SILK PLUS DYEING FACTORY (H.K.) LTD. v. LUCKY FACT GARMENTS LTD.

HCA 1185/1996

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO. 1185 OF 1996

____________

_______________

J U D G M E N T

_______________

Background:

1. The plaintiff operates a dyeing business with a factory in Dongguan called Modern Silk Dyeing and Washing Factory (hereinafter called "Modern Silk"). At the material time, there were three directors in the plaintiff company. They were Mr Chan, Ms Tiu and Mr Cheung. Mr Cheung was also the manager in charge of Modern Silk. His brother Mr Derek Cheung was the sales representative responsible for the transactions which formed the subject matter of the present proceedings.

2. The defendant is a garment manufacturer with a subsidiary in Dongguan called Sunrise Fashions Limited (hereinafter called "Sunrise"). It sub-contracted part of its garment manufacturing work to sub-contractors, one of which was called Shenzhen Hua Gan Clothes Co Ltd (hereinafter called "Hua Gan"). The defendant had placed orders with the plaintiff for dyeing and processing of fabric since 1990 until July 1995.

3. The plaintiff is now claiming $746,969.21 in respect of work done and materials supplied between August 1994 and July 1995 under 86 invoices. The defendant admits liability for these invoices but counterclaims $949,943.00 against the plaintiff in respect of damages for the plaintiff's defective workmanship under two of these invoices, number H0938 and H0980 dated 21 and 30 December 1994 respectively. The first invoice was related to the Defendant's purchase order No. 01080 for dyeing and sandwashing 28,322 yards of fabric. The second invoice was related to the Defendant's purchase orders No. 02205 and 02206 in respect of 2,079 yards of similar fabric. The defendant's merchandiser, Ms Connie Tsang, was responsible for following up on these purchase orders with Derek Cheung of the plaintiff.

4. Part of the fabric processed under these three purchase orders were used for making 5,200 ladies' blazers for export to the defendant's client, Otto Versand of Germany (hereinafter called "Otto"). However, Otto only accepted 2,030 pieces. The rest of the blazers were rejected by Otto because of colour shade variations. Subsequently on 22 April 1995, another shipment of 1,076 pieces was delivered. While the defendant claims this as a shipment under a separate replacement order, the plaintiff suggests that it was part of the 5,200 pieces delivered under the three purchase orders.

5. The defendant withheld payment in respect of the processing charges under these 86 invoices while negotiating with Derek Cheung. In about April 1995, Connie Tsang left the defendant's employ. In July 1995, Derek Cheung also left Modern Silk with his brother to start their own factory, taking with him the business the plaintiff had been doing with the defendant. He has since maintained social contact with Connie Tsang.

6. Upon the defendant's refusal to pay the 86 invoices, the plaintiff instituted the present action. The defendant then counterclaimed the FOB price of the rejected garments. A special feature of this case is that while one would expect Connie Tsang to be an important witness for the defendant on its counterclaim, she was actually called as a plaintiff's witness to rebut the defendant's counterclaim. As the defendant did not dispute the plaintiff's claim in respect the 86 invoices, the defendant was given the right to open its case on the counterclaim as if it were a plaintiff.

Defendant's case:

7. According to Mr Wong, the defendant's director, Connie Tsang had reported to him that some of the fabric processed by the plaintiff had serious colour shade variation and he told her not to accept the fabric. A few days later, Connie Tsang informed him that Derek Cheung requested the defendant to use the fabric by avoiding the defective areas (this process is hereinafter called "avoidance cutting") and undertook to compensate for the FOB price of any garments rejected. Wong agreed. The defendant's counterclaim is based on the plaintiff's defective workmanship in dyeing and sandwashing the fabric under the defendant's three purchase orders (hereinafter called the "1st set of agreements"); or alternatively, on Derek Cheung's request in December 1994 to the plaintiff to perform avoidance cutting and his undertaking to pay the FOB price of any garments rejected by Otto (hereinafter called the "request and undertaking").

8. Mr Chiu for the defendant contended that the main thrust of the defendant's counterclaim is on Derek Cheung's request and undertaking in December 1994 under which the defendant counterclaimed the FOB price in respect of 3,287 pieces of rejected garments. If the request and undertaking was not proved, then the defendant would fall back on the 1st set of agreements. He reasonably conceded that under the 1st set of agreement, the defendant was only entitled to nominal damages, as proof of actual loss was wanting. In any event, the 1st set of agreements provided the necessary factual background giving rise to the alleged request and undertaking by Derek Cheung.

9. The manufacturing work for the entire lot of 5,200 pieces of garments was sub-contracted to Hua Gan. Towards the end of the lunar year, Wong accompanied Otto's quality control inspector, Mr Johnny Ng, to inspect the 5,200 pieces of garments in Hua Gan, as he anticipated problems with the garments because of colour shade variation in the fabric. Upon a random inspection, Johnny Ng wanted to reject the entire lot of garments because of colour shade variation in the garments. Wong, however, persuaded him to perform a piece by piece check. Eventually, 2,030 pieces only were accepted. 850 pieces were shipped on 30 January and 1,180 pieces were shipped on 20 February 1995. From the two lots shipped, Otto further rejected 117 pieces. Thus only 1,913 pieces were accepted and 3,287 pieces were rejected. Derek Cheung and his brother were aware of the rejection of the garments by Otto.

10. The defendant tried to salvage the rejected garments by re-patching the pieces with the wrong colour shading. Only a few pieces were thus salvaged. Connie Tsang ordered additional fabric to be dyed by other factories as she considered the plaintiff's workmanship not up to standard. The new fabric was used in making 1,076 pieces of garments shipped on 22 April 1995. This was a delivery under a separate replacement order and did not form part of the 5,200 pieces under the three purchase orders.

11. Wong has since liaised with Derek Cheung about compensation. After Derek Cheung left Modern Silk in July 1995, he continued visiting the defendant to procure business for by his brother's factory. Derek Cheung also continued to liaise with him about compensation and requested a letter from Otto as confirmation that the garments had been rejected because of colour shade variation. However, subsequently Derek Cheung told him that there were difficulties from the partners in Modern Silk. After receipt of a letter before action from the plaintiff's solicitors, Wong had a meeting with Chan and Derek Cheung to discuss the plaintiff's claim. There he informed Chan that he would counterclaim for damages as a result of the defective dyeing. Derek Cheung kept silent throughout the meeting.

