence of opinion, because the issue involved a debatable point about the proper interpretation of the overall effect of the contract.
140. Phillips J said at [1991] 1 Lloyd's Rep 277, 287-288:

"The obligation broken, assuming breach to be established, was a single discrete obligation to procure the appropriate loading documents. No question of alternative obligations arises. No question of rendering the sellers liable for not doing that which they were not bound to do arises. The question is not: what was the least burdensome method of performing the obligation which the sellers failed to perform?...When assessing damages, the board should not have proceeded on the premise that, had the vessel berthed promptly, the sellers would have berthed so slowly that they would have used all the available lay time. The board's task was to ask how long loading would have taken, had World Navigator berthed promptly without losing her place in the queue. If the sellers were in a position to influence the loading rate, then the board should have considered how they would have done so. That question fell to be answered, not by assuming that the sellers would have loaded at the slowest permissible rate, but by considering all the factors that would have been likely to influence their conduct."

141. The Court of Appeal analysed the position differently. Parker LJ accepted that the implied term was a discrete obligation to which there was no option or alternative, but he said that it was nevertheless necessary to consider what were the consequences of the breach and what was required to put the buyers in the same position as if the breach had not occurred. Because of the speed at which the sellers loaded, the effect was that the buyers got that to which they would have been entitled had there been no breach. He said at 29:

"The truth of the matter is that by their assumed breach the sellers had in effect deprived themselves of the ability, if they were to escape their liability to pay damages, to load only at the minimum rate prescribed, and put on themselves the necessity of loading at a higher rate in order to prevent such a liability occurring."

142. On that approach, Abrahams v Reiach was of no assistance to the buyers. The case is a further example of the critical importance of the interpretation of the contract.
143. Drawing a line between a contract which creates alternative obligations, or one which prescribes (or provides a method of calculating) a quantifiable minimum level of performance, or one which creates a single obligation with an element of discretion how it is performed, involves an exercise of construction. Mustill J in Paula Lee considered it wrong that in the last category damages should be measured, in a case of repudiatory breach, by estimating how the contract would probably have been performed, rather than by reference to some identifiable minimum mode of performance. His solution was to seek to identify a range of reasonable methods of performance, of which the least advantageous to the claimant would provide the measure of damages. I can see the theoretical argument in favour of that approach, if it were viable; but against that, where the parties, intending that there should be a contractual obligation, refrain (for what may be good commercial reasons) from laying down a minimum level of performance or a method for calculating it, I question whether they should be regarded as expecting that the court should seek to construe the contract as if it did, for the purpose of assessing damages in the case of a repudiation, especially given the difficulties of that approach to which I have referred. In such a case, the approach of Atkin LJ seems to me to afford a just way of assessing loss in the event of a repudiation of the contract, and one which might fairly be assumed to reflect the parties' intention.
144. The present case is analogous to Abrahams v Reiach in that the airline's obligation to maintain a x2 based operation was a single obligation but it allowed the airline a wide discretion how it performed it. In the assessment of damages for repudiation of the obligation, I consider that the court is bound by the majority decision in Abrahams v Reiach, which I also respectfully consider to be right for the reasons stated. Alternatively, if we are not bound to follow the approach of Atkin LJ (whether on the ground that there was not a clear majority opinion, and that we are free to adopt the reasoning in the case which we prefer, or on the ground that the decision is at most of persuasive authority when applied to a long term agreement rather than a single transaction – a distinction emphasised by Mr Crane), I consider that we should follow it as the approach best fitted to serve the cardinal compensatory principle in the present instance.
145. I have no doubt that the parties must have contemplated and intended that both aircraft should ordinarily be in daily use, but I can conceive of many eventualities which might lead to the grounding of an aircraft for shorter or longer periods. Besides technical breakdown or routine maintenance, I have referred to wars, strikes, pandemics, terrorism or bad weather and there may be other factors which would affect a flying programme. I do not therefore think that it is possible to specify a minimum number of flights which the airline would be bound to operate throughout the duration of the contract. Even if Mustill J's approach would be the most appropriate way of measuring the claimant's loss if it were viable, I do not see that it would be in the present case.
146. If an assessment of damages would necessarily require construing the contract in such a way as to identify a contractual minimum number of flights, I would accept that the airport would be entitled only to nominal damages, but that would not be just. If it is right to conclude that there was a valid contract which obliged the airline to maintain a x2 aircraft based operation, it may be difficult for a court to assess how the airline would have been likely to have fulfilled that obligation, if it had not repudiated the contract, but that is no reason for not making the best assessment that it can. As Abrahams v Reiach illustrates, the material may be exiguous, although it would be wrong to assume in advance that it will be exiguous in this case. Given the unusual length of the contract and the uncertainty of any prediction about the long term future of the economy in general and the aircraft industry in particular, the court carrying out the assessment may have strong reason to be cautious, but that is a matter for another day. As to the way in which the assessment should be conducted, I agree explicitly with what Patten LJ has stated in para 79.

Issue 4: Ought the judge to have found that it was an implied term of each agreement that the airline would operate the two aircraft based at the airport in a manner which was reasonable in all the circumstances?

147. On the hearing of the appeal Mr Brealey QC advanced his implied term argument as a fall back position. His primary submission was that on the basis of the judge's finding that on the terms of the two agreements the airline was obliged to establish by 30 April 2006, and thereafter to continue, a two aircraft based operation at the airport, he ought to have ordered an assessment of damages resulting from the airline's repudiatory breach. If the court accepted that argument, he did not seek to argue for the introduction of an implied term. It is therefore unnecessary to address the point further. However, I also agree with Patten LJ that it is unnecessary to imply such a term. The court conducting the assessment will assume that the airline would have acted as a sensible commercial party, taking proper account of its own interests but not acting uncommercially in order to disadvantage the airport. It is in my view unfortunate that the argument about implied terms was introduced into the case by the airport in the way that it was, and this seems to have led to a misfocusing of the arguments at the trial. The judge referred to the question of implied terms as to him the real difficulty in the case and he devoted approximately a third of his judgment to it. His rejection of the argument on implied terms led him to give judgment for the airline notwithstanding that he had found that, without the implied terms, the airline had repudiated its contractual obligations. I cannot help feeling that his dismissal of the claim resulted from the airport's unsuccessful attempt to introduce an implied term or terms when none was needed. On the appeal Mr Brealey wisely adopted a different approach from the way in which he apparently argued the case before the judge.

Conclusion

148. I agree with the disposal proposed by Patten LJ.

Lord Justice Mummery :

150. I agree with both judgments.