Louis Vuitton Malletier v Ng Hoe Seng (formerly trading as EMCASE SG) [2025] SGHC 122

This judgment deals with the assessment of damages for trade mark infringement where the defendant has used a counterfeit mark. I had earlier granted default judgment to the claimant. Throughout these proceedings, the defendant was absent and unrepresented. Ordinarily, the claimant would have to face the arduous task of having to prove its actual losses, a task made all the more difficult because of the dearth of evidence occasioned by the defendant’s absence.

To cater precisely for this situation, Parliament has legislated for the provision of statutory damages as an alternative to the traditional remedies of damages and an account of profits. This negates the need for a claimant to prove its actual losses. How, then, is a court to calculate the quantum of statutory damages to be awarded, bearing in mind the need to vindicate the claimant’s rights while ensuring that the award is proportionate and fair to the defendant? That is the question before me.

Facts

The claimant is a French-incorporated company and owner of, inter alia, the “Louis Vuitton” mark. The defendant is a Singapore citizen who was formerly the sole proprietor of EMCASE SG (“EMCASE”). EMCASE’s principal activity was stated to be the retail sale of clothing for adults.

The claimant is the registered proprietor of the following trade marks in Singapore (the “Registered Marks”):

S/N

Trade Mark

Filing Date

Classes

1

Trade mark no (“TM No”): T1214158Z

April 2012

3, 4, 6, 8, 9, 12, 26, 34

2

TM No: T9100128Z

January 1991

14

3

TM No: T9007042C

October 1990

18

4

TM No: T7773776B

November 1977

18

5

TM No: 40202103540P

December 2020

9

6

TM No: T0514536B

January 2005

18

7

TM No: 40202103531W

December 2020

9

8

TM No: T0512684H

January 2005

18

9

TM No: 40202103537T

December 2020

9

10

TM No: T0514539G

January 2005

18

11

TM No: T1116259A

September 2011

3, 9, 14, 16, 18, 24, 25, 34

12

TM No: 40201916373U

May 2019

18

13

TM No: 40201916232X

May 2019

18

The defendant operated an online store through an Instagram page with the user name “emcase_sg” (the “EMCASE IG Page”). In or around July 2022, it came to the claimant’s attention that the defendant advertised, offered and/or exposed for sale various goods, including mobile phone cases, watch straps, purses, pouches, wallets, passport cases, key cases and card cases, affixed with signs identical to one or more of the Registered Marks (the “Offending Goods”) without its consent. It was also discovered that the defendant represented through the EMCASE IG Page that his products were upcycled goods which were derived from the claimant’s “authentic” products.

About a month later, the claimant instructed its representative to make a trap purchase of 12 of the Offending Goods. The claimant’s representative did so for a total price of $2,100. The claimant then issued a cease-and-desist letter on 21 March 2023 to the defendant, demanding that he cease offering and/or exposing the Offending Goods for sale.

The defendant caused EMCASE to cease registration as of 26 April 2023, and the EMCASE IG Page became inactive since around May 2023. However, the defendant started operating another Instagram page, this time with the username of “emcrafts_sg” (the “EMCRAFTS IG Page”).

On the EMCRAFTS IG Page, the defendant continued to advertise, offer and/or expose for sale the Offending Goods without the claimant’s consent. Likewise, there were posts on the EMCRAFTS IG Page representing that products on offer were repurposed from “authentic” second-hand products from the claimant.

The claimant instructed its representative to make a further trap purchase of one of the Offending Goods. The claimant’s representative did so once again, along with another item of a different brand to avoid arousing suspicion, for a total price of $471. It has also adduced evidence to show that the defendant continued his acts of advertising, offering and/or exposing for sale the Offending Goods up until as recently as May 2024.

For clarity I will, in the rest of this judgment, refer to the trap purchases mentioned above at [6] and [9] collectively as the “Sample Purchases”,

The claimant commenced action against the defendant for trade mark infringement on 15 August 2023. As the defendant did not file a notice of intention to contest or not contest, I granted judgment in default on 30 November 2023 in HC/JUD 486/2023 (“JUD 486”). In JUD 486, I ordered, amongst other things, an inquiry as to damages or, at the claimant’s option, an account of profits or statutory damages.

The claimant has since prayed for an award of statutory damages pursuant to s 31(5)(c) of the Trade Marks Act 1998 (2020 Rev Ed) (the “TMA”) or, alternatively, an award of general compensatory damages.

The defendant was absent and unrepresented throughout the entire course of proceedings.

The claimant’s case

The claimant submits that the defendant has used counterfeit trade marks in relation to the Offending Goods, thus satisfying the requirement for an award of statutory damages. Having applied the factors set out in s 31(6) of the TMA, the claimant says that it should be awarded a substantial quantum of statutory damages. In the absence of an established basis for the quantification of statutory damages in Singapore, the claimant relies on the scale for the quantification of damages for trade mark infringement in the Canadian jurisprudence (the “Canadian Scale”), at least as a starting point.

Given the similarities between the trade mark principles in our jurisprudence and those of the United Kingdom, I directed the claimant to file further submissions on the UK’s jurisprudence. I also directed the claimant to file a second set of supplementary submissions on the jurisprudence in the United States, as the provisions dealing with statutory damages in the TMA were modelled on those in the US legislation.

In the main, the claimant submits that the UK and US jurisprudence does not offer much guidance on the quantification of statutory damages, and that the Canadian Scale remains the best approach to be adopted in the present case. Based on this approach, the claimant submits that the quantum of damages to be awarded would be $4,840,000.00. However, it claims that, as the maximum limit for the present case under the TMA is $2,900,000.00, this should be the quantum awarded to it.

I will elaborate more on my views in relation to the approaches in each of these jurisdictions below. However, I think it would be useful to first provide a brief overview of the statutory damages regime for trade mark infringement in Singapore, including how it found its way onto our shores.

The law on statutory damages for trade mark infringement in Singapore

The genesis of statutory damages

The statutory damages regime for trade mark infringement was introduced in 2004 as part of a larger suite of legislative changes to fulfil Singapore’s obligations under the US-Singapore Free Trade Agreement (the “USSFTA”). As then Minister for Law, Prof S Jayakumar, remarked when introducing the Trade Marks (Amendment) Bill incorporating these suite of changes, the negotiations for the USSFTA gave the Government an opportunity to “study the key features of the US IP framework, and to examine which are the ones relevant to [Singapore]” (see Singapore Parliamentary Debates, Official Report (15 June 2004) vol 78 at cols 108–114 (Prof S Jayakumar, Minister for Law) (the “Parliamentary Debates”)).

One of these key features was the remedy of statutory damages for trade mark infringement that involves the use of a counterfeit trade mark. As the Minister explained, the remedy of statutory damages was meant to complement the existing process of assessing damages, because “in certain cases, it may be difficult to prove actual losses or obtain an account of profits”. In such a situation, the court will “assess the quantum of statutory damages on compensatory principles, taking into account the guidelines in [s 31(6) of the TMA]”.

In other words, statutory damages, whilst being compensatory in nature, would obviate the need for a claimant to prove its actual or foreseeable losses as a result of the infringing activity.

Sections 31(5) and 31(6) of the TMA

In putting pen to paper, Parliament enacted a statutory damages regime in ss 31(5) and 31(6) of the TMA.