12. Ms Ada Lam, the Senior Merchandiser, corroborated Wong's evidence. When Connie Tsang left in April, Connie Tsang told her that these three orders with the plaintiff had been settled. She did not contact Derek Cheung about this matter as she assumed everything was in order. She said that the practice was for the plaintiff to count the garments rejected and take delivery of the garments. Then a debit note would be issued to the plaintiff to be set off against other sums due to the plaintiff. However, she said Derek Cheung never came to count the garment, though he never declined liability.

Plaintiff's case:

13. According to the plaintiff, the fabric was dyed in accordance with the colour sample attached to the purchase orders. It is not unusual that defects, such as uneven colour shading would occur between different rolls of fabric or sometimes even within the same roll. It does not dispute that some of the fabric dyed had been returned for re-processing, but alleged that the problem had been cured after re-dyeing. It denied ever requested the defendant to perform avoidance cutting. According to Connie Tsang and Derek Cheung, the three orders progressed very smoothly.

14. After Derek Cheung left the plaintiff's employ, Chan followed up with the defendant for payment. He could not get in touch with Wong despite repeated attempts and fax. Eventually, towards the end of 1995, he instructed his solicitors to issue a letter before action to the defendant. In the meantime, Derek Cheung continued visiting the defendant to solicit business for his brother's new factory, but he heard no complaint from Wong about defective fabric until Wong received the letter before action. He then arranged a meeting between Chan and Wong to resolve the dispute. At the meeting, Wong said he would counterclaim damages for the plaintiff's defective dyeing but did not mention anything about the alleged request and undertaking by Derek Cheung.

The issues:

15. The issues are whether the plaintiff had failed to dye the fabric in accordance with the sample provided by Connie Tsang or within the lot colours approved by her; and whether as a result Derek Cheung requested the defendant to perform avoidance cutting and offered an undertaking to pay the FOB price of any garments rejected by Otto.

16. The defendant's expert, Mr Wicky Wong had not compared the rejected garments with the colour sample or lot colours approved by Connie Tsang. As such, the defendant's case was mounted on the basis of colour shade variation between the different panels in the finished garments. The defendant sought to rely on that as evidence of uneven dyeing of the fabric. Secondly, it sought to rely on the alleged report by Connie Tsang that the fabric delivered by the plaintiff was defective and thirdly that as a result Derek Cheung allegedly requested avoidance cutting and undertook to pay the FOB price of any garment rejected. These allegations were denied by Connie Tsang and Derek Cheung.

Credibility:

17. For reasons as will become apparent in the later part of this judgment, I find that apart from Chan and the defendant's expert witness, Wicky Wong, none of the witnesses were telling the whole truth.

18. Wong was and is the boss of the defendant company. He was in charge of production and travelled regularly to China to visit his subsidiary factory. He did not impress me as having command of all the facts which one would expect him to have especially when mounting a case of this nature. This is certainly not a case in which his merchandisers should have exclusive knowledge of the events. He did not stand up to cross examination. When being questioned as to whether 3,170 pieces of garments had been rejected by Otto, he had to admit that he was probably mistaken. I find him insincere and his evidence exaggerating.

19. Ada Lam was evasive and appeared to be impatient under cross examination. Her evidence is inconsistent with the documentation. She did not impress me as an honest and credible witness either.

20. Ms Koopmann certified that of the 5,200 pieces of garments ordered, only 2,030 pieces were accepted with the balance of the delivery cancelled. This is only partial truth as there was a third shipment delivered under the same orders on 22 April 1995 for 1,076 pieces. There is no reason why she should have overlooked that shipment when writing the letter on 22 January 1996 as she said in that letter that the balance of the order was cancelled. That I assume was an oversight or the result of undue reliance on the defendant's information rather than deliberate dishonesty. This shipment does assume some importance in my deliberation.

21. Wicky Wong, the defendant's expert, who examined the garments rejected by Otto, can be regarded as an independent witness. His evidence was not seriously challenged. He examined the rejected garments with his unaided eye. His conclusion about colour shade variation in the finished garment was reached on the basis of his subjective observation and not supported by any scientific or objective data. But I have no doubt about his honesty and credibility and accept his evidence.

22. Chan impressed me as an honest businessman. He did not appear on the scene until after the issue of the letter before action. However, the conversation which took place between him and Wong in the meeting in December 1995 is significant. The plaintiff pleaded two exemption clauses displayed in its invoices and stickers attached to the dyed fabric which excluded liability if no complaint was made within three days of delivery or if the fabric had been cut. However, in evidence, like a fair and sensible businessman, Chan frankly admitted that the three day limitation was just of an advisory nature and would not be enforced, and that depending on circumstances the plaintiff would also entertain complaint after the fabric had been cut. It appears that the defence of exemption clause as pleaded is more of a counsel's wisdom than of the true intention of the plaintiff. I consider Chan an honest and responsible witness. I accept his evidence.

23. Connie Tsang appeared unimpressive. She was evasive and answered most of the questions asked of her equivocally by saying "I cannot recall". She has no recollection of the rejection of the defendant's garments. She was the merchandiser in charge of that order. That was a big order from an important client and her last big order before her resignation. It is unlikely that she would not have known about such a large scale rejection by Otto. She said that the defendant's dyeing orders placed with the plaintiff progressed very smoothly. That could not be true in view of the indisputable colour shade variation. I do not consider her a wholly credible witness. I treat her evidence with greatest caution. However, upon the final analysis and in the light of the totality of the evidence, I find her credible on the major issues.

