GOARM ENGINEERING LTD. v. SHIMIZU CORPORATION

HCCT21/1999

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

CONSTRUCTION AND ARBITRATION PROCEEDINGS
NO.21 OF 1999

------------------------

------------------------

---------------------

J U DG M E N T

---------------------`

1. In May 1996 the defendant ("Shimizu") was appointed as the main contractor for the construction of a 43-storey residential block at 11 May Road, Hong Kong called Valverde Tower. Originally the excavation and structural frameworks sub-contract was awarded to Amen Construction Company ("Amen") in August 1996. That sub-contract was terminated by Shimizu in about January 1997. Thereafter the plaintiff, Goarm Engineering Limited ("Goarm"), stepped in and completed the excavation and structural frameworks.

2. The primary issue on liability in this case is the terms upon which Goarm agreed to take over from Amen and finish the job. There is no doubt that Amen's sub-contract was in the form of a written contract for a lump sum price with an agreed completion date together with a clause for liquidated damages for delay. It is the plaintiff's case that it took over from Amen on very different terms.

3. In a nutshell the primary issues on liability are (a) if, as Shimizu contends, Goarm agreed :

(i) to a fixed lump sum payment of $7 million to finish the works and

(ii) to be bound by the calculation of rates and quantities as were provided for in the Amen sub-contract and

(iii) to a completion date for the contract being 30 June 1997.

or (b) as Goarm contends, the parties verbally agreed that Goarm would complete the remaining works as quickly and as efficiently as possible, i.e. within a reasonable time and at a reasonable price.

A. LIABILITY

1. Shimizu's case

4. Shimizu's pleaded defence is that Goarm verbally agreed to be bound by all the terms and conditions in Amen's sub-contract save that the price was to be calculated according to the balance remaining from Amen's sub-contract after it (Amen) had left the site. Such a sum was $7 million. Shimizu also pleaded that Goarm agreed to a specific completion date, 30 June 1997. The documentary evidence upon which they primarily rely is a 'Form B1" which, Shimizu submits, evidences the contract. This was signed, for Goarm, by Ms Leung Siu Lan on 18 March 1997. It is a single piece of paper which contains reference to the contract sum being "$7 million - Provisional" and the completion date. Shimizu concede that this was one of a number of similar documents which had to be signed by Goarm before any interim payments could be released to them.

5. In fact, Goarm was paid a total of $20,135,073 on ten progress payment applications. Shimizu contend that, after adjustments, it has overpaid them $10,135,073 ($3 million was estimated for variations).

6. In addition, Shimizu has pleaded a substantial counterclaim. Firstly it seeks $12.6 million in liquidated and ascertained damages. This would be due if there was a completion date of 30 June 1997 and 126 days of delay. The term in the Amen sub-contract being $100,000 a day for delay. Shimizu also claim contra-charges of just under $1 million and the cost of rectifying defects at $172,000. In all, Shimizu claims that Goarm's liability to them is approximately $24 million. On the face of it, considering Goarm took over from an unsatisfactory sub-contractor and finished the excavation and structural frameworks on an up-market 43-storey residential block in May Road, without significant complaints or problems, this is surprising.

2. Goarm's case

7. I will deal with Goarm's case in more detail when I refer to the more important aspects of the evidence hereafter. In short, they say there was no sub-contract incorporating Amen's terms, nor any other specific terms, save an agreement to complete the works in a reasonable time and at a reasonable cost. Ten interim payments were made - all were adjusted and then paid. The final result was that they were paid $20,141,073. They say their net costs were in fact $21,805,628. They were entitled to make a 15% profit which results in them being underpaid by $4,941,400. This is their claim.

3. Evidence

8. As will be seen hereafter, having seen and heard the witnesses of fact, I have concluded that I should find in Goarm's favour on all key issues of liability.

9. Goarm called two factual witnesses, Mr Wong Chi Ming, the executive director of Goarm and Ms Leung Siu Lan, Goarm's accounts officer. Shimizu called four factual witnesses; Mr Yau Chung Ming, a site agent, Mr Allen Li Kwok Wah, the project quantity surveyor until July 1997, Mr Gordon Mak Ting Wai, who took over from Mr Allen Li and Mr Ricky Yeung Wai Kwok, a supervisor. All were employees of Shimizu at material times.