Section 31(5)(c) provides that in any action for infringement of a registered trade mark where the infringement involves the use of a counterfeit trade mark in relation to goods or services, the claimant is entitled, at its election, to statutory damages not exceeding $100,000 for each type of goods or service in relation to which the counterfeit trade mark has been used (s 31(5)(c)(i)), and not exceeding in the aggregate $1 million, unless the claimant proves that its actual loss from such infringement exceeds $1 million (s 35(5)(c)(ii)).

Section 31(6) lists the factors which the court awarding statutory damages is to have regard to. These are:

Whether the defendant used counterfeit trade marks in the present case

Before I deal with the quantification of statutory damages, I first address whether the defendant’s infringement involved the use of a counterfeit trade mark such that the remedy of statutory damages would be engaged.

The claimant submits that the defendant committed at least 121 instances of infringement, as summarised in the following table: `

As the claimant has used “advertising”, “offered for sale” and “exposed for sale” interchangeably, I will, for the sake of simplicity and unless the context requires otherwise, use the term “advertising” to describe all of these acts in the remainder of this judgment.

The instances showing sale of the Offending Goods which the claimant has listed include the Sample Purchases as well as Instagram posts by others showing that they have purchased the Offending Goods (see the examples at [124(c)] below). Some examples of the latter are posts by the same account and appear to show the same product. I refer specifically to the screenshots of posts by accounts with the usernames “k*****xjl”, “t*****kin, and “s****73”. Each of these accounts posted two posts showing what appears to be the same product. Hence, there may be double counting of these as separate instances of the sale of the Offending Goods when, in reality, each pair of posts may just show a single sale of the Offending Goods. These are more appropriately characterised as separate instances of Offending Goods being advertised.

Subject to my observation in [27], and having reviewed the evidence tendered by the claimant, I am satisfied that the instances of infringement summarised in the table (at [25]) are made out. The claimant’s representative, Mr Marcus Lim (“Mr Lim”), lists out the instances of infringement in annexes to his affidavit of evidence-in-chief. He refers to numerous screenshots from the EMCASE IG Page and EMCRAFTS IG PAGE showing messages exchanged regarding the Sample Purchases, posts in which the defendant advertised the Offending Goods for sale, as well as posts and “stories” showing the Offending Goods bought by the defendant’s customers.

The question, then, is whether these instances of infringement involved the use of a counterfeit trade mark.

The definition of a counterfeit trade mark can be found in s 3(6) of the TMA, which I reproduce here:

I preface my analysis by emphasising that not every mark that is the subject of infringement proceedings would be a counterfeit trade mark. This much is clear from the definitions of “infringing goods” and “counterfeit goods” in ss 3(2) and 3(5) of the TMA respectively. Section 3(2) defines goods as being “infringing goods” in relation to a registered trade mark “if they or their packaging bears a sign identical with or similar to that mark ...”. Section 3(5) defines goods as being “counterfeit goods” in relation to a registered trade mark if “(a) they are infringing goods in relation to that mark; and (b) the sign borne by them or their packaging is a counterfeit trade mark” [emphasis added].

Clearly, counterfeit goods are more than mere infringing goods, and a counterfeit trade mark is more than a mere infringing mark. Were it otherwise, every mark that is the subject of infringement proceedings would be a counterfeit trade mark, and that is clearly not how the TMA is intended to operate. The TMA gatekeeps the more expansive remedy of statutory damages by imposing a higher bar on the claimant to meet. As such, care must be taken to ensure that an infringing mark meets the necessary requirements for it to be a counterfeit trade mark before the remedy of statutory damages is engaged.

I now turn to assess whether the marks used by the defendant in the present case meet those requirements such that they were counterfeit trade marks.

First, I address the requirement that the mark used by the defendant be identical with or so nearly resembling the claimant’s registered trade mark so as to be calculated to deceive and, in so doing, state my views on how the phrase “calculated to deceive” should be understood and applied. Prof David Llewelyn, in the article “Statutory Damages for Use of a ‘Counterfeit Trade Mark’ and for Copyright Infringement in Singapore: A Radical Remedy in the Law of Intellectual Property or One in Need of a Rethink?” (2016) 28 SAcLJ 61 (“Llewelyn”) at 74, identifies the difficulty raised by the use of the words “calculated to deceive” in s 3(6) of the TMA. He notes that the definition of a “counterfeit trade mark” in s 3(6) introduces notions such as the phrase “calculated to deceive” which no longer have a place in the modern law of trade mark infringement outside the US.

Trade mark infringement under s 27 of the TMA encompasses only the notion of “likely to confuse”. It does not include the concept of “calculated to deceive”.

According to the Collins Dictionary, the definition of “confuse” includes “to bewilder; perplex”, “to mix up (things, ideas, etc); jumble”, “to make unclear” and “to fail to [recognise] the difference between; mistake (one thing) for another”. On the other hand, “deceive” means “to mislead by deliberate misrepresentation or lies” [emphasis added]. The term “deceive” cannot therefore be equivalent to “confuse”. To induce a state of confusion is to sow the seeds of uncertainty in the mind of another. To deceive someone, however, is to plant a specific lie. A confused person may not be able to tell whether two marks are linked, but a deceived person would positively mistake one mark for the other.

I also note that the term “deceive” implies that there must be an element of intention to mislead another. I am of the view, however, that having a subjective intention to mislead on the part of the defendant is not a necessary requirement that has to be made out. This is because when one looks at s 3(6)(a) as a whole, especially the words “as to be” before the phrase “calculated to deceive”, it would appear that the provision imputes an intention to deceive onto the defendant by virtue of the degree of resemblance between the registered trade mark and the sign used by the defendant. In other words, it suffices that the defendant’s sign is identical with or so nearly resembling the claimant’s registered trade mark such that the inexorable inference to be drawn is that the use of the former was intended to mislead members of the public into believing that it was the latter. There is no need to prove, as an additional element, that the defendant had a subjective intention to deceive.

Seen in this light, there can be no doubt in the present case that the trade marks used by the defendant on the Offending Goods are so nearly resembling, if not identical with, the corresponding Registered Marks so as to be calculated to deceive. The defendant has essentially used exact copies of the Registered Marks, such that the only inference one can draw is that he intended to mislead members of the public into thinking that those marks are the Registered Marks.

This is apparent from a table comparing some of the products offered by the claimant and the defendant in the claimant’s written submissions, which I reproduce below:

Secondly, it goes without saying that the marks identical with or nearly resembling the Registered Marks so as to be calculated to deceive were applied to the Offending Goods without the claimant’s consent. Mr Lim has deposed that the Offending Goods were not manufactured by the claimant or under its licence, and he has explained his reasons for arriving at this conclusion.

Thirdly, I am also of the view that the marks were applied to falsely represent the Offending Goods to be the genuine goods of the claimant. The manner of the application of the marks to the goods leads to only one conclusion: that the purpose of applying the marks was to falsely represent the Offending Goods to be the genuine goods of the claimant. The defendant applied marks which were calculated to deceive and there is nothing else, for example, the use of the defendant’s own distinctive trade mark, that can lead the court to conclude that it was not a false representation (see the example discussed in [156]–[158] below).

In this regard, I agree with the claimant’s submission that the defendant’s reference to the Offending Goods as “upcycled” goods does not change the fact that he had used counterfeit trade marks. On one level, such a reference may be construed as a “disclaimer” that the goods are not the genuine goods of the claimant. However, in my opinion, such a disclaimer is an extraneous factor which has no place in the inquiry into whether a counterfeit trade mark has been used.

I find support for this in Staywell Hospitality Group Pty Ltd v Starwood Hotels & Resorts Worldwide, Inc and another and another appeal [2014] 1 SLR 911 (at [90]–[92]), where the Court of Appeal explained that there is no role for extraneous factors, such as an express disclaimer, to feature in the inquiry on whether there is a likelihood of confusion as a result of the use of an identical or similar mark on similar or identical goods. This is for the following reason:

As Prof Ng-Loy Wee Loon explained, Parliament could not have intended to allow matters apart from the two factors, namely (a) the degree of similarity between the marks, and (b) the degree of similarity between the goods or services to which the marks are applied, to feature in the likelihood of confusion inquiry. This is because the provisions on infringement only mention those two factors and provide for a finding of infringement where a likelihood of confusion results from only those two factors.

Although Prof Ng-Loy’s remarks pertain to the likelihood of confusion inquiry in s 27(2) of the TMA, in my view, a similar reasoning can be applied to the inquiry into whether an infringing trade mark is a counterfeit trade mark.

Much like s 27(2) of the TMA, s 3(6)(a) mandates a comparison of the mark used by the defendant and the registered trade mark only, and asks if the former is identical with or so nearly resembling the latter such that the inexorable inference to be drawn is that the use of the former was calculated to deceive. A finding of deception arises only from a comparison between the two marks and an assessment of the degree of resemblance between them. There is no room for extraneous factors to operate.

At a subsequent stage of the inquiry, the focus then shifts to how the mark used by the defendant is applied to goods. The inquiry under s 3(6)(b)(ii) focuses on whether the manner in which the mark was applied to goods was such that the purpose of doing so was to falsely represent those goods to be the genuine goods of the claimant. It is clear from the provision that the inquiry is limited to determining whether a false representation arises from the application of the defendant’s mark.

Accordingly, any “disclaimer” used by the defendant is an irrelevant consideration in so far as assessing whether the defendant’s mark was calculated to deceive or whether there was a false representation that the defendant’s goods were the claimant’s.

Even if the claimant’s material was used in the Offending Goods, it still does not change the fact that the Offending Goods were not the genuine goods of the claimant. The minute the defendant uses the claimant’s material in an unauthorised manner to make the Offending Goods, those goods become the defendant’s own. The reference to the Offending Goods being “upcycled” products cannot operate as a free pass to negate the defendant’s false representation, which is inherent in the very use of the claimant’s marks on goods that are not its genuine goods. If it were otherwise, many trade mark counterfeiters would be able to escape a finding of liability all too easily.

Having found that the defendant had used counterfeit trade marks in relation to the Offending Goods, I now proceed to assess the quantum of statutory damages to be awarded.

The quantification of statutory damages

Difficulties in quantifying statutory damages in the present case

It should be noted that the claimant’s submission for the Canadian Scale to be adopted relates only to factor (b) of s 31(6), which is the loss that the claimant has suffered or is likely to suffer by reason of the infringement. As such, it is not the be-all and end-all of the inquiry if it is to be adopted at all.

As alluded to in the introduction of this judgment, the claimant faces especial difficulties proving its loss when the defendant has been absent throughout the entire course of proceedings. Besides the screenshots which the claimant took showing the defendant advertising the Offending Goods, as well as the Sample Purchases, there is no other evidence to go on to demonstrate the actual scale of the defendant’s infringement.

Moreover, this is compounded by the problem which claimants dealing in luxury goods face. In the ordinary nature of things, trade mark infringement usually results in lost sales and/or profits. This forms the basis for the quantification of losses suffered by the claimant. The case, however, is quite different for vendors who sell luxury goods.

As Roy J very astutely observed in Louis Vuitton Malletier S.A. v Audrey Wang [2019] FC 1389 (“Wang”) (at [118]) in relation to products like those sold by the claimant (which incidentally happened to be the same claimant as in this case), it is less likely that the sale of counterfeits of such products will result in lost sales, because “they are luxury goods for which the significant attraction for counterfeiters and customers is the difference in price between the counterfeits and the actual product”. Hence, “someone who buys a ‘knock off’ would not necessarily have otherwise bought a genuine product” (Wang at [118], citing Oakley, Inc. v Doe (2000) 8 CPR (4th) 506 at [9]).

In other words, counterfeit luxury goods are not necessarily substitutes for actual luxury goods. Hence, it would be difficult for a claimant to say that the sale of these counterfeit goods and/or the profits earned from them would have otherwise been made by the claimant.

Therefore, the focus shifts to quantifying the loss suffered by the claimant in terms of the depreciation of goodwill associated with its business using those marks. This is how the Assistant Registrars had crystallised the claimant’s losses in Converse Inc v Ramesh Ramchandani and another [2014] SGHCR 11 (“Converse”) (at [32]) and Louis Vuitton Malletier v Cuffz (Singapore) Pte Ltd [2015] SGHCR 15 (“Cuffz”) (at [29]–[30]). I agree with them. Such claimants usually enjoy widespread reputation associated with their trade marks. Their goods are also known for their exclusivity. When a defendant sells counterfeit goods that are not necessarily in keeping with the high standards the claimant holds itself to, then the exclusivity of the goods sold under the trade marks is diminished. The same can happen when the defendant deals in them at much cheaper prices.

It is in quantifying the depreciation of goodwill that a claimant faces difficulties. This is because goodwill is intangible and not easily susceptible to being reduced to a precise figure.

I now turn to analyse how each jurisdiction has dealt with the issue of quantifying statutory and/or general compensatory damages for trade mark infringement.

Singapore

Not much ink has been spilt on the issue of quantifying statutory damages in Singapore. There have only been two published decisions relating to the award of statutory damages for trade mark infringement – Converse and Cuffz. Both are decisions of the High Court Registry and, as such, they are not binding on me. However, they are useful for the assessment of statutory damages in this case. I will refer to them when I come to determining the quantum to be awarded (at [151]–[160] below).

I also note that the Assistant Registrar in Converse declined to apply a “tariff” approach (ie, quantifying the damages based on a fixed figure for each of the infringing goods), saying that statutory damages must be awarded in accordance with the express wording of the TMA, which mandates that a court have regard to and apply the factors in s 31(6) (Converse at [27]).

United States

The next natural port-of-call to turn to is the US, given that s 31(5) of the TMA is loosely modelled after 15 USC (US) § 1117(c) of the US Trademark Act of 1946 (commonly known as the “Lanham Act”). There are, however, significant differences in that the Lanham Act (i) provides for a minimum quantum of statutory damages to be awarded per mark for each type of goods, (ii) provides for the maximum quantum of damages awarded per mark for each type of goods to be raised in the event of wilful infringement, and (iii) does not prescribe a limit on the overall quantum of damages that can be awarded.

It is also noteworthy that the Lanham Act does not contain an equivalent provision to s 31(6) of the TMA, hence leaving courts with an even greater bandwidth of discretion in determining the factors to consider when awarding statutory damages.

The factors considered by the courts have been helpfully summarised in the International Trademark Association’s “Trademark Statutory Damages Trends Summary” released in May 2019 (the “INTA Summary”) (at p 3). They are:

As the claimant rightly points out, many of these factors already overlap with the guiding factors prescribed under s 31(6) of the TMA. Indeed, I would venture to say that s 31(6) of the TMA is, in theory, wide enough to encompass all of these factors, given that factor (e) empowers the court to consider “all other relevant matters”.

However, in my view, whether a factor is to be considered “relevant” would hinge on whether it is relevant for furthering the purposes of the TMA.