24. Derek Cheung appeared to be quite forthcoming in the witness box. However, he left the plaintiff's company with his brother under surreptitious circumstances. They quit by giving the plaintiff one month notice and leaving the plaintiff's factory crippled. Immediately, they operated a competing business, which must have been set up while he was still working for the plaintiff. Not only that, Derek Cheung also diverted the plaintiff's business with the defendant to his brother's factory. All these happened within six months of Otto's rejection of the defendant's garments. Firstly, there is no reason why in a rejection of such a large quantity of garments worth a million dollars, Wong would not have raised the matter with Derek Cheung, irrespective whether there was prior undertaking to compensate for the FOB price. For Derek Cheung to say that Wong never complained until receipt of the letter before action, was just a blatant lie. Secondly, the garments were rejected towards the end of January 1995 when Derek Cheung must be planning the setting up of his brother's factory and planning to take with him the defendant's business from the plaintiff. He had every reason to appear to be accommodating towards Wong so as to secure the defendant's business for his brother in future. Thirdly, he also had every reason to conceal the defendant's complaint from Chan. If Chan were to know about this potential claim, it would adversely affect his brother when finalising their partnership account. He is therefore another witness whose evidence I have to treat with extreme caution. Again, in the final analysis, on the totality of the evidence, and after duly warning myself of the need for caution, I accept his evidence that he had neither requested avoidance cutting nor undertaken to pay the FOB price of any rejected garments.

25. Thus the witnesses whom I have found honest and credible have no personal knowledge of the matters in dispute, while those who have were not telling the whole truth. I therefore have to build a picture of the true facts from the pieces of jigsaw given to me by these witnesses. I shall start with the uncontroverted evidence, test the witnesses' evidence against the uncontroverted evidence and then draw inference as to what was the truth. The case has also to be resolved on the burden of proof.

The missing defective garments:

26. According to the defendant, all 5,200 pieces of garments under the three orders had been manufactured and inspected before the end of the lunar year. 2,030 pieces were accepted. Thus 3,170 pieces were rejected upon inspection while another 117 pieces were returned after shipment. Thus the total number of garments rejected was 3,287. However, only 1,281 pieces were available for inspection by its expert, Wicky Wong. 2,006 pieces of defective garments were therefore missing. According to Ada Lam, the 1,076 pieces shipped on 22 April 1995 were manufactured pursuant to a replacement order and did not constitute a further delivery under the three purchase orders. She said these garments were not salvaged from the 2,006 defective pieces, but were made from new fabric.

27. The defendant could offer no explanation whatever why such a significantly large quantity of garments was missing. This immediately raises a number of questions. How credible is the defendant's counterclaim or how credible are the defence witnesses when their evidence is tested against the missing garments? Was it true that all the 5,200 pieces had been manufactured and inspected by Johnny Ng at the end of the lunar year or was it less? Was it true that 3,287 pieces have been rejected or was it less? Were the 1,076 pieces shipped in April 1995, or at least part of that shipment, salvaged from some of the rejected garments that are now missing? Were those 1,076 pieces part of the 5,200 pieces ordered by Otto? Is the defendant inflating its claim?

28. Otto placed three orders. The first one, Repeat 1 Order was for 4,300 pieces for delivery on 31 January 1995. It was dated 23 November 1994. It bore the impression of a chop dated 28 November 1994, which presumably indicated the date when the order was received or responded to by the defendant. The defendant subsequently sub-contracted the production work to Hua Gan for that quantity via a sub-contract agreement dated 24 December 1994. On that agreement was the impression of a chop, which indicated that the garments had been delivered on 24 January 1995, which was some time before the end of the lunar year. All these tied in very well with the defence case.

29. The second and third orders, Repeat 2 Order and Repeat 3 Order, were for 500 and 400 pieces respectively. These orders were dated 12 and 16 December 1994 respectively. The chops on the orders suggested that these were either received or responded to by the defendant on 19 December 1994 and 4 January 1995 respectively. These garments were scheduled for delivery on 28 February 1995. Wong claimed that they were manufactured at the same time as the 4,300 pieces under Repeat 1 Order. But the production of these 900 pieces was not included in the sub-contract agreement dated 24 December 1994. The defendant did not produce any sub-contract agreement in respect of these 900 pieces of garments or similar documentation showing that these 900 pieces had been delivered by Hua Gan, though similar document for the 4,300 pieces under Repeat 1 Order was available.

30. Wong's evidence that Johnny Ng was able to examine 5,200 pieces of garments within a day is hardly convincing. Allowing for travelling time to and from Dongguan and meal time, 10 hours was probably all that could be spent for inspecting the garments. Thus Johnny Ng only had 7 seconds to inspect each piece of garment. It is impossible for him to perform any responsible or meaningful inspection at that speed. The truth probably was either that only 4,300 pieces were available for inspection or that Johnny Ng only performed a random check. Though Koopmann said that according to her record Otto performed a random check, I do not think that is correct. This is because the shipment invoices show that 2,030 pieces were shipped and hence on the defendant's case, 3,170 pieces had been rejected by Johnny Ng before shipment. He could only have done that if he had performed a piece by piece check and not a random check. If only 4,300 pieces were available for inspection, Johnny Ng would have 9 seconds to inspect each piece, which is less unrealistic.

31. When cross examined about the April delivery of 1,076 pieces, Wong eventually had to admit that he might have been mistaken when he said 3,170 pieces had been rejected by Johnny Ng during that inspection. I entertain serious doubts whether in fact there were 5,200 pieces available for inspection. This is especially so as Johnny Ng, whose identity is known to the defendant, has not been called, and the defendant chose to call Koopmann who does not have first hand information about the inspection.

32. On the other hand, there were two memo issued by Connie Tsang dated 15 and 17 February 1995. By the first memo, she asked Hua Gan for an estimate of supplementary fabric required for the Repeat 2 and Repeat 3 Orders as a result of colour shade variation. In the second memo, Connie Tsang confirmed that the supplementary fabric would be delivered on the following day and requested Hua Gan to return all defective fabric, presumably for the purpose of claim against the plaintiff.

33. The defendant also relied heavily on these two memo as evidence of the plaintiff's defective dyeing. Connie Tsang, herself, could not recall anything about these two memo. Mr Chiu submitted that these memo are consistent with Wong's evidence that all the 5,200 pieces had been inspected and the new fabric was for the purpose of salvaging the rejected garments. This is inconsistent with Ada Lam's evidence that none or only a few pieces of garments were salvaged. If there were garments salvaged, these salvaged garments have not been accounted for. On the other hand, these memo were dated two weeks before the scheduled date of shipment for the garments under Repeat 2 and Repeat 3 Orders. This is more consistent with the fact that these two orders were still under production as at the date of the memo and Johnny Ng could not have inspected all the 5,200 pieces during the inspection at the end of the lunar year. I consider this the only reasonable inference to be drawn.