10. Mr C.M. Wong was undoubtedly the key witness. He spent three days giving evidence at trial. All four defence witnesses took less than two days in total. Generally speaking I found Mr Wong to have a good recollection of events and he gave evidence in a straightforward manner. I found him to be a reliable witness. He had been directly involved in discussions and negotiations with Shimizu from beginning to end. The defence witnesses, however, were less "hands on". On a number of occasions they were unable to remember things due to the passage of time. They were only brought back into this litigation relatively recently in order to make their witness statements. Their lack of certainty on certain matters gave an impression of evasiveness. On key issues of fact, I preferred the evidence given by Mr Wong for the plaintiff.

11. The defence case was in fact marred, not so much by the witnesses who they did call but by the fact that two particular witnesses were not called and had not made witness statements. One stage higher up the Shimizu hierarchy in Hong Kong at the time, stood a Mr Kobayashi and a Mr Miki. The former, in particular, was directly involved with Mr Wong at all material times. They had known each other for many years previously, as Mr Wong had himself been an employee of Shimizu prior to setting up his own firm, Goarm. They were on good terms and had a good working relationship. There was no evidence from either of them to challenge or contradict Mr Wong's evidence concerning their discussions concerning the basis upon which Goarm would take over from Amen. This serious weakness in the defence case was not repaired by the evidence which was in fact called.

12. Both counsel, Mr Martyn Richmond for Goarm and Mr Danny Choi for Shimizu, agree about what the court must decide on issues of liability. It comes down to five questions, each of which I propose to deal with with relative brevity. They are all questions of fact.

(a) What was the nature of the original agreement?

13. As a matter of background it is important to remember that Goarm had made a tender at the same time as Amen. Amen won the tender. Amen had been very slightly cheaper than Goarm but Amen's tender had been for excavation and structural frameworks whereas Goarm's tender had not included excavation work. In reality, therefore, there was quite a difference between their original tenders. Amen's work started in August 1996.

14. By January 1997, Mr Kobayashi had decided to terminate Amen because he was not satisfied with their progress (there was 73 days delay on the Podium works and 43 days delay on the Tower works when Goarm took over). Mr Kobayashi engaged Goarm to continue the project. Mr Wong had been his deputy project manager for many years. There was an atmosphere of trust and goodwill between them.

15. Shimizu was obviously desirous of as smooth and as quick a take over as possible. Continuity and speed were uppermost in Shimizu's thinking. Just one relatively brief meeting took place before Chinese New Year 1997. A lot of material resources and manpower remained on site, Mr Kobayashi successfully negotiated a termination settlement with Amen, and Goarm stepped in.

16. I accept Mr Wong's evidence, uncontradicted by any witness present at that meeting, that he did not agree a lump sum contract price at that meeting, nor did he agree a completion date. He made it clear to the court, and I am sure he made it clear at the time that to complete the works for $7 million by 30 June 1997 was out of the question. I accept also that details such as interim payments, retention monies and so on were not agreed either. Mr Wong did not have the time to do a detailed re-tendering exercise. It was plain that Mr Kobayashi wanted Goarm to carry on, on site, as quickly as possible.

(b) Was there any agreement about price?

17. The simple fact is that neither Mr Wong nor any other contractor could have or would have agreed a fixed price of $7 million or any other similar figure in January 1997 to complete this contract.

18. Goarm submitted interim payment applications on a monthly basis from the outset. By the original completion date of 30 June 1997 costs had already exceeded $11 million. Shimizu did not express any alarm or concern about the "fixed price" being exceeded. This demonstrates that the price was being negotiated on an on-going basis. By the time of the fifth payment on 17 June 1997 Shimizu had paid Goarm $7.7 million for work done, approved and certified. Even the defence evidence, through Mr Allen Li, conceded that by June 1997 the lump sum figure and Amen's rates were redundant.

19. It was at this time that Mr Gordon Mak took over Mr Allen Li's job. He continued to receive and consider Goarm's payment applications. He adopted a very rigid approach. He accepted that discussions and negotiations were taking place about Goarm's rates and quantities, thus accepting that Amen's rates were not written in stone, however, in the absence of specific instructions as to agreements, he simply adopted and paid out at Amen's old rates. As such a dispute was inevitable.

20. It is interesting to note that from the third application to the tenth (final) application (namely between May and December 1997) the difference between Goarm's claim and Shimizu's payment steadily decreases. In May, Goarm say they were owed $6.165 million. By December it was down to $1.66 million, on this contract. This trend reflects the fact that negotiations and discussions about rates and quantities were on-going and that increasing agreements about some rates and quantities resulted in a narrowing of the differences between them. At the end of the day Mr Wong very fairly said, in evidence, that if Shimizu had offered him something over $3 million to settle the final account, he would have accepted it and this expensive litigation would have been avoided. He refused the $101,917 which Shimizu offered. It seems that thereafter the parties got further apart rather than closer together.

(c) Was a completion date agreed?

21. Again, the answer to this question is no. Goarm inherited delay from Amen. Amen had been on site for five to six months of what was originally intended to be a 10-month contract. The time available to Goarm from its commencement of work to 30 June 1997 was about five months. It would have been fanciful to think that anyone could have both caught up Amen's delays and finished the work in that time. In approximate terms two-thirds of the Podium work was outstanding and about 55% of the Tower work was outstanding. Both were significantly behind schedule.

22. Mr Wong's evidence was consistent. His evidence concerning the completion date was the same as his evidence concerning the price. The only thing he could sensibly agree to was to complete the work within a reasonable time and at reasonable cost.

(d) Was Goarm, at any stage, bound by any of the terms in the Amen sub-contract?

23. For substantially the same reasons as already stated, the answer is again no. Given the court's findings thus far, it would be illogical to find otherwise. Amen's sub-contract was a lengthy document. Had Shimizu wanted it to apply to Goarm or even apply in part to Goarm, it would have been a relatively easy task to put such an intention in writing to Goarm by correspondence. Neither was any letter of intent sent to Goarm.

24. In short, the evidence on this issue is somewhat one sided. The only defence witnesses who might have been able to challenge Mr Wong's assertion were not called.

(e) What is the effect of Form B1?

25. "Form B1", upon which Shimizu rely as evidence of a contract, is a single page document and is headed "Form B1 Sub Contractor's Copy" . It was accepted by Shimizu that a similar document was generated and signed in respect of each interim payment. For Goarm it was signed by an accounts officer, Ms Ada Leung, who attended Shimizu's offices to collect each interim payment. On each occasion she was required to sign Form B1 before the cheque was released to her. She was not authorized by Goarm, nor did she have ostensible authority, to sign contractual documents on Goarm's behalf. Her signature is evidence of collection of money and no more.

26. With or without her signature the question remains - it is evidence of what the parties agreed. It contains, inter alia, the following :

"'Quotations Amount HK$7,000,000 (Provisional)'

'Period of Job 13/2/97 to 31/6/97'"

27. Given that the nature of the document is, in truth, a Shimizu accounting document and given my findings concerning the true nature of the agreement between the parties, Shimizu's contention that Form B1 is evidence in support of a different agreement, must fail.

B. QUANTUM

1. General approach

28. Given my findings on liability Goarm is entitled to be paid on a "quantum meruit" basis. In deciding how to approach a quantum meruit assessment in this case, it is unnecessary to go beyond basic principles.

29. Quantum meruit denotes the payment of a reasonable sum. Chitty on Contracts, 28th Edn states :

" The courts have laid down no rigid guidelines to be applied in the assessment of a reasonable sum although it is clear that the contractor should be paid a fair commercial rate for the work done in all the relevant circumstances. The basis for any such assessment of a reasonable sum can be a source of controversy, since the employer may wish to confine recovery to the contractor's actual or tender costs, whereas the contractor may want the reasonable sum to reflect the value which his work ultimately represented to the employer. Although in some instances a reasonable sum will be calculated on the basis of actual cost plus an uplift for profit and overheads, it will be a matter of importance in the particular case whether, for example, the contractor who has obtained his resources at an especially low cost will be obliged to share that benefit with the employer, or whether a more objective view of 'fair commercial rate' should be adopted."

30. In this case, a fair approach is to award Goarm a sum equal to its net costs reasonably incurred (after deductions for defects and contra charges) plus a reasonable profit. If this figure turns out to be similar to the amount that Mr Wong says he would have settled for at the time of the dispute, it would be a comforting pointer towards its reasonableness.

2. Goarm's performance

31. There is no evidence from witnesses of fact which suggests any serious criticism of the quality of Goarm's work. Goarm finished the work on 25 November 1997. In some respects they made in-roads into Amen's delay in other respects they did not. It is not necessary to consider the question of delay in the context of Goarm's performance for two reasons. Firstly, as quantum is being assessed on the basis of there being no contractual terms, Goarm cannot be liable for any liquidated damages. Secondly, as a matter of fact, Shimizu did not pay any liquidated damages for delay to the employer. Apparently, the matter was negotiated and settled without Shimizu paying anything. The details of the settlement are not known. It is reasonable, however, to assume that the employer is unlikely to have abandoned damages to which he would otherwise have been entitled. This being the case, it would be wrong to criticize Goarm's performance because of a delay for which Shimizu never had to pay.