In this regard, I note that some of the factors considered by the US courts are more in line with a deterrent aim rather than a purely compensatory aim. It is therefore prudent to explore whether s 31(5) of the TMA likewise contains the same deterrent aim on top of being compensatory in nature. This would determine whether the factors taken into account by the US courts in relation to the deterrent aim of the Lanham Act, such as a reduction of damages on account of the defendant’s mitigating conduct, could feature in the assessment of statutory damages under the TMA. If, for instance, an award of statutory damages under the TMA was meant to be purely compensatory in nature, there would hardly be any space for a court to take into account the defendant’s conduct when determining the quantum to be awarded.

Undertaking such an analysis is also relevant because the factors in s 31(6) of the TMA are largely modelled after s 119(4) of the Copyright Act 1987 (Cap 63, 1999 Rev Ed) (the “CA”), which provides for an award of additional damages. These factors, in turn, were taken from s 115(4) of the Australian Copyright Act 1968 (the “Australian Copyright Act”). The claimant, however, submits that the Australian jurisprudence associated with the award of additional damages is of limited assistance as an award of additional damages is punitive in nature.

In support of its submission, the claimant relies on the AR’s comments in Cuffz (at [42]) that

The AR then went on to say (at [44]) that “statutory damages under the [TMA] are awarded on a compensatory basis as an alternative, and not an addition, to damages or an account of profits” [emphasis in original], relying on the wording of the TMA itself as well as the Parliamentary Debates for this conclusion.

To the extent that additional damages under the Australian Copyright Act may be said to be predicated on a purely punitive basis, since they are awarded in addition to general compensatory damages, I agree with the AR’s approach in treating the jurisprudence on these additional damages with circumspection. Nevertheless, I hesitate to conclude that the jurisprudence is of no relevance at all on the footing that additional damages in the Australian Copyright Act are predicated on an “entirely different basis” from statutory damages under the TMA, or from the mere fact that such additional damages are punitive in nature. Such a conclusion would only follow if statutory damages under the TMA were purely compensatory in nature. That, however, is not clear to me at present, and it is a matter which needs to be considered.

The normative basis for statutory damages under s 31(5) of the TMA

As a starting point, it bears noting that the plain wording of the TMA itself indicates that statutory damages under s 31(5) are not meant to be purely compensatory in nature. In particular, factor (d) under s 31(6), namely “the need to deter other similar instances of infringement”, indicates that Parliament had intended for deterrent considerations to feature in determining the quantum of statutory damages to be awarded. In other words, the award should be high enough to deter the defendant and others from committing the same or similar acts of infringement. This, in and of itself, is at odds with a purely compensatory aim.

There is also factor (e), namely “all other relevant considerations”, which contemplates a certain degree of latitude being afforded to the courts when conducting the assessment of statutory damages.

It is worth noting as well that the loss suffered by the claimant, which is encapsulated in factor (b), is only one out of a range of factors provided for in s 31(6) of the TMA. If statutory damages were meant to be purely compensatory in nature, then, strictly speaking, only factor (b) would be relevant to the assessment of damages. After all, the main aim of compensatory damages is to “put the [claimant] in question (so far as is possible) in the same position it would have been if the wrong(s) had not been committed”, and this is usually done through requiring the claimant to “[bear] the burden of specifically proving its loss” (see Kickapoo (Malaysia) Sdn Bhd and another v The Monarch Beverage Co (Europe) Ltd [2010] 1 SLR 1212 (“Kickapoo”) at [55], referring to General Tire & Rubber Co v Firestone Tyre & Rubber Co Ltd [1975] 1 WLR 819 (“General Tire”) at 824).

It would therefore appear that, on a plain reading of ss 31(5) and (6) of the TMA, statutory damages under the TMA are not meant to be purely compensatory in nature.

Next, I turn to consider the relevance, if any, of the Parliamentary Debates, which formed the basis for the AR’s conclusion in Cuffz that statutory damages under the TMA are awarded on a compensatory basis.

Curiously, as mentioned above at [19], Minister Jayakumar, in introducing the Trade Marks (Amendment) Bill, explained that “the court will assess the quantum of statutory damages on compensatory principles, taking into account the guidelines in section 31(6)” [emphasis added]. This was cited by the AR in Cuffz at [13]. It was also relied upon by the AR in Converse (at [11]) to conclude that “statutory damages were intended to be compensatory in nature – to compensate rights owners for losses suffered particularly where such losses are difficult to prove”.

This, however, is odd, given that at least one of the factors in s 31(6) of the TMA (ie, factor (d)), is clearly not compensatory in nature. Indeed, this oddity has been the subject of academic commentary. For example, Prof Llewelyn notes that it is surprising that the AR in Converse drew attention to Minister Jayakumar’s statement in the Parliamentary Debates that statutory damages are based on compensatory principles, and then proceeded to apply the factor in s 31(6)(d), which on no reasonable analysis is compensatory in nature (see Llewelyn at para 34).

It is clear that there is an apparent contradiction between the language used in the statutory provision and the explanation given in the Parliamentary Debates. While s 31(6) contains a factor that makes express reference to deterrence, the extract from the Parliamentary Debates does indeed only mention that the assessment of statutory damages is to be done on compensatory principles. It makes no mention at all of punitive or deterrent principles.

As the Court of Appeal emphasised in Tan Cheng Bock v Attorney-General [2017] 2 SLR 850 (“Tan Cheng Bock”) (at [43]), “in seeking to draw out the legislative purpose behind a provision, primacy should be accorded to the text of the provision and its statutory context over any extraneous material”, as the text which Parliament has chosen to embody is the law enacted by Parliament. The Court also noted (at [48]) that if any extraneous material referred to when determining legislative intent does not confirm the ordinary meaning of the statute ascertained by looking at its text, or even calls that meaning into question, “the court is not permitted to use that extraneous material as a basis for departing from the ordinary meaning”.

In my view, it is clear, without resort to any extraneous material, that s 31(6) of the TMA mandates for deterrence to feature as a consideration in the assessment of statutory damages.

In that regard, I do think that the rubric of deterrence is wide enough for a consideration of the defendant’s conduct to feature in the analysis, to the extent that such conduct would increase or decrease the need for deterrence, as the case may be. For example, an uncooperative and recalcitrant defendant who has wilfully continued his infringing conduct in violation of orders of court would increase the need to send a strong message to that defendant and other would-be infringers that such conduct will not be tolerated. Conversely, a cooperative defendant’s participation in the proceedings would carry some mitigating weight in so far as it reduces the need to deter him from committing other similar instances of infringement.

Accordingly, these considerations in the US jurisprudence would be of relevance to an assessment of statutory damages under the TMA.

I now turn to the Canadian Scale, which the claimant asks me to adopt and apply in the present case.

Canada

The Canadian courts have, since 1997, utilised a scale for the quantification of damages in cases involving the sale of counterfeit goods. It should be noted that this scale is for general compensatory damages, as Canada does not have a statutory damages regime for trade mark infringement.

The scale approach is neatly encapsulated in Burberry Limited v Juvilyn Billones Ward [2023] FC 1257 (“Ward”) at [98]:

In short, the Canadian Scale envisages the award of fixed amounts of damages on a scale for each instance of infringement. There are three fixed amounts on the scale, and they vary according to the nature of the infringer. As explained in Wang (at [124]), “the scale increases with the scope of the operation, from a short-lived operation, to one where fixed premises exist, to those who manufacture or distribute the counterfeit goods to retailers”. It is also pertinent to note that the Canadian Scale has been described as having “achieved the status of precedent” (Wang at [123], citing Nike International Ltd and Nike (Ireland) Ltd v Goldstar Design Ltd et al., T-1951-95 at [11]), and has been applied consistently (albeit with suitable modifications over time) in trade mark counterfeiting cases in Canada.