34. As for the third shipment of 1,076 pieces on 22 April 1995, Ada Lam said these garments were manufactured pursuant to a replacement order, which had nothing to do with the Repeat 1, Repeat 2 and Repeat 3 Orders. The replacement order from Otto was not produced. No explanation was given to explain the absence of such documentation either from the defendant or from Otto. The defendant's allegation is not supported by a separate work order to its sub-contractor, Hua Gan or to its own subsidiary, Sunrise. This delivery was not referable to any so called replacement order in the shipment invoice, but was only referable to the Repeat 1, Repeat 2 and Repeat 3 Orders.

35. This then brings me back to Connie Tsang's two memo dated 15 and 17 February 1995 regarding delivery of supplementary fabric for the Repeat 2 and Repeat 3 Orders. The memo specifically referred to these two orders and the total quantity of 900 pieces of garments. The only reasonable inferences to be drawn from these two memo and the reference to the three purchase orders on the shipment invoice are that (i) the garments under the Repeat 2 and Repeat 3 Orders had not been manufactured at the time of Johnny Ng's inspection in January 1995 and that (ii) the supplementary fabric arranged by Connie Tsang on 17 February 1995 was to manufacture the garments for the Repeat 2 and Repeat 3 Orders which were then shipped on 22 April 1995. This was almost two months behind the scheduled delivery date. That is by no means surprising as the second shipment had also been delayed by a month after the garments were ready while awaiting cargo space. I therefore reject Ada Lam's evidence that the 1,076 pieces shipped on 22 April 1995 were related to a replacement order,

36. Ada Lam said these 1,076 pieces were produced by using new fabric ordered from other dyeing factories because Connie Tsang considered the workmanship of the plaintiff not up to standard. She produced an invoice from Dickful Investment Ltd (hereinafter called "Dickful") dated 31 March 1995 addressed to Connie Tsang and another invoice from Tung Shing Dyeing Factory (hereinafter called "Tung Shing") dated 6 April 1995 in support of her evidence. Connie Tsang could not remember why she ordered from these two factories and what the fabric was for. This is understandable in view of the lapse of time. She also said that the defendant had a good stock of similarly dyed fabric for making other styles of garments. She could not be sure whether the defendant used the fabric ordered from Dickful and Tung Shing for making the garments for Otto or that it used some other fabric in its warehouse earmarked for some other styles to fulfil these orders.

37. According to Connie Tsang's memo, the fabric supplied on 18 February was expressly for the Repeat 2 and 3 Orders. This fabric must have been used for the production of those 1,076 pieces, which was the only shipment after the two lots totalling 2,030 pieces. There is no documentation between 17 February and April to show that additional fabric was required for the so called replacement order. It is therefore incredible that Connie Tsang would have ordered fabric from Dickful and Tung Shing in April for making those 1,076 pieces. It then follows that the second part of Ada Lam's assertion that Connie Tsang was not satisfied with the workmanship of the plaintiff was probably a gratuitous creation on her part. I also reject her evidence that Connie Tsang ordered from other factories because she was not satisfied with the plaintiff's workmanship. There could be other reasons for so ordering. In view of the lapse of time, it is understandable that Connie Tsang could not remember why she ordered fabric from Dickful and Tung Shing.

38. If those 1,076 pieces were part of the delivery under the Repeat 1, 2 and 3 Orders, the defendant should have excluded this quantity from its counterclaim. If they were manufactured as part of the defendant's effort in mitigating its loss, it should have honestly said so and claimed production cost as part of its costs of mitigation rather than making an empty assertion of a separate replacement order thereby inflating its claim.

39. The defendant's assertion about this replacement order was unsupported by any documentary evidence, whether from Otto, or Hua Gan or from itself. It was never able to explain why 2,006 pieces of rejected garments were missing and why none of the dismantled garments or defective pieces are kept for the purpose of claim, despite Connie Tsang had specifically instructed in her memo that the defective fabric should be retained. However, it is remarkable to note that the total quantity of garments shipped in January, February and April less the 117 pieces returned plus the stock of defective garments now in the defendant's warehouse was just 30 pieces short of 4,300 pieces required for meeting the Repeat 1 Order (2,030 + 1,076 - 117 + 1,281 = 4,270). Having rejected Ada Lam's evidence that these 1,076 pieces were produced under a separate replacement order from Otto, I find these pieces were part of the garments produced under the Repeat 1, Repeat 2 and Repeat 3 Orders and that only 4,300 pieces had been produced by the end of January 1995 for inspection by Johnny Ng. This finding adversely affects my assessment of Wong's and Ada Lam's credibility. Wong was dishonestly inflating his claim, even if he had a valid one. And Ada Lam was dishonestly corroborating in that evidence.

40. In view of the large quantity of unexplained missing rejected garments, it is likely that some of those rejected garments had been salvaged and formed part of the shipment of 1,076 pieces in April. The quantity of rejected garments missing is probably between 30 and 930 pieces and not 2,006 pieces as it would be according to the defendant's case. Whether this is the case is unnecessary for my decision. I am not suggesting that these 1,076 pieces were all salvaged items as I accept that the 900 pieces under the Repeat 2 and 3 Orders were being produced after the Chinese New Year with the supplementary fabric arranged by Connie Tsang. Suffice it is to say that the above analysis has shown Wong and Ada Lam were exaggerating and were dishonestly inflating the defendant's claim.