32. In this light, it is helpful briefly to consider Goarm's performance as part of the court's task of assessing the reasonableness of their claim.

33. I find one particular letter from Goarm, written by Mr Wong, to Shimizu's management to provide considerable assistance on this issue. It was written on 3 July 1998 when Mr Wong had become increasingly frustrated that his final account was not be paid. It is a nine-page letter which Mr Wong concedes he had help in writing, from a friend who was not a solicitor or barrister.

34. I have read the letter carefully. It deals with all the areas of dispute. It is too lengthy to recite herein but I find it to be comprehensive in its content, measured in its language, reasonable in its requests and a fair and honest submission of Goarm's understanding and performance.

35. Shimizu did not specifically reply in writing although some further negotiations followed. The letter's penultimate paragraph is as follows :

"We first agreed to participate in this project to provide much needed assistance and with an eye to maintaining a good and close working relationship with Shimizu Corporation. From the evidence provided above, it is patently obvious that we have acted in an extremely reasonable manner, our attitude has been one of co-operation and support and whenever possible, we have gone out of our way to assist Shimizu to overcome the innumerable problems encountered on this project. All that we seek in return is your acknowledgement of our reasonable entitlements, agreement of our final account and payment of the outstanding balance."

3. The expert witnesses

36. Three experts were called to give evidence. All had prepared detailed and helpful reports. Mr Paul Barrett was Goarm's expert. Mr John Battersby was Shimizu's. Additionally, Shimizu called Mr John McNaught who had reported on the separate issue of delay and liquidated damages and the extent to which delay was attributable to Goarm's performance.

37. Expert reports from quantity surveyors are the result of considerable work, meticulous attention to detail and very time consuming trawling through an enormous volume of documents. Mr Barrett's and Mr Battersby's reports are both over 60 pages in length. Both are commendably concise bearing in mind that the litigation has generated 26 arch files of documents which Mr Battersby has reduced to 725 pages of appendices attached to his report.

38. Having read the reports and having heard the experts' evidence, the court's task is to make a judgment. In this case, bearing in mind my findings on liability, the judgment to be made is to decide how much Goarm deserves to be paid for its work. The parties themselves have never been able to agree the answer to this question and neither can the experts.

39. The result of Mr Barrett's detailed analysis of all the costs incurred on the project and his opinion on their reasonableness is that reasonable net costs amounted to $21,805,628. He also considered an alternative approach in which he added his figure for direct costs to his calculation for overheads and site preliminaries (indirect costs) and arrived at a not dissimilar figure of $22,111,626. His entire report is based on the assumption that this was a "quantum meruit" case. He was "instructed to examine the manner in which the financial claims of the plaintiff set out in the Statement of Claim have been calculated and to comment on their reasonableness and composition".

40. Mr Battersby on the other hand conducted, on instructions, a wider investigation. After setting out all the facts (actual and assumed) upon which his opinions were based he did three separate exercises which were :

(a) "The amount to be paid in respect of the works carried out by Goarm on the basis that there was a sub-contract executed between Goarm and Shimizu in which the sub-contract price was a lump sum price of HK$7,000,000 subject only to adjustment for variations.

(b) The amount to be paid in respect of work carried out by Goarm on the basis that there was a sub-contract executed in which the sub-contract price was provisional and was to be re-calculated on the basis of :

.1 Amen's Sub-Contract rates; or

.2 rates agreed between the parties and reasonable rates (fair valuation) in absence of agreed rates; or

.3 fair valuation.

(c) The amount to be paid in respect of work carried out by Goarm on the basis that there was no contract between Goarm and Shimizu and the work carried out by Goarm is to be paid for on a quantum meruit."

41. In view of my decision on liability only (c) is now relevant. As a result of his investigation of (c) he produced a six-page schedule in which he listed approximately 80 items of work and in respect of each item set out Goarm's claim (quantity, rate and amount), Shimizu's payment (quantity, rate and amount), the ASD rates for that work and finally his own opinion as to the appropriate quantity and rate and amount under each item. Finally he calculated the amounts that would have been paid based on Goarm and Shimizu's quantities but using his rates (which was largely ASD rates adjusted downwards). Mr Battersby's final "amount" calculation had taken the quantities from the drawings supplied to him. All calculations had been done by his staff but checked by him on a random basis. This helpful and illuminating exercise produced the following result. Goarm had claimed $21,199,333. Shimizu had paid $18,901,462 and Mr Battersby's calculation suggested that they should have been paid $12,728,573.

42. Generally speaking, the reason Mr Battersby's figure is so much lower is because of his figures for rates. He explained how he arrived at his figures for rates and no criticism is made of his method. However, on many occasions, Goarm's and Shimizu's figures for rates were much higher and were the same. The fact that they were the same is compelling evidence that they had been agreed. Mr Battersby accepted that if the parties had agreed a higher rate then "so be it". His rate was his opinion on what a reasonable rate should have been.

43. One particular item in his schedule merits particular mention, namely "excavation". It was the largest item in the entire project. Goarm claimed 6,526 cubic metres at $750m3. Shimizu allowed 6,240 cubic metres at $750 m3. Goarm had claimed $4.89 million and had been paid $4.68 million. Mr Battersby measured the quantity at 5,414 m3 and allowed a rate of $154.8 m3 resulting in a payment of $0.84 million. In evidence he very fairly conceded that his rate was based on the excavation of soft soil whereas, in fact, there had been a mixture of soil, rubble and hard rock. Accordingly he made a substantial upward variation to his figure. He acknowledged that if the rate of $750 had been agreed the real difference between the parties was in quantities only and the parties lower amount (Shimizu's) was about $4 million higher than his. Taking a similar item, formwork, the same exercise results in a difference between an agreed amount between the parties of $1,095,840 and Mr Battersby's calculation of $175,892.

44. Looking at the schedule as a whole, if one assumes that where the parties rates are the same they had been agreed and that that rate is used in preference to Mr Battersby's adjusted downward ASD rate, Mr Battersby's final figure would not be $6 million lower than Shimizu's but would be in the same region.

45. Mr Battersby's schedule is helpful in that it gives the entire picture. His method of calculation was consistent. However, in the court's search for "an amount which the plaintiff deserves", I think Mr Barrett's approach provides a better starting point. I merely highlight five factors why Mr Battersby's final figure of $12.78 million takes second place. Firstly, he very fairly made upward concessions in the course of his evidence. Secondly, the ASD rate is more of a benchmark for tendering in maintenance contracts than a widely used rate on site. Thirdly the adjustments downward of the ASD rate would not have been the starting point for discussion on this project which was an up market residential block in May Road. Fourthly, the best evidence of measurements must be Mr Wong's who was the only witness who actually calculated measurements on site. Fifthly, the circumstances in which Goarm took over this project, namely as a matter of urgency with a desire towards continuity and a resolve to catch up Amen's delays, were valid grounds for higher rates being agreed.

4. The court's approach

(a) Costs

46. Mr Wong would have been content had his final account been paid. Such a payment would have been of a lesser sum than is now being claimed. He said in evidence that he would have accepted that sum to maintain goodwill but now that any pre-existing goodwill has been smothered by this dispute, he claims his reasonable costs plus 15% profit. This is not an unreasonable stance. The fact of the matter is that Shimizu did not pay up and did not offer more than about $100,000. No progress was made towards a settlement and so Goarm now makes its quantum meruit claim.

47. However I find it difficult to accept that had Goarm's final account been paid they would have only regarded themselves as having achieved a "break even" situation, as contended by Mr Wong. More likely, in my judgment, they would have regarded it as a successful project upon which they had made money.

48. The amount they would have been content with was $24,540,458 (for all works). In this action their claim is for $25,076,473 which includes 15% for profit (based on Mr Barrett's net costs). It seems obvious that Goarm's invoices contained a profit element.

49. The court's starting point is to take Mr Barrett's revised costs figure of $21,805,628. In my judgment, it is reasonable to make a downward adjustment. The reason for so doing is twofold. Firstly to acknowledge the genuineness of the competing arithmetic. Shimizu's sums or Mr Battersby's sums should not be arbitrarily ignored. They have been conscientiously gathered and prepared. The reality is however that when two parties are locked into a dispute it is both inevitable and understandable that the plaintiff's case on quantum is presented in its most favourable light. The plaintiff starts from the optimum scenario and works downwards, the defendant starts from the worst scenario and works upwards. Secondly, Goarm's own invoices would contain a small profit element. Mr Wong did not deny this in evidence

50. With these general considerations in mind and in a desire to achieve reasonableness I have decided to reduce Mr Barrett's figure by 5%. The figure for net costs therefore becomes $20,715,347.

(b) Profit

51. 15% has been claimed. Mr Barrett broke this down further as to 13% pure profit and 2% office overheads. Given the time when this work was being done, 1997, such a profit might have been regarded as slightly optimistic. Had Goarm's costs been as calculated by Mr Barrett ($21.805 million) and had Shimizu paid Goarm's final account ($24.540 million) the profit resulting would have been approximately almost exactly 12.5%. I regard this as a better percentage to work on. $20,715,347 plus 12.5% is $23,304,765. Therefore the balance due to Goarm's on their claim, after deduction of the amount paid ($20,135,073) is $3,169,692.

(c) Counterclaim

52. Three matters fall for consideration under the counterclaim.

(i) Delay

53. As I have found there to be no agreement between the parties whereby the terms of Amen's sub-contract were incorporated there can be no claim for liquidated damages. Mr McNaught's helpful report was only concerned with this issue. It therefore requires no further comment.

54. Neither do I consider it proper to reduce Goarm's claim for prolongation costs. As an alternative to liquidated damages Mr Battersby has made a calculation for prolongation costs. He has concluded that $3,008,537 represents a figure for loss incurred by Shimizu as a result of having its preliminaries (site overheads) resources on site longer than necessary. The burden is on Shimizu to prove loss flowing from prolongation. Mr Battersby's calculation is the high water mark of that loss. It is based on Goarm being liable for 81 days of delay. Mr McNaught opined that a measure of delay was "attributable" to Goarm but expressly stopped short of saying they were "liable". There could have been other factors, such as the performance of other trades or inclement weather, which Mr McNaught was not instructed to investigate. In short, I am not satisfied that Shimizu have proved any prolongation costs to the required standard.

(ii) Defects

55. As a measure of the standard of Goarm's work it is worthy of note that only one issue arose under the heading of defects. A sum of $172,000 is counterclaimed by Shimizu for rectification work on the concrete slabs in the car park.

56. As soon as it was first raised Goarm disputed liability for it. However they did not simply dig their heels in, they suggested an independent report be obtained by which they would abide. Shimizu however did not respond to this suggestion, they merely pressed their claim. In my judgment they have not sufficiently proved that such defects as there were in the car park fell to be rectified by Goarm.

(iii) Contra-charges

57. Goarm concede that $95,514 for contra-charges should be deducted from their claim. Shimizu on the other hand claim $922,854 should be deducted under this heading.

58. Mr Battersby's evaluation of contra-charges comes to $653,624. Mr Gordon Mak, in his witness statement, states it to be $715,874. He has added a 15% handling charge. Mr Battersby considers 15% to be "high". Shimizu's claim is based on a 20% handling charge. All this demonstrates that there are a variety of different calculations on contra-charges. Mr Wong has gone through every item on the list of contra-charges, of which there are about 170 in total, and has given a percentage (ranging from 0% to 100%) to denote what he considers to be his liability. A few of the items were the subject of examination at trial. There are four items which are significantly larger than the others. They are $35,000 for repairs to a damaged curtain panel, $106,000 for purchase of plywood, $106,922 for purchase of material and $14,516 for repair to a concrete mixer. Mr Wong has put 0% against all of these.

59. I consider the only way to deal with this issue is to take a general approach. It would not be right to make a judgment on one particular item and speculate about many other small items about which no evidence has been given. In the course of his evidence Mr Battersby, (on another issue) said that sometimes the best one can do is to "split the difference". I consider that to be a reasonable approach to this issue. I do so on the factual finding that Mr Wong has underestimated his liability. By his letter of 14 January 1998 he submitted his schedule and assessment of $95,514 for Shimizu's "review" and to "arrange a further discussion". Because of the history of the matter it now falls to the court to make that review based on limited information. I take $95,514 as Goarm's figure and $653,624 (per Mr Battersby) as Shimizu's. A median figure between the two is $374,569. I allow Shimizu's counterclaim for contra-charges in this sum.

60. On the claim there will be judgment for the plaintiff in the sum of $3,169,692. On the counterclaim there will be judgment for Shimizu in the sum of $374,569. The net effect is a liability on Shimizu to pay Goarm $2,795,123.

61. I award interest on this sum at the rate of prime plus 1% from the date of the writ.

62. As Shimizu has only succeeded on a very small part of its counterclaim, the costs order nisi I make is that the costs of the claim be to the plaintiff and that Shimizu pays 90% of Goarm's costs on the counterclaim.