While the Canadian Scale has the virtue of imbuing the process of assessing statutory damages with a degree of consistency, I have difficulty applying it to the assessment of statutory damages under the TMA.

On a normative level, the Canadian Scale was developed to assist the Canadian courts in determining the quantum of purely compensatory damages to be awarded. This much is clear from Wang (at [122]), where Roy J stated that “damages should not become a penalty for the violation of one’s trade-mark, providing a measure of deterrence”. Hence, the Canadian Scale is somewhat at odds with the deterrent aim of statutory damages under the TMA.

It is also hard to see how the Canadian Scale would square with the way the statutory damages regime under the TMA is structured. I say this with reference to two features of the TMA.

First, s 31(5)(c) of the TMA imposes limits on the amount of statutory damages that can be awarded in any given case. On the other hand, the Canadian Scale exists within a regime which technically does not have limits on the amount of damages that can be awarded.

Secondly, s 31(6) of the TMA envisions a multi-factorial inquiry in the assessment of statutory damages. The Canadian Scale, however, is used as a heuristic to quantify the loss (of goodwill) suffered by a claimant. Such loss is only one out of the five statutorily-enshrined factors that the TMA mandates a court is to have regard to in awarding statutory damages.

The statutorily-enshrined limits and factors constrain the court’s discretion and necessarily shape the contours of the court’s assessment. Their presence means that a court would have to be mindful of the need to ensure that the quantum it awards adequately reflects the considerations that underpin the statutory damages regime while being proportionate to the legislative scheme.

It may, of course, be said that the Canadian Scale can be used to address the compensatory aim of the statutory damages regime, with the quantum of damages raised to account for the other factors. That may well be true in theory but, given the statutory limits in the TMA, it is difficult to see how that would work in practice. The disjunct between the way the Canadian Scale operates in practice and the manner in which the TMA is meant to work is clear in the present instance.

Based on the number of instances of infringement (ie, 121 instances), as well as an adjustment to the Canadian Scale to account for inflation, the claimant submits as follows. The quantum of damages to be awarded under the Canadian Scale would be $4,840,000.00 but as the TMA would impose a limit of $2,900,000.00 in the present case, that should be the amount awarded.

As I will subsequently discuss (at [101]–[114] below), the limit should instead be $900,000.00 here. However, even if one were to accept the claimant’s submission that the statutory limit in this case is $2,900,000.00, that limit would still be exceeded by a very large margin. If the statutory limit is so easily busted, it effectively renders the other factors in the TMA otiose.

Moreover, it is, in my view, obviously untenable for the maximum quantum to be awarded in the present case on account of the Canadian Scale. This is clearly disproportionate to the legislative scheme, as it cannot be seriously contended that the defendant’s infringing acts are reflective of the most egregious acts that could happen and/or that the depreciation in the goodwill associated with the claimant’s marks is the highest that could possibly occur.

Finally, using the Canadian Scale to anchor the amount of statutory damages to be awarded would, in effect, be tantamount to placing undue weight on the loss that the claimant has suffered. This is antithetical to the approach enshrined in s 31(6) of the TMA, which requires that all five factors be considered in the round. The loss suffered by the claimant is only one of these five factors. I see no reason to accord it with the pre-eminence that would effectively come with adopting the Canadian Scale.

Therefore, I decline to adopt the Canadian Scale for the purposes of assessing statutory damages under the TMA. To import the Canadian Scale into the statutory damages regime would be akin to transplanting the engine of a truck into a car. It may well have worked well in its own context, but it would be ill-suited for a completely different body with its own limitations and make-up.

Be that as it may, I do think that the principles behind the gradations in the Canadian Scale offer some guidance that could be usefully applied to the assessment of statutory damages under the TMA. Essentially, the scale is calibrated based on the scope of the defendant’s operations, with the following three categories: (a) where the defendant operates from temporary facilities (flea markets); (b) where the defendant operates from conventional retail premises (stores); and (c) where the defendant is a manufacturer, importer or distributor of counterfeit goods.

While the numbers in the scale are not relevant for present purposes, I am of the view that the three-pronged classification of the defendant’s scope of operations is useful in terms of guiding a court in assessing how flagrant the defendant’s infringement is (ie, factor (a)). A defendant who runs a large-scale manufacturing operation would necessarily have committed more flagrant acts of infringement than one who runs a stall in a night market or festive bazaar.

Assessment of statutory damages in the present case

I now proceed to assess the quantum of statutory damages in the present case. However, before I do so, I address the claimant’s arguments in relation to the statutory limits under s 31(5) of the TMA.

Determination and application of the statutory limits under s 31(5) of the TMA

The statutory limits determining the maximum quantum of statutory damages that can be awarded in an action for infringement are set out in s 31(5)(c) of the TMA:

The claimant submits that the statutory limits in s 31(5)(c) apply on a per mark basis. Since 13 of the Registered Marks were infringed, the theoretical overall limit on statutory damages would be $13 million under s 31(5)(c)(ii), which imposes an aggregate limit of $1 million for the amount of damages awarded in an action in relation to the use of a counterfeit trade mark (unless the claimant is able to prove that its loss exceeds $1 million).

However, this would still be subject to the cap of $100,000 for each type of goods or service in relation to which the counterfeit trade mark has been used under s 31(5)(c)(i) of the TMA, as ss 31(5)(c)(i) and (ii) operate conjunctively.

It is the claimant’s case that the limit imposed by s 31(5)(c)(i) in the present case would be $2,900,000.00, which is derived based on the following types of the Offending Goods and the different counterfeit trade marks applied to them:

I do not agree with the claimant that the limits imposed by s 31(5) operate in relation to each counterfeit trade mark used by the defendant.

The claimant says that its position is supported by the TMA’s framing of the prescribed monetary limits with reference to “a” counterfeit trade mark used. However, I do not think this necessarily means that the statutory limits were intended to be contingent on the number of counterfeit trade marks used.

The claimant also argues that its position is consistent with the US statutory damages regime under the Lanham Act, which specifically states that the statutory limits apply “per counterfeit mark per type of goods or services sold, offered for sale, or distributed” [emphasis added].

For ease of reference, I reproduce the text of s 1117(c) of the Lanham Act as it stood in 2004 (ie, the provision from which s 31(5)(c) of the TMA was derived) here:

In my view, this undermines rather than supports the claimant’s case. It is instructive that the phrase “per counterfeit mark” in the two limbs imposing the statutory limits is absent from the equivalent limbs in s 31(5)(c) of the TMA. Moreover, s 1117(c) of the Lanham Act also starts off with “In a case involving the use of a counterfeit mark ...”, which is similar to how s 31(5)(c) of the TMA begins, namely with “In any action for infringement of a registered trade mark ...” [emphasis added]. As such, I am not persuaded that the provision’s use of the phrase “a registered trade mark” in the singular form alone leads to the conclusion that the statutory limits are meant to apply on a per mark basis. The claimant is asking the phrase “a registered trade mark” to bear a burden that it does not deserve.

Moreover, having an aggregate limit in s 31(5)(c)(ii) of the TMA, a feature absent from the Lanham Act, would make greater sense if it were not interpreted as being meant to apply on a per mark basis. Section 31(5)(c)(ii) itself does not make reference to “a registered mark” or “the counterfeit trade mark” at all, and seems to imply that the aggregate limit of $1 million contained therein simply applies as it is to each case. That is the plain wording of the provision, and the claimant would be hard pressed to show that Parliament intended to enact a shifting aggregate cap depending on the number of trade marks infringed.

I also add that I do not find attractive the claimant’s argument that such an interpretation would have the effect of encouraging defendants involved in trade mark counterfeiting operations to use as many counterfeit trade marks as possible. I do not think that those who run counterfeiting operations would consciously slap on their goods as many counterfeit trade marks as possible just because the statutory limits are not determined by the number of counterfeit trade marks used. Defendants usually seek to imitate the design and style of genuine products, rather than simply apply counterfeit trade marks for the sake of doing so. In any event, the number of counterfeit trade marks used is something that can be taken into account when assessing the flagrancy of the defendant’s infringement (ie, factor (a)).

From a policy perspective, it would also not make sense for the statutory limits to depend on the number of counterfeit trade marks used because that is not determinative of the egregiousness of the defendant’s infringement and/or the loss suffered by the claimant. In my view, the number of counterfeit trade marks per se, whilst a factor to be taken into consideration, cannot be used to say that for each mark used, the statutory limit is $1 million.

Ultimately, common sense must prevail when it comes to the interpretation of a statutory provision. Tellingly, the claimant only took the position that the statutory limits apply on a per mark basis in its third set of submissions (ie, its second supplementary written submissions). In its first set of submissions, the claimant construed the limits in s 31(5)(c) of the TMA as applying in exactly the same way as I have done here. The claimant did not contend that the limits apply on a per mark basis, and I would hazard that this was because such an interpretation is not tenable on a plain reading of s 31(5)(c). If one has to use a fine-tooth comb to plough through a statutory provision in search of an alternative to the obvious interpretation staring one in the face, that alternative is likely not how the provision was meant to be interpreted and applied.

As such, the maximum amount of damages that can be awarded in the present case would be $900,000.00, as the counterfeit trade marks have been applied to nine different types of goods and this is lower than the aggregate limit of $1 million.

I add that, in the present case, it is clear that the Offending Goods can be classified as nine different types of goods. However, if the facts were different, the court may well need to consider in further detail how a defendant’s infringing goods or services should appropriately be broken down into different types for the purposes of arriving at the damages that will be payable and determining if those damages breach the statutory limit in s 31(5)(c)(i) of the TMA.

Consideration of the factors in s 31(6) of the TMA

I now consider each of the factors in s 31(6) of the TMA in turn.

At the outset, I highlight that while I may make certain points more than once under the rubric of the different factors that I have to take into consideration, I will bear this in mind to ensure that there is no double-counting of factors when I arrive at the final quantum to be awarded.

Factor (a): the flagrancy of the infringement of the registered trade mark

I agree with the claimant that the defendant’s infringement is highly flagrant.

The defendant’s infringement was wide in breadth. He used counterfeit trade marks on nine different types of goods.

The marks used by the defendant were also applied to the Offending Goods in a manner that was similar to the way they were applied to the claimant’s goods, as can be seen from the table above (at [39]). This surely strengthened the false representation perpetrated on members of the public by the use of the marks.

However, I should clarify that while the number of counterfeit trade marks used is a factor which I will need to take into consideration, I will not place undue weight on it. This is because, ultimately, the number of counterfeit trade marks used is but a function of how the claimant uses the Registered Marks on its goods. For instance, if the claimant uses three of the Registered Marks on a bag, a defendant who seeks to imitate that bag would use counterfeits of those three Registered Marks as well. The flagrancy of the defendant’s infringement therefore does not necessarily increase in direct proportion to the number of counterfeit trade marks used.

Quite apart from that, I am of the view that weight ought to be placed on the manner in which the defendant promoted the Offending Goods. To be clear, I am not referring to the defendant’s infringing acts of advertising the goods per se, but rather the way the defendant had showcased and presented the Offending Goods to the public.

The defendant had committed his infringing acts through online means, viz, the EMCASE IG Page and EMCRAFTS IG Page. This would have enabled the defendant’s advertisements and posts to reach a larger audience. Indeed, the EMCASE IG Page had around 16,100 followers, and the EMCRAFTS IG Page had around 17,100 followers as at May 2023. It is clear that the defendant has amassed a sizeable following which would have been on the receiving end of the posts and “stories” advertising the Offending Goods for sale. Theoretically, these posts and “stories” would have been visible to the public at large as well.

The manner in which the defendant promoted the Offending Goods was aggravating. I list a few examples of this:

(a) First, when the claimant’s representative made the Sample Purchases, the defendant used the “LV” name to term his products, as if all of the products were the claimant’s. This can be seen from a screenshot of the conversation between the claimant’s representative and defendant, in which the defendant listed out the products to be purchased:

(b) Secondly, the defendant represented the Offending Goods as being “100% authentic”. Although this was clearly not the case, the defendant went to great lengths to drive home the message that his products were “authentic”. This can be seen from a screenshot of a post on the EMCRAFTS IG Page, where the defendant even represents that every material used is sourced from independent secondary market sources with “authenticity guaranteed and verified through a 3rd party authenticator” [emphasis added]:

It was rather disingenuous of the defendant to represent that every material used was verified, since Mr Lim, who actually has the credentials to verify the authenticity of these materials, has deposed that some of the materials used did not meet the claimant’s standards and thus could not have been authentic.

(c) Thirdly, the defendant’s modus operandi when it came to promoting his products was to re-post Instagram posts and/or “stories” by customers who had purchased his products, ostensibly with the aim of thanking them and showing off their rave reviews of the products. In my view, this has the effect of compounding the defendant’s false representations, since it reinforces the message that the Offending Goods were the claimant’s authentic products.

Some posts, for instance, show the customer referring to the “LV DESIGN” and saying that the purchased product is “MADE FROM AUTHENTIC LV LEATHER” or “Authentic lv skin”:

Additionally, other posts show that it was indeed the defendant’s practice to ask those in his circle to promote his products, as he made reference to his circle of “EmCollabFam” who would post (about his products) in some of the “stories” on the EMCRAFTS IG Page:

In other words, the defendant would deploy his coterie of “influencers” to propagate the misrepresentation about his “authentic” products more widely to his followers and the public at large.

The instances I have highlighted go toward demonstrating the defendant’s use of online means to increase the reach of his infringing acts.

Further, I consider it aggravating that, in some instances, the defendant’s representation that the Offending Goods were “upcycled” from the claimant’s genuine products and thus “authentic” was false in and of itself. This is because some of the Offending Goods, which the defendant had represented to be “upcycled”, were not even derived from the claimant’s genuine products. This was a lie upon a lie which compounded the false representation perpetrated upon members of the public by the mere application of the Registered Marks to the Offending Goods.

That being said, the defendant’s infringing acts must be viewed in perspective, in so far as his scope of operations is concerned.

As I mentioned above (at [98]–[99]), the three categories of infringers in the Canadian Scale are useful for guiding a court in assessing the flagrancy of the defendant’s infringement. This would help a court to achieve a result which is proportionate to the legislative scheme where factor (a) is concerned (ie, it would ensure that only the most flagrant acts of infringement warrant an amount closer to the statutory cap and vice versa).

In this regard, the claimant submits that the highest benchmark rate (ie, the rate for a manufacturer, importer or distributor of counterfeit goods) should be applied, given that the defendant was the manufacturer of the Offending Goods. In the alternative, the claimant says that the rate for conventional retail premises should be adopted in line with the court’s characterisation in Ward.

I note that Ward likewise involved sole proprietors who offered for sale and sold the counterfeit merchandise through online means. Walker J described them (at [100]) as having “an online business of significant reach involving continuing sales and an established, well-stocked showroom”. In the end, however, she applied the established rate for retail sellers, as she found (at [105]) that the defendants’ operations were “at least equivalent to a traditional physical establishment”.

In my view, care must be taken to ensure that the categories of operations are not construed and applied in a literal and mechanistic fashion, without regard to the principles underlying the Canadian Scale. The categories in the Canadian Scale are meant to reflect the scope of the defendant’s operations, not merely the nature of activities carried out by him. As such, the activities of the defendant cannot be viewed in isolation. In this regard, the following remarks in Wang (at [168]) are particularly apposite:

I am not satisfied that the defendant’s operations in the present case are on such a large scale that he should be considered a “manufacturer, importer or distributor” of counterfeit goods. He may have been a “manufacturer” in the literal sense, but he was not the sort of manufacturer who made the Offending Goods for the purposes of distribution to other retailers. He had only manufactured the Offending Goods for himself to sell through his online store. Accordingly, I find that the most appropriate category in which to place the defendant’s scope of operations is that of a retailer (ie, one who operates “conventional retail premises”).

This should not be taken to diminish the seriousness of the defendant’s infringement through online means, particularly when counterfeiting activities online are capable of reaching a wide audience. Indeed, it is on account of this that the US courts have tended to award a higher quantum of statutory damages when the defendant’s infringing activities were carried out online (see, eg, the remarks of the court in Luxottica USA LLC v The Partnerships and Unincorporated Associations identified on Schedule “A” (2015) (Case No. 14 c 9061), that “[i]n cases involving the online sale of counterfeit goods, courts have found substantial damages awards appropriate”; see also the INTA Summary at pp 4–5).

The defendant’s acts of infringement would also be on the more flagrant end of the spectrum as far as retailers are concerned. This is because the defendant was the manufacturer of the Offending Goods, at least on his own representation on the EMCASE IG Page that the Offending Goods were “handmade” and “repurposed” from the claimant’s products, as well as posts on the EMCASE IG Page stating “We are working around the clock to manufacture your orders as fast as we can” [emphasis added]. Hence, while the defendant did not rise to the level of a large-scale manufacturer and distributor, his infringement was certainly more flagrant than a retailer who merely sold products manufactured by someone else.

Factor (b): any loss that the claimant has suffered or is likely to suffer by reason of the infringement

Broadly speaking, the claimant’s submissions on the loss it has suffered or is likely to suffer are two-fold. First, the defendant’s infringement damages the exclusivity of the Registered Marks as consumers are now able to purchase the same type of products which are in a similar style to the claimant’s own goods at significantly cheaper prices (ie, damage to exclusivity). Secondly, the availability of the Offending Goods could also result in consumers choosing to buy the Offending Goods instead of the claimant’s genuine goods since they may believe that they are offering a broadly equivalent product at significantly cheaper prices (ie, lost sales). In this regard, the claimant has set out a comparison of the price ranges between its own products and the Offending Goods:

I agree with the claimant that it is likely to suffer damage to exclusivity. The claimant strictly controls the distribution of its luxury products so as to maintain the exclusivity of the Registered Marks. In fact, so exclusive is its brand that one can only buy the claimant’s products at its own stores, and not any other store, as the claimant does not grant licences to third parties to distribute its products. The claimant is therefore bound to suffer some damage to its exclusivity, reputation and goodwill.

However, I have my doubts as to whether the claimant will suffer lost sales in any significant way. As I have explained above (at [54]–[56]), knock-offs of luxury goods are usually not substitutable with the genuine goods. Some consumers may, in this case, choose to buy the Offending Goods instead of the claimant’s products because of the defendant’s false representations (see [41]–[49] above). Nevertheless, most consumers who are inclined to buy the claimant’s products would likely know that the Offending Goods are not genuine. If one looks at the prices of the parties’ phone bags (as seen in the table at [134] above), there is a huge gulf between the prices of the claimant’s phone bags (at $2,310 to $2,710) and the price of the defendant’s equivalent product (at $295). Some of the claimant’s phone bags are almost eight times the price of the defendant’s equivalent product. It would be obvious to a discerning customer that the defendant’s phone bag is anything but genuine. Indeed, the claimant itself acknowledges, in its supplementary written submissions, that “it is not likely that ... the Defendant’s infringement would have the effect of diverting to the Defendant sales that would otherwise have been made by the Claimant.”

Even for the damage to exclusivity argument, I do not think that the defendant’s infringing acts are likely to lead to the level of depreciation in goodwill and/or exclusivity attached to the Registered Marks that the claimant alleges. One must bear in mind that the defendant is a sole proprietor operating through a social media channel. He is not a large-scale manufacturer who has distributed the Offending Goods to other retailers and sparked other chains of infringement. As such, the losses likely to be suffered by the claimant need to be put into perspective. It would be bizarre if one were to take the view that a sole proprietor selling goods online could wreak so much havoc on the claimant’s reputation and the goodwill associated with its business under the Registered Marks.

Factor (c): any benefit shown to have accrued to the defendant by reason of the infringement

I accept the claimant’s argument that the defendant was able to leverage the Registered Marks to sell the Offending Goods at far higher prices. This is set out in a comparison between the prices at which the Offending Goods were sold and the prices for the same types of products without trade marks associated with luxury goods, such as the Registered Marks, sold by other online sellers on e-commerce platforms including Lazada and Shopee:

On a more intangible level, the defendant would have been able to free-ride off the quality associated with the Registered Marks to enable the Offending Goods to gain traction. As the screenshots (at [124] above) show, the defendant’s use of the Registered Marks allowed him to label his products “LV” products. Consequently, this was how his customers came to identify the Offending Goods as well. They were not termed “Emcrafts” or “Emcase” products. Clearly, the defendant benefited from being able to make use of the Registered Marks.

Factor (d): The need to deter other similar instances of infringement

The claimant adduced evidence showing that the “Louis Vuitton brand” is one of the most counterfeited “brands” in the world. This is something I will take into consideration. Trade in counterfeit goods cuts across borders and it will be useful for the court to be aware of the level of counterfeiting activity in other countries and globally. However, I bear in mind that the focus of the inquiry is on the defendant’s potential future infringement as well as other instances of infringement in Singapore. This flows from the trite principle that intellectual property rights are territorial in nature.

In view of this, the need to deter similar instances of infringement by the defendant and others is acute.

The claimant also highlights the defendant’s evasive and uncooperative conduct in its submissions as a factor to be taken into account under “all other relevant matters” (ie, factor (e)). I am of the view that this is clearly borne out in the present case, and is a matter that ought to be taken into consideration under the rubric of specific deterrence instead.

In JUD 486, I granted the claimant an injunction to restrain the defendant from continuing his infringing acts. Even after this was brought to the defendant’s attention through his daughter, the defendant continued his infringing acts. There were still posts on the EMCRAFTS IG Page advertising the Offending Goods for sale. This was a blatant violation of a court order.

In fact, not only did the defendant not obey a court order, he also changed the EMCRAFTS IG Page to a private account. This meant that those who did not follow the account before it was changed to a private account were no longer able to view the posts by that account. Conversely, those who were following the account before it was changed to a private account were still able to view the posts on the account after the change. This came four days after the claimant’s solicitors sent him a letter reminding him to comply with the injunction in JUD 486, and it was clearly an attempt by the defendant to mask his infringing activities (and potentially continue the infringement).

The defendant also did not participate in any part of these proceedings, thus depriving the claimant of an opportunity to discover the full extent of his infringement to prosecute its claim and quantify its losses.

In the premises, a strong message needs to be sent to the defendant that he may be able to run from the claimant, but he will not be able to hide from the long arms of the law. The defendant has shown himself to be a recalcitrant infringer, and he will need to face the consequences accordingly.

The final amount to be awarded

While there has hitherto not been a decision of the High Court on the assessment of statutory damages under the TMA, it would be useful to look to the two High Court Registry decisions which have dealt with the issue, as well as other cases involving the use of counterfeit trade marks. The amounts awarded in those cases form useful guides for present purposes.

Crucially, the AR had found the defendants’ infringement to be flagrant because they were not merely middlemen, but were heavily involved in arranging for the goods to be sold (at [37]). In other words, the scope of the defendants’ operations was larger than that of the defendant in the present case, as they did not just sell the infringing goods from retail premises, but were involved in their distribution to other retailers. The AR also found that there was a need to deter other similar instances of infringement because, inter alia, there had been more than one attempt by the defendants to infringe the claimant’s rights (at [44]). In fact, they were experienced dealers in counterfeit goods and had been doing so on an organised and regular basis for some time (see [45(d)]). In the round, the AR awarded statutory damages of $100,000, which was also the maximum amount that the plaintiff could have recovered in that case (since there was only one type of infringing goods being sold) (see [46]).

In Cuffz, the claimant was the same claimant as in the present case. However, the registered mark found to be infringed was a different one and it was used on leather wallets. A private investigator hired by the claimant had made a “test purchase”, from the defendant’s outlet, of a wallet bearing a mark identical or similar to the claimant’s registered mark. Subsequently, three other wallets bearing the allegedly counterfeit mark were seized during a raid. These three wallets were the infringing goods which were the subject of the action.

While the AR noted that the flagrancy of the infringement would increase with the quantity of infringing items (at [19]), he declined to draw an inference that the defendant must have sold much more than the three wallets seized (at [21]). He recognised that the defendant’s refusal to participate in the proceedings was the reason why the claimant could not obtain evidence on the actual number of infringing goods held in stock and sold, but decided that the defendant’s uncooperative conduct could instead be taken into account under the rubric of “all other relevant matters” (ie, factor (e)). In addition, the AR placed significant weight on the fact that the defendant was not merely a seller or distributor of the infringing goods, but was a manufacturer of them (at [24]).

Similar to the present case, the AR also considered, inter alia, the damage or potential damage to the claimant’s goodwill and reputation (at [29]–[30]) and the defendant’s free-riding on the resources spent by the claimant to enhance its own commercial reputation and profile and gain “brand prestige” (at [33]). Although the claimant had asked for the maximum amount of $100,000 to be awarded, the AR saw fit to award only $35,000, having regard to the statutory factors and the principle of proportionality.

Before moving on, I say a few words about the AR’s finding that the defendant had used a counterfeit trade mark in Cuffz. The registered mark in Cuffz was described (at [3]) as being characterised by “interleaving ridges and valleys applied to the whole or predominant area of the surface of the product, with the ridges being of a darker shade and the valleys of a lighter shade, thus giving it an immediately recognisable two-tone effect.” In addition, the AR noted (at [24]) that the defendant had affixed its own “distinctive brand logo” on the infringing goods, this being “two small symmetrical diamond shapes under the words ‘Cuffz’”.

I have my doubts about whether the infringing mark in Cuffz was a counterfeit trade mark. This is also the view of Prof Llewelyn, who notes that the defendant’s use of its own “distinctive brand logo” (as the AR called it) in Cuffz, in addition to the Registered Mark, made it difficult to understand how this was use by the defendant of a sign that was identical to the claimant’s registered mark, or use such as to falsely represent its goods as being the claimant’s genuine goods (Llewelyn at para 41).

Putting aside the issue of whether the defendant had used a sign identical to the claimant’s registered mark, the defendant’s affixing of its own “distinctive brand logo” onto the infringing goods calls into question whether the manner in which it had used the mark would have given rise to an inference that the purpose of applying the infringing mark was to falsely represent that the infringing goods were the claimant’s genuine goods. The claimant’s registration required its application to the surface of the product. There is no doubt that the defendant applied an infringing mark to its products. But crucially, it had also applied its own “distinctive brand logo” to the surface of its products. The defendant’s inclusion of its own “distinctive brand logo” was not an extraneous factor, but inherent in the way the goods were represented.

As explained above (at [31]–[32]), not every infringing mark is a counterfeit trade mark. Accordingly, it is far from clear as to whether the remedy of statutory damages should even have been engaged in Cuffz.

That being said, on the assumption that the defendant did use a counterfeit trade mark in Cuffz, the application of the s 31(6) factors by the AR and the final quantum awarded still remain relevant as guides for the assessment of statutory damages in the present case.

I next turn to Bosch Corp v Wiedson International (S) Pte Ltd and others and another suit [2015] 3 SLR 961, which concerned the use of counterfeit trade marks. Although the remedy sought and awarded by the court there was an account of profits rather than statutory damages, I consider the quantum awarded by the court as a relevant guide to determining an award of statutory damages that is proportionate in the present instance.

In that case, thousands of potentially infringing goods were seized from the first defendant’s premises. There were also documents showing the first defendant’s infringing sales which were seized. However, these documents only represented about 10% of the defendant’s total infringing sales, and thus Tay Yong Kwang J (as he then was) saw it fit to extrapolate from those documents the profits which the defendant had earned from all of its infringing sales over a six-year period. He finally landed at a figure of $275,000 to be awarded to the first claimant and $65,000 to be awarded to the second claimant.

Finally, I consider Creative Technology Ltd v Cosmos Trade-Nology Pte Ltd and Another [2004] SGHC 5, a High Court Registry decision which was for the assessment of general compensatory damages for trade mark infringement. A raid conducted on the defendants’ premises yielded 56 infringing sound cards and one infringing CD-ROM. These were all counterfeit items and were in all material aspects identical to the claimant’s sound cards, and carried the claimant’s marks (at [7]). The claimant obtained evidence showing that the defendants had for some time been dealing in large quantities of counterfeit sound cards, namely 14,428 sound cards (see [8]). The AR considered different heads of damages, but what is relevant for present purposes is that she awarded the claimant a sum of $70,000 for the loss of business reputation and goodwill.

Considering the factors under s 31(6) of the TMA and the four cases I have referred to in the round, I am of the view that an appropriate amount to be awarded to the claimant would be $200,000.00.

Conclusion

For completeness, I add that although the claimant had tendered a set of supplementary submissions on the UK jurisprudence pursuant to my directions, I did not find the UK position to be relevant for this case as the UK does not have a statutory damages regime.

On a more general note, I should stress that I only had the benefit of the claimant’s arguments in coming to my decision. As the defendant was absent and unrepresented, the contents of this judgment have to be viewed in that light.

In addition, the assessment of statutory damages under the TMA will necessarily not be one that can be done with mathematical precision, especially since the TMA envisages a multi-factorial inquiry. Such an approach is not perfect, but the perfect cannot be the enemy of the good. It is the best a court can do when the claimant cannot prove its actual losses. Ultimately, the final award must be proportionate to the legislative scheme and to the factors in s 31(6) of the TMA. Seen in this light, the claimant’s proposed quantum of $2,900,000.00 is grossly excessive.

Finally, I am grateful to Mr Ravindran, Mr Chan and Mr Neo for their comprehensive submissions and the assistance which they rendered to me in providing me with the relevant authorities from other jurisdictions at my request.