Colour shade variation in the finished garments:

41. According to Chan, it is impossible to dye to an identical colour. Colour shade variation invariably occurs in dyed fabric between different rolls and sometimes even within the same roll because the fabric itself may not be consistent. Thus a narrow range of colour shade variation is accepted. The variation within the acceptable range is identified by "lot colours". Fabric dyed from different tanks exhibit different lot colours. Hence, even the tank in which the fabric has been dyed is identified on each roll of dyed fabric by a sticker for the benefit of the garment workers. Shipment samples of the different rolls of fabric dyed from the first tank would be sent to the defendant for approval and determination of permitted lot colours. The variation in colour shading was also acknowledged by Wong as he recognised the importance of sewing only pieces from the same roll of fabric into the same piece of garment.

42. According to the plaintiff, the usual practice was that when the fabric was delivered, it was checked by the defendant's workers. Any fabric found defective would be returned for re-dyeing. If that could not cure the defect, the plaintiff would perform the dyeing using new fabric and compensate the defendant with the cost of the fabric wasted. It would then sell the defective fabric, with or without further processing, to make good part of its loss. Where the defect affected only a small area of the fabric, say less than 5%, and was well confined, the defective part of the roll would be cut and discarded. If the defect were spread throughout the roll, the fabric would be cut into pieces sized for sewing. The pieces with defect would then be removed. In either event, the plaintiff would compensate the defendant for the cost of the defective fabric so cut away. This is called "avoidance cutting". This would only be employed where the defective area is small, so that it is undesirable to have the entire roll re-dyed. This is just practical sense.

43. The plaintiff does not dispute that there were colour shade variation among the sewing panels in some of the garments, but denies that it was due to defective dyeing. Though not specifically pleaded, it was submitted by Mr Mak on behalf of the plaintiff that the colour shade variation could be due to workers sewing panels with different lot colours into the same piece of garment. According to the defendant's own expert, Wicky Wong, this is a possibility. Thus colour shade variation in the finished garment is not conclusive evidence that the fabric had been defectively dyed. The issue remains whether the fabric was dyed in accordance with the sample provided by Connie Tsang or within the lot colours approved by her.

44. The defendant relied on the rejection by Otto, Wicky Wong's opinion, Connie Tsang's report of defective fabric and Derek Cheung's undertaking to pay the FOB price of rejected garments as evidence of defective dyeing and sandwashing of the fabric by the plaintiff.

Otto's opinion as to cause of colour shade variation in the garments:

45. At the request of the defendant, Otto's Product Manager, Ms Koopmann wrote a letter dated 22 January 1996 as follows:

"Regarding the above mention orders which had delivery on Jan - Feb, 95.

At that time, during the final inspection, it was found that the blazer had a serious colour shading, and it was caused by the uneven dyeing stuff."

Koopmann was not the person who inspected the garments and was certainly not qualified to give expert evidence that the colour shade variation was due to uneven dyeing stuff. While I accept that a substantial quantity of garments had been rejected because of colour shade variation, for reasons as explained above, I cannot conclude on the basis of Koopmann's evidence that the colour shade variation was caused by uneven dyeing stuff or was due to the plaintiff's defective workmanship.

46. Further, in the course of her evidence, it became clear that the garments were made of two types of fabric. The body of the garment was made of dyed fabric processed by the plaintiff, while the sleeves were made of white fabric with blue stripes printed by another factory, For Ngai. Both Otto and the defendant accepted that because of the different colouring process involved there were bound to be colour shade variation between the printed stripes and the dyed body. Koopmann said Otto had allowed tolerance for such colour shade variation. But what that tolerance was is unknown.

47. According to Wicky Wong, 80% of the 125 pieces of garments he examined showed colour shade variation between the body and the printed sleeves, though only about one third of the sample showed colour shade variation within the body of the garments. This suggests that the rejection by Otto was due more to colour shade variation between the printed and the dyed fabric than to the colour shade variation within the dyed fabric.

48. Mr Chiu submitted that the plaintiff contracted to dye in accordance with the colour sample of the printed sleeve and as Wicky Wong has observed 80% of the garments showed colour shade variation between the sleeve and the body, this must be because the fabric was not dyed to conform to the sample. With respect, he overlooked the very important and undisputed fact that because of the different colouring process, there were bound to be colour shade variation between the printed fabric and dyed fabric. The extent of that tolerance was unknown both to this Court and to Wicky Wong and Wicky Wong had not been asked to give an opinion on the extent of the colour shade variation between the dyed and the printed fabric. Besides, the colour shade variation could be due to the colour of the printed fabric processed by For Ngai not being consistent or that the colour of the fabric dyed by the plaintiff was inconsistent with the colour sample or the approved lot colours. I therefore reject Koopmann's evidence that the colour shade variation was due to uneven dyeing stuff and I ignore any colour shade variation between the colour on the sleeve and the body.

Wicky Wong's expert opinion as to the cause of colour shade variation in the garments:

49. Wicky Wong studied a two year course in garment manufacturing in Hong Kong Polytechnics, worked for two years in a garment factory and then 13 years with SGS Hong Kong Ltd. He examined the rejected garments with his unaided eye and applied his subjective standard. His finding was therefore not supported by any scientific analysis or data, but purely subjective. However, his finding was not challenged by the plaintiff. He examined 125 pieces out of the 1,281 pieces of rejected garments. He was only given a piece of garment as the control sample for his comparison. He had not been given the colour sample attached to the defendant's purchase order to the plaintiff or the lot colours approved by Connie Tsang for comparison. Hence, he was not in any position to say that the fabric had not been dyed in accordance with the sample or within the approved lot colours. That is the greatest limitation of his evidence.

50. Wicky Wong found colour shade variation in 49 out of the sample of 125 pieces of garments he examined. His analysis is as follows:

Colour shade variation	Quantity
Front/back panel	21
Back/back panel	17
Pocket flap/pocket	2
Pocket/body	4
Lapel/body	2
End to end on front	1
Uneven colour streak	1
Fold mark	1
Total:	49

He regarded 7 rejects out of a sample of 125 as acceptable according to ANSAI/ASQC Z1.4-1993 standard, whatever that standard means. I do not consider that standard to be relevant for the present proceedings.

51. Of the 49 pieces, he regarded one as defective because of fold mark and another because of uneven colour streaks, which according to him could be due to the fabric itself or to defective dyeing process. I do not think these defects should be attributed to the plaintiff's workmanship.

52. Accepting that 47 pieces out of the sample of 125 had colour shade variation among the dyed fabric, there would be 482 such pieces (47/125 X 1,281) among the 1,281 pieces of rejected garments now available or among the entire lot of 4,300 pieces manufactured. This is equivalent to 11.20% of the garments manufactured.

53. Wicky Wong said that apart from defective dyeing, the colour shade variation in the finished garments could be due to the length of time of storage, the colour of the interlining and workers sewing panels of different lot colours or from different rolls into the same piece of garment. The first of these possible causes need not concern us because there had not been undue storage before the fabric was made into garments.

54. As for the second cause, Wicky Wong said that the interlining was affixed to the underside of the fabric by a heating process. Apart from the possible changes in colour as a result of the heating process, the presence of the interlining and its colour affected the appearance of the fabric from the surface. I have examined the three pieces of garments exhibited in court. On a visual examination, I find there is no interlining on the underside of the back panels and pockets of all these three pieces of garments. Of the 47 pieces of garments which showed colour shade variation, 21 pieces had colour shade variation between the front and back panels, 4 pieces had colour shade variation between the body and pocket and 2 pieces had colour shade variation between the pocket and the pocket flap. There is no evidence whether these 27 pieces were rejected solely because of the colour shade variation as a result of the lack of interlining. If 21 pieces had colour shade variation between the back panel and the body while 104 pieces had not, if only 4 pieces had colour shade variation between the pocket and the body while 121 had not, and if only 2 pieces had colour shade variation between the pocket and the pocket flap while the other 123 had not, the inference is that the absence of interlining is not the contributory cause for the colour shade variation.

55. The real cause of concern is the third cause, i.e. sewing panels of different lot colours or from different rolls into the same piece of garment. Mr Chiu submitted that the defendant was not a novice to the trade and would not have ignored such a basic requirement. I have no doubt that the defendant and its director, Wong, were not novices to the trade and would have appreciated the importance of sewing only panels from the same roll into the same piece of garment. However, this does not mean the defendant's sub-contractor, Hua Gan and its workers down the line would appreciate that importance. Even if they did, it does not mean that they had observed that requirement. It would have been extremely easy for Hua Gan's workers to have neglected to cut all the component sewing pieces in appropriate numbers from the same roll of fabric or to have neglected segregating the cut pieces from the different rolls. There is no evidence as to the competence of Hua Gan and its standing in the trade or evidence of the procedures adopted by Hua Gan. Quite on the contrary, as the pocket flaps and the lapels show, Hua Gan's workmanship was rather inconsistent. On my examination, the top panels of some of the pocket flaps and lapels have interlining and the under panels had not, while in other cases, the under panels are interlined while the top panels are not.

56. Mr Chiu submitted the possibility that Hua Gan's workers sewed panels of different lot colours or from different rolls into the same piece of garment as speculation. With respect, I do not agree. The burden of proving defective dyeing rests on the defendant. Colour shade variation in the finished garment is not conclusive of colour shade variation in the fabric. There is no direct evidence of the plaintiff's defective workmanship. But on the contrary, there is evidence from the defendant's own expert that the colour shade variation in the finished garment may be due to workers piecing together in the same piece of garment panels of different lot colours or from different rolls of fabric. In view of the many possible causes for colour shade variation in the finished garments, and bearing in mind the limitation of Wicky Wong's evidence, I cannot be satisfied from his evidence that there were colour shade variation within the individual rolls of fabric dyed by the plaintiff.

Defendant's lack of documentation:

57. There is a total lack of documentation as to the quantity of defective fabric allegedly returned for re-dyeing and as to Derek Cheung's request and undertaking. Wong's evidence is that when fabric was delivered to the defendant's warehouse in China, the deliveries would be recorded. There were records showing return of other fabric to the plaintiff for remedial works. However, the defendant was unable to produce any record showing return of any of the fabric in question for re-dyeing. Ada Lam suggested that the fabric were rejected right at the time of their delivery. This is hardly convincing. The defect alleged was uneven colour shading. This could not have been detected unless the fabric was unrolled and examined. There were more than 30,000 yards of fabric dyed. The fabric was delivered in batches of 50 to 100 rolls of about 40 yards each. It is difficult to believe that the fabric could all have been immediately examined upon delivery. The likelihood is that the fabric was only examined before cutting. If so, any defective rolls returned would have been warehoused upon delivery and documented upon return for re-dyeing. But there is a total absence of such documentation. There was documentation for other minor complaints such as shortage of the fabric in the dyed rolls. The absence of documentation for such a serious complaint involving such an extensive quantity of fabric as alleged does cast doubts on whether any such complaint had been made.

58. Similarly lacking is contemporaneous documentation of the alleged request and undertaking. On many previous occasions when loss resulted from the plaintiff's defective workmanship, the defendant would lodge a written complaint and the plaintiff would be required to acknowledge liability in writing. For example, in her memo to the plaintiff dated 27 February 1995 addressed to the attention of Derek Cheung, Connie Chan specified the defect that required supplementary materials of 700 yards and she expressly held the plaintiff responsible for the materials so wasted. This memo related to another order but was issued a month after the rejection of the defendant's garments by Otto. There are other similar memo holding the plaintiff responsible for other damage in respect of other orders. There was at lease one occasion in which Derek Cheung himself acknowledged such liability on a memo issued by the accounts department of the defendant. But there is no record of similar documentation in respect of the alleged request and undertaking by Derek Cheung. Ada Lam's explanation was that Connie Tsang had taken away her document files when she left the defendant's employ. When Connie Tsang left, the garment had been rejected. The defendant should be aware of this million dollars worth of claim against the plaintiff. It is hardly convincing that under such circumstances, the defendant would have let Connie Tsang leave without securing from her the documentation about these three orders.

59. It was also the parties' practice that after liability has been agreed, the damaged garments would be counted by the plaintiff's representative and the defendant would issue a demand note for damages. In the present case, no such demand note had been issued. Ada Lam's explanation was that as Derek Cheung never attended the defendant's factory to count the rejected garments and hence no demand note could be issued. She also put the blame on Connie Tsang, alleging that Connie Tsang had told her prior to her resignation that the matter had been settled. She was Connie Tsang's superior. This was a large scale rejection of a very large quantity of garments by a very important and regular client. The FOB value of the rejected garment was almost $1 million. On her evidence, Derek Cheung visited the defendant's factory frequently to solicit orders for his brother and said that he would follow up with these three particular orders despite he had left the plaintiff's employ. It is hardly credible that as Connie Tsang's superior Ada Lam would have simply assumed that the matter had been settled without pressing for an acknowledgement of liability or insisting on Derek Cheung to count the rejected garments or referring the matter to Wong or some other colleagues for follow up action.

60. The total absence of documentation suggests the likelihood is that there was never any complaint made about defective fabric before the fabric was cut and turned into garments. Accordingly, it is more likely than not that there was never any request for avoidance cutting or undertaking by Derek Cheung to pay the FOB price of rejected garments.

Defendant's response when pressed for payment:

61. Derek Cheung left the employment of the plaintiff in July 1995. He had never informed Chan that there was problem relating to these three orders or of any undertaking to pay the FOB price of the garment rejected. After Derek Cheung left, Chan followed up on the orders and pressed for payment a number of times. He was not given any reason why payment was withheld. Eventually in reply to his fax reminder for payment, Ada Lam and Wong's secretary replied that the delay in payment was due to changes in personnel in the plaintiff's and defendant's company resulting in ambiguity in certain accounting matters and that the defendant would advise him of the date of payment on 10 November 1995. Then on the following day, i.e. 11 November 1995, they replied saying that Wong was not in Hong Kong.

62. The payment had been outstanding for 8 to 9 months by then. By that time, the rejection by Otto of garments worth almost $1 million must be well known to Ada Lam and Wong's secretary. Ada Lam in particular, on her own evidence, was aware of Derek Cheung's undertaking to pay the FOB price of the rejected garments. If there was any truth in that allegation, she should have so replied to Chan, instead of talking about ambiguities in accounting as a result of changes in personnel. This is another pointer that the alleged request and undertaking by Derek Cheung is a fiction that has not taken shape by November 1995.

The negotiation meeting in December 1995:

63. At Wong's request, Derek Cheung arranged a meeting between Chan and Wong in December 1995 after Wong received the letter before action from the plaintiff's solicitors. Chan and Derek Cheung had been extensively cross examined about this meeting. According to Chan, Wong never mentioned anything about Derek Cheung's undertaking or any particular problem about the dyeing. He said that Wong only asked for compensation for goods damaged. When Chan asked him what was damaged, Wong did not answer but said that it did not matter and litigation would take a long time. Chan said when asked why Wong did not raise the complaint earlier, Wong replied that in the past the plaintiff paid compensation for damaged goods. Derek Cheung did not take part in the conversation and remained silent.

64. Mr Chiu submitted that it was strange that Derek Cheung would have remained silent during the negotiation meeting and that if he had not undertaken to pay the FOB price of rejected garments he ought to have voiced that out at the meeting. I have no doubt about Chan's credibility. I am satisfied that during that meeting Wong had not raised the issue of Derek Cheung's request and undertaking. The entire basis of Wong's refusal to pay, as disclosed during that meeting, was that in the past the plaintiff paid compensation for goods damaged by their defective workmanship. If Wong had not alleged Derek Cheung of having requested avoidance cutting, there is of course no reason why Derek Cheung would have brought up the topic that he had never requested avoidance cutting or offered an undertaking. This topic simply could not have crossed his mind then. On the other hand, it is understandable that as Derek Cheung was at that stage soliciting business from Wong for his brother's new factory, he had every incentive not to say anything at the meeting that might offend Wong. For him, in the circumstances, silence was gold.

65. Since Wong never raised the issue of Derek Cheung's request and undertaking at the meeting, the inference is that there was never any request or undertaking by Derek Cheung. All along, Wong was counting on the plaintiff's goodwill in the past in making compensation. The allegation about the request and undertaking is an afterthought, which had not even taken shape at that meeting. This also provides support for Connie Tsang's and Derek Cheung's evidence that Derek Cheung had not requested avoidance cutting or offered any undertaking to pay the FOB price of any garments rejected.

Alleged report by Connie Tsang of defective dyeing:

66. The defendant's case rests heavily on Wong's and Ada Lam's allegation that Connie Tsang had reported to them that the fabric dyed by the plaintiff was defective and Derek Cheung had requested the defendant to perform avoidance cutting and had undertaken to compensate the defendant the FOB value of any rejected garments. That was flatly denied by Connie Tsang. To a certain extent, Connie Tsang appeared to be evasive in failing to recall some very striking events in relation to this order, which occurred prior to her resignation, e.g. the large scale rejection of garments by Otto.

67. On the other hand, Wong and Ada Lam appeared to be more incredible. Firstly, Wong was obviously lying about having manufactured 5,200 pieces before the end of the lunar year to inflate his claim. Secondly, Ada Lam, in support of Wong's allegation, said that the 1,076 pieces shipped in April 1995 were made pursuant to a separate replacement order and from new fabric ordered from Dickful and Tung Shing because Connie Tsang considered the plaintiff's workmanship not up to standard. For reasons as explained above, I have rejected Ada Lam's evidence as being incredible.

68. On Connie Tsang's evidence, the three dyeing orders placed with the plaintiff were uneventful. She could not recall anything unusual about them. She was shown documents in which she made some mention of defective dyeing by the plaintiff. Defective dyeing and re-dyeing is nothing unusual in the industry. What was unusual was the request for avoidance cutting and the offer of undertaking to pay the FOB price of rejected garments made from the defective fabric. She was very firm that there was never such an undertaking discussed between her and Derek Cheung. Having rejected Wong's and Ada Lam's evidence, and in view of the total lack of documentation from the defendant about return of defective fabric for re-dyeing and about the undertaking, I do not think on this important issue she was discredited by any of the documents she was cross examined on. I therefore accept Connie Tsang's evidence that she had not reported to Wong or Ada Lam about defective dyeing by the plaintiff or Derek Cheung's request and undertaking.

Alleged request by Derek Cheung for avoidance cutting and undertaking to pay the FOB price of rejected garments:

69. The defendant's evidence of Derek Cheung's request and undertaking again came from what Wong and Ada Lam alleged Connie Tsang as having reported. This was also flatly denied by Derek Cheung.

70. According to Derek Cheung, he would not request avoidance cutting if the defective fabric amounted to more than 1 to 2 % of the fabric in a particular roll. Chan suggested a maximum of 5%. The reason is based on risk. If for whatever reason, including one of inadvertence, a defective piece has found its way into a completed piece of garment, the finished garment would be rejected and the plaintiff would have to pay the FOB price or cost of workmanship of the rejected garments, which is usually very high as compared with the cost of compensating for the defective fabric. The defective cut pieces could be of no use to the plaintiff. Furthermore, the plaintiff would have difficulties in selling the rejected garments at a fair price because the plaintiff is not in the garment business and has no access to the market for garments and also because of the limited range of colour and size of the rejected garments. On the other hand, the defective fabric, so long as it has not been cut into sewing pieces, could be re-dyed and sold as entire rolls of fabric. Being in the business of dyeing, the plaintiff has access to the market for such second rate fabric. Thus avoidance cutting is not an option which would be lightly requested by the plaintiff. If the defect is more than 5%, it is probably wiser to have the fabric re-dyed and if that could not cure the defect, to compensate the cost of the fabric, than to run the risk of avoidance cutting. Where the defect in the fabric is more than 5%, performing avoidance cutting is not a good bet. It is just gambling the heavier manufacturing cost for the cheaper cost of the fabric. After all the defective fabric could be dyed to a darker colour and sold at a reasonable price to reduce the loss. It is against this background that Derek Cheung said he had not requested for avoidance cutting or offered an undertaking.

71. According to Wong, in the normal course of production the fabric would be folded and then cut into sewing pieces as a pile. While performing avoidance cutting, a worker when folding the fabric would put a sticker onto the part of the fabric that showed colour shade variation. After the fabric was cut, those pieces with stickers would be discarded and substituted by other pieces cut from the same roll. All the other component panels required to make up the same piece of garment were to be cut from the same roll. If this process had been followed, there would be no colour shade variation in the garments produced. Indeed, as several past incidents show, the defective garments so generated were usually much less than 1% (22 out of 2,500 pieces, 8 out of 4,000 pieces, 7 out of 550 pieces, 803 out of an on-going order of tens of thousands of pieces), say between 0.2 and 1.26% only. In the present case, the defective garments amounted to 11.20%, i.e. between ten to fifty-six times more. This must cast serious doubt on whether avoidance cutting had been performed. If it had not, then the entire basis of the defendant's case of request for avoidance cutting and undertaking by Derek Cheung must fail.

72. In the light of the above and in the light of the total absence of documentation from the defendant about return of defective fabric for re-dyeing, the total absence of documentation of comtemporaneous complaint of defects, the defendant's response to the plaintiff's request for payment and the lack of reference by Wong to the undertaking during the December 1995 meeting, I cannot be satisfied that Derek Cheung had requested avoidance cutting and offered an undertaking to pay the FOB price of any garments rejected. I am able to reach this conclusion without relying on the evidence of either Connie Tsang or Derek Cheung, whose evidence I have treated with extreme caution.

73. The most probable scenario is that Hua Gan turned the fabric into 4,300 pieces of garments in fulfillment of the Repeat 1 Order without making any complaint of defective fabric and without performing avoidance cutting. Hence there was never any request for avoidance cutting and undertaking by Derek Cheung. There might be defect in the dyeing process or there might not. Hua Gan's workers might have ensured that only panels from the same roll of fabric were sewed together into the same piece of garment, or they might not. Whatever the truth, the defendant has failed to discharge the burden of proving defective dyeing on the part of the plaintiff. About 2,270 pieces were rejected by Otto. Some of the rejected garments might have been salvaged easily by replacing the mis-matched panels, such pockets or pocket flaps, and then shipped with the 900 pieces of garments produced under the Repeat 2 and 3 Orders in April 1995. Knowing that Hua Gan had cut the fabric without making any prior complaint to the plaintiff, Wong realised the weakness of his case and he sought to rely on the plaintiff's goodwill in the past for compensation. He negotiated with Derek Cheung. As Derek Cheung was planning to divert the defendant's business to his brother's factory, Derek Cheung engaged delaying tactics towards Wong's request for compensation and withheld that request from Chan. The matter then came to light when Chan took over from Derek Cheung and pressed the defendant for payment. Yet between then and the meeting in December 1995, Wong never raised the alleged request and undertaking by Derek Cheung as there never was any such a request and undertaking. These allegations were just Wong's afterthought.

Whether the fabric had been defectively dyed:

74. While it could not be disputed that some of the garments exhibited colour shade variation, for reasons as explained earlier, this does not necessarily mean that the plaintiff had failed to dye in accordance with the colour sample or within the lot colours approved by Connie Tsang. As Wicky Wong suggested, this could be due to workers sewing panels of different lot colours or from different rolls of fabric into the same piece of garment. The fatal aspect of the defendant's case is that Wicky Wong had not been asked to compare the colour of the finished garments with the colour sample or the lot colour approved by Connie Tsang. There is, therefore, no evidence that the colour of fabric making up the garments was outside the acceptable range of lot colours approved by Connie Tsang. As the burden of proving defective workmanship rests on the defendant in the counterclaim, I must conclude that the defendant has failed to discharge its burden of proof. Thus the defendant also failed in its fall back position.

Conclusion:

75. In conclusion, I am not satisfied that the defendant has proved on a balance of probability that there were defects in the fabric dyed and sandwashed by the plaintiff and that as a result Derek Cheung requested the defendant to perform avoidance cutting and undertook to compensate the defendant the FOB price of any garments rejected. Accordingly, I dismiss the defendant's counterclaim. As the defendant does not dispute the plaintiff's claim and the quantum, I enter judgment in favour of the plaintiff against the defendant in the amount of $746,969.21 as claimed, with interest at judgment rate from 27 January 1996. I also make an order nisi that the defendant shall pay the plaintiff's costs including half a day's counsel fee for preparation of written submission.