1 The plaintiffs are Singapore companies operating gymnasiums and fitness centres (collectively “True Group (Singapore)” and are part of the “True Group”). True Group has companies in China and Taiwan (“True Group (China)” and “True Group (Taiwan)” respectively). It also had companies in Malaysia and Thailand as well (“True Group (Malaysia)” and “True Group (Thailand)” but they had ceased operations there.

2 The defendant was the former Group Chief Executive Officer (“CEO”) of True Group from 19 March 2008 until 9 May 2018. He was the director of the plaintiffs until 30 July 2021. He was found liable in breach of his contractual duties and director’s duties (see True Yoga Pte Ltd and others v Wee Ewe Seng Patrick John [2022] SGHC 155 at [31] and Wee Ewe Seng Patrick John v True Yoga Pte Ltd and others [2023] 2 SLR 323 (“True Yoga (AD)”) at [74]–[92] and [105]). True Group (Thailand) ceased operations on 9 June 2017 whilst True Group (Malaysia) ceased operations on 10 June 2017 (True Yoga (AD) at [17]-[18]). This judgment concerns the quantification of the plaintiffs’ losses.

3 Mr Tam Chee Chong (“Mr CC Tam”), who testified as the plaintiffs’ expert, is a Fellow Chartered Accountant of both England & Wales and Singapore. Mr Tham Chee Soon (“Mr CS Tham”), the defendant’s expert, is also a Chartered Accountant of Singapore. Their reports and a joint expert statement dated 5 April 2024 were duly filed. Mr CC Tam estimates the plaintiffs’ losses to be as high as S$13.852m and as low as S$5.863m, depending on which of the 12 different permutations of quantification is adopted. Mr CS Tham maintains that the plaintiffs’ losses would only be S$284,280, based on his model. Mr CS Tham did not provide other figures derived from alternative approaches.

4 The significant differences between both experts’ quantifications stem from three main areas of contention, namely:

The first issue is a matter of law, but the second and third issues are matters of fact.

5 The plaintiffs’ losses are “financial loss arising from damage to the brand equity of the ‘True’ brand” (True Yoga (AD) at [98]). The assessment of such a loss presents some difficulties because brand equity is not reflected as an objective metric, unlike the loss of the contract price of a sales agreement or damage to a physical asset. It is the intrinsic value of the brand, and can only be estimated. In the present case, both experts agree that the plaintiffs’ losses should be calculated from the loss of cash inflow (“Cash Sales”) less expenses. They agree that this method of quantifying the plaintiffs’ loss is preferable to using the loss of accounting revenue less expenses approach. Mr CC Tam explains that comparing the plaintiffs’ loss of Cash Sales “after the closure of True Malaysia and True Thailand” would reflect “the direct impact of the loss suffered by the [p]laintiffs due to the [d]efendants mismanagement”.

6 In spite of that consensus, their final results were vastly different because of many uncertain factors. They vigorously disagree as to how the plaintiffs’ loss of Cash Sales should be measured and for how long such loss should be accounted for. Both experts disagree as to whether the plaintiffs’ loss of Cash Sales had been affected by other potential factors besides True Group (Singapore)’s brand equity. There is thus uncertainty about the extent to which the defendant’s breaches of his contractual duties and director’s duties contributed to the plaintiffs’ losses. Counsel for the plaintiffs, Mr Jordan Tan, submits that since it is undisputed that the defendant had breached his fiduciary duties to the plaintiffs, any uncertainty as to what would have happened had there been no breach would be resolved against the fiduciary. Counsel for the defendant, Mr Daniel Soo, submits that the uncertainty as to what caused the plaintiffs’ loss of Cash Sales is fatal to the plaintiffs’ claim for damages because they have not proved that the defendant caused the loss in Cash Sales.

7 When, as in this case, equitable compensation is sought in a non-custodial breach of fiduciary duties, the claimant bears the legal burden of “showing a breach of a core fiduciary duty and establishing that a loss has been sustained” to make out its claim. Thereafter, “a rebuttable presumption that the fiduciary’s breach caused the loss” arises and the fiduciary bears the legal burden “of proving that the [claimant] would have suffered the loss in spite of the breach”. The upper limit “of equitable compensation is the position that the [claimant] would have been in had there been no breach” (Credit Suisse Trust Limited v Ivanishvili, Bidzina and others [2024] SGCA(I) 5 (“Credit Suisse”) at [63] citing Sim Poh Ping v Winsta Holding Pte Ltd and another and other appeals [2020] 1 SLR 1199 (“Sim Poh Ping”) at [254]). The rationale behind this burden-shifting approach is explained in these cases.

8 The onerous nature of the burden shifting approach is mitigated in cases “where there is patently no linkage between the fiduciary’s breach and the losses sustained”. In such a situation, “the discharge of the legal burden would not be an onerous one where there is clearly no causative link”. If the fiduciary shows “evidence pointing to the lack of linkage between the breach and the loss, the evidential burden may shift to the principal to adduce evidence to show otherwise” (Sim Poh Ping at [243] referring to Loo Chay Sit v Estate of Loo Chay Loo, deceased [2010] 1 SLR 286 at [14]). This of course depends on the facts.

9 It is not disputed in the present case that the defendant had breached his non-custodial fiduciary duty of good faith to act in the best interests of the plaintiffs. He therefore bears the legal burden of proving that the plaintiffs would have suffered the loss despite his breach. The quantification of the plaintiffs’ losses (see [10]–[46] below) follows this framework. Although the shift in the burden of proof (see [7]–[8] above) is not applicable to the defendant’s breach of his contractual duties, it does not change the approach regarding quantification in the present case – the plaintiffs are entitled to elect between equitable compensation for the defendant’s breach of his fiduciary duty or damages for breach of contract. They appear to have elected the former by relying on the shift in the burden of proof with regard to the uncertainties in quantification.

10 Both experts agree that the plaintiffs’ losses are represented by the loss of Cash Sales less expenses. However, they do not agree as to how the result is derived. They disagree over whether the loss of Cash Sales of the plaintiffs should be estimated with reference to:

They argue over what expenses should be subtracted from the Cash Sales.

11 The plaintiffs’ actual Cash Sales are known. The benchmarks represent various estimates of the plaintiffs’ expected Cash Sales in the hypothetical situation where the defendant had not breached his fiduciary duty. The difference between the actual Cash Sales and the expected Cash Sales reflects the loss in Cash Sales. The budget benchmark (“Budget Benchmark”) “uses the financial budget that [was] prepared internally and approved by the management to track the [p]laintiffs’ expected Cash Sales and expenses” as the plaintiffs’ expected Cash Sales. The historical benchmark (“Historical Benchmark”) estimates the plaintiffs’ expected Cash Sales and expenses by using “actual historical financial performance as the reference”.

12 The plaintiffs say that the Budget Benchmark is appropriate because the budgets in FY 2017 and FY 2018 were prepared contemporaneously and reflected what the defendant himself thought would be the plaintiffs’ performance then, and is thus a reasonable projection of the plaintiffs’ expected performance for FY 2017. The defendant maintained that there was “no fraud or dishonesty involved in the preparation of the financial budgets” and that he “would not have tolerated figures being put in the financial budgets without honest belief”. The plaintiffs further say that the Budget Benchmark is more appropriate because they reflect the growth expected of the plaintiffs in FY 2017 and FY 2018. Moreover, Mr CC Tam had compared the FY 2015 and FY 2016 budgets against the actual figures then and found that they were an accurate predictor of actual performance. Therefore, it follows that the FY 2017 and FY 2018 budgets would likely be reliable as well.

13 The defendant says that the FY 2017 and FY 2018 budgets cannot be reliable estimations of future performance. He claims that the FY 2017 and FY 2018 budgets were prepared when True Group was courting “Tongfang” as a potential acquirer. The budgets were thus presented as “the best possible picture” which “required exceptional performance” to be achieved. Additionally, the defendant says that the historical data shows that the FY 2017 budget was not realistic because it had failed to project the first six months of 2017 accurately. Actual performance was significantly lower than what the FY 2017 budget had projected. Hence, he claims that the Historical Benchmark, based off actual figures from 2017, ought to be used instead.

14 In my view, the Budget Benchmark is not reliable. It mattered that the FY 2017 budget figures for January to May 2017 significantly exceeded the actual performance of the plaintiffs in those months. In particular, actual performance was between 13% to 19% lower than the budgeted figures for the four months of February to May 2017 (the figures from June 2017 should not be used in the assessment because they would be affected by the closures of True Group (Thailand) and True Group (Malaysia) on 9 and 10 June 2017 respectively). Both experts agree that any variance that is less than 5% “is not material”. This large variance over four months suggests that the FY 2017 budget is not a reliable benchmark and supports the defendant’s claim that he had prepared unrealistic (and difficult to achieve) projections to impress Tongfang. Mr CC Tam acknowledged that this variance poses a problem as well and had made a downwards adjustment in his calculations to reflect it. His adjustments entailed a 12% downward adjustment to the figures for budgeted cash sales.

15 Mr CC Tam’s evidence (with Mr CS Tham agreeing) is that accountants would not accept a month-by-month approach towards comparing variance between the budgeted Cash Sales and the actual performance of the plaintiffs. Instead, it would be preferable to look at the variance for the entire year. Mr CC Tam explained that sometimes in a certain month, sales that “was projected to happen” did not happen and instead “happened in the following month”. He referred to this as the “compensating factor”. He explained that when there is a variance “of up to 8% to 12 %” in particular months, he would need to “understand why and see whether subsequent months have compensated that”. His evidence does not assist the plaintiffs in establishing the reliability of the FY 2017 budget. There is simply no empirical data available to check on the reliability of the FY 2017 budget after May 2017 (since the news of True Group (Thailand) and True Group (Malaysia) broke in June 2017). It is thus not possible to determine the variance of the FY 2017 budget for the entire year. The best and only option left would be to assess the variance (and reliability) of the FY 2017 budget using the actual performance from January 2017 to May 2017.

16 In this connection, the inability to establish the reliability of the FY 2017 budget means that Mr CC Tam’s downward adjustments to account for its variance remains unreliable. Mr CC Tam himself had explained that when months of higher variance occur, one has to “see whether subsequent months have compensated that”. That is why it was preferable to look at variance for the entire year rather than month-by-month variance. Unfortunately, there is no empirical evidence after May 2024 to show if the higher variance from February to May 2024 had been compensated for. It is also not known if the variance had increased or decreased.

17 Moreover, the purported reliability of the FY 2015 and FY 2016 budgets are not relevant to the reliability of the FY 2017 budget, given the existence of contemporaneous data showing that the FY 2017 budget is unreliable. For the same reason, what the defendant thought the plaintiffs could achieve in FY 2017 and FY 2018 (with or without the input of advisors) is also not relevant. The need to account for the growth expected of the plaintiffs is not a relevant consideration too. No evidence has been adduced to show why growth should be expected of the plaintiffs in FY 2017 and 2018. On the contrary, as Mr Tan had pointed out and Mr CS Tham had agreed, Mr CS Tham’s calculations of Cash Sales per outlet from the first half of 2015 to the first half of 2017 show “no specific trend”. There is no reason to assume that the plaintiffs’ Cash Sales would have grown had the defendant not breached his fiduciary duty.

18 In contrast, the Historical Benchmark does not have the same problems. It is based on the estimate of the plaintiffs’ expected Cash Sales to its historical performance, and avoids overestimation of the plaintiffs’ losses because it does not assume that the plaintiffs’ Cash Sales have grown when there is no empirical evidence of such future growth, and is thus the more appropriate gauge.

19 There are disputes as to how the Historical Benchmark should be implemented. The plaintiffs want their performance in July 2017 to June 2018 to be estimated by a direct comparison with the preceding one-year period (ie, July 2016 to June 2017). This includes a downward adjustment by Mr CC Tam to the Cash Sales of July and August 2016 to account for the closure of California Fitness, a rival of the plaintiffs in the fitness industry. On the other hand, the defendant wants to extrapolate the plaintiffs’ expected performance in June 2017 to August 2017 by using the plaintiffs’ median Cash Sales of the first five months of 2017. He says that avoids using “the most drastic overperformance” that was “seen in the months of July and August 2016” due to the closure of California Fitness. It obviates the need to estimate the impact of California Fitness’s closure on the plaintiffs’ Cash Sales as well.

20  I prefer and adopt the plaintiffs’ approach as to how the Historical Benchmark should to be applied. I accept that having a direct comparison with the preceding 1-year period would capture the seasonality in the fitness industry across the various months. It also avoids any issues with variance (see [15] above). I agree with the plaintiffs that contrary to Mr CS Tham’s claims, his taking of the median for five months in the year does not eliminate the seasonal highs and lows in the full calendar year. I accept that Mr CC Tam’s proposed downward adjustment to the Cash Sales of July and August 2016 to account for the closure of California Fitness is adequate. The Cash Sales for September to December 2016 are roughly within the same range as the Cash Sales for January to June 2016. The large increase in Cash Sales from June 2016 to July and August 2016 disappears in September 2016 as well, where there is a significant drop in Cash Sales for that month. There is no evidence to suggest that the impact of California Fitness’ closure in July 2016 continued till the end of 2016, and that further downward adjustments to the September to December 2016 Cash Sales ought to be made.

21 As for the dispute over whether the sums collected from existing members on a monthly basis (“EFT Income”) should be included in the quantification exercise, I am of the view that there is no reason to exclude them. If, as the plaintiffs claim, the financial data shows that there were losses of EFT Income, the defendant bears the legal burden of proving that the EFT Income would have been lost in any case. The onus thus lies on the defendant to prove his claim that recurring memberships could not be cancelled, and that no customer had in fact stopped paying for their membership upon learning of the negative publicity. The same applies to the defendant’s claim that Cash Sales from EFTs could have tapered off regardless of any negative publicity. If there are no losses of EFT Income, the plaintiffs cannot claim such compensation from the defendant anyway.

22 Parties disagree over the length of time over which the defendant’s breach of fiduciary duty affected the plaintiffs’ losses. The longer this length of time, the more the defendant would have to compensate the plaintiffs. The plaintiffs say that their losses should be assessed for a period of 12 months from July 2017 to June 2018, in line with the objective financial data and the documentary evidence. The defendant disagrees and says that the plaintiffs’ prolonged poor performance cannot be a basis to adopt a longer period of loss. He says that the period of one year advanced by the plaintiffs is arbitrary, and that it is not possible to measure their losses in terms of Cash Sales after January 2018 due to changes in the membership model, and moreover, there is evidence that the impact of the negative publicity had diminished after a couple of months.

23 My view is that the plaintiffs’ time period of 12 months from July 2017 to June 2018 should be adopted for the present quantification exercise. The objective financial data shows that the average number of memberships sold every month had fallen from July 2017 to June 2018 following the announcement of the closures of True Group (Thailand) and True Group (Malaysia) on 9 and 10 June 2017 respectively. This is consistent with the pattern of the new member sales figures for June 2017 (the month where closures were announced). The sales from new members during the period of 1 to 9 June (before the closures were announced) was $1.18m. In contrast, the sales from new members from 10 to 30 June (after the closures were announced) was just $0.44m.

24 The defendant relies on Continental Steel Pte Ltd v Nippon Steel & Sumitomo Metal Southeast Asia Pte Ltd and another [2023] 5 SLR 445 (“Continental Steel”) at [256] and says that a fall in demand cannot be used to determine the period of loss unless it had been established that the fall in demand was wholly caused by the damage to reputation. Continental Steel is a case about a tortious claim for defamation and does not apply where there is a breach of fiduciary duties and a shift of the burden of proof (see [7]–[9] above). Similarly, his assertion that the one-year period is arbitrary and that changes in the membership model rendered it impossible to measure the plaintiffs’ losses in terms of Cash Sales does not assist him (see also [41]–[46] below).

25 The documentary evidence relied on by both parties do not, in my view, change the above assessment. The plaintiffs say that the defendant had prepared a “Revitalisation Business Plan” which he presented to Tongfang to seek further funding, and this plan had referenced the impact of the closures on business. In a slide deck (dated 25 January 2018) that was prepared for a management meeting, the difference between the FY 2017 forecast and actual performance was explained to be (in part) because the “[c]losure of Thailand and Malaysia in mid-June 2017 has affected our members’ confidence”. In the revised “Revitalisation Business Plan” (dated 3 March 2018), one of the weaknesses of True Group’s business was stated to be “[b]rand perception due to negative publicity in 2017”. The defendant says that Tongfang had published a FY 2018 Interim Report (on 27 September 2018) stating that the closure of the clubs in Thailand and Malaysia had a short-term impact on the clubs in Singapore, and “had mostly diminished after the initial couple of months”.

26 The contradictory documentary evidence is not consistent as to whether the effect of the plaintiffs’ breach was short-lived. Furthermore, I must take the circumstances into account. All the comments contained in the documentary evidence put forth by the parties were expressed in a general sense, lacking explanations as to how those conclusions were reached as well as references to the basis for supporting such conclusions. They do not show how long (and to what extent) the impact of the closures on the plaintiffs’ business had persisted. For instance, the plaintiffs point out that the FY 2018 Interim Report was prepared for the period from January to June 2018, and does not relate to the second half of 2017. Furthermore, none of the comments were intended to address the issue of the plaintiffs’ losses resulting from the defendant’s breaches. The comments were instead made in the context of achieving certain aims. For the “Revitalisation Business Plan”, it was to seek further funding from Tongfang. For the FY 2018 Interim Report, it was to update shareholders. They thus cannot be taken as a reliable indication of the length (and extent) the impact of the closures had on the plaintiffs’ business.

27 The parties agree that sales commissions need to be taken into account when calculating the plaintiffs’ loss. However, they disagree as to what those sales commission figures are. The plaintiffs say that the category of “staff costs” already includes the salaries and commissions paid by the plaintiffs. The defendant wants sales commission to be directly accounted for using a blended sales commission figure (which averages out the sales commission rates for new member sales and personal training sales). I accept the evidence of Ms Reina Lim (“Ms Lim”), a director of the plaintiffs, that the staff costs of the plaintiffs which is reflected in the monthly management accounts, “refers to the salaries and commissions being paid to the employees of the [p]laintiffs”. Since the category of “staff costs” already reflects the commission component, it is unnecessary to further complicate the quantification exercise by directly estimating sales commissions using a blended sales commission figure. This is especially so since Mr CC Tam had explained that:

Any cost saving from the decrease in staff costs would be “taken into account in the quantification of loss… by reducing the damages due to the [p]laintiffs”.

28 Besides the dispute over the correct way to model the plaintiffs’ loss of Cash Sales, another significant part of the defendant’s case is that there were other unrelated developments that affected the plaintiffs’ financial performance. Many of these unrelated developments concerned changes in the plaintiffs’ membership and pricing structure. For context, there are four different types of membership packages offered by the plaintiffs. Mr CC Tam summarises them in his expert report as follows:

29 On 12 June 2017, the plaintiffs replaced the option for a prepaid 24-months Term Membership (the “2-Year Plan”) priced at $2,000, with two separate memberships of one year each, with the first-year costing $1,850 and the second-year costing $150 (the “1+1 Plan”). The defendant says that although the final price for two years of memberships remained the same when customers paid this sum upfront, this affected customers that used instalment payment plans (“IPPs”) as they had to make a higher monthly payment. Under an IPP, the customer’s credit card company would make full payment of the Term Membership to the plaintiffs, while the customer would pay their credit card company in monthly instalments. According to the defendant, under the 2-Year Plan, customers had to pay $83 over 24 months, whilst under the (new) 1+1 Plan, customers had to pay $304 in the first month and $154 for the next 11 months. This was thus an unpopular plan with customers. It also drove customers to memberships where fees were charged on a monthly basis (“Dues Memberships”) because there was a lower payment of $119 per month for a Dues Membership. Both of these consequences would have led to a negative impact on the plaintiffs’ Cash Sales. The defendant says that his hypothesis is borne out by the empirical data examined by Mr CS Tham.

30 In my view, there are crucial problems with the defendant’s case. First, the assertion that the 1+1 Plan drove customers to Dues Memberships because there was a lower monthly fee is flawed. On the defendant’s own case, the 1+1 Plan required instalment payments of $304 in the first month and $154 per month for 11 months, plus a one-time payment for $150 for two years of membership. The total sum due under the 1+1 Plan is clearly less (and thus more desirable to consumers) than the sum due under the Dues membership, which is $119 for 24 months for two years of membership.

31 Second, the evidence which the defendant relies on, ie, the larger decline in Cash Sales from new members (33%) as compared to the drop in new members signing up (10%) in the months of July, August and September (as compared to March, April and May) does not assist him. These figures do not show that the change from the 2-Year Plan caused the drop in new members signing up instead of the defendant’s breach. Even if I accept the defendant’s contention that new members were still signing up (albeit fewer than usual), and that it is clear a majority of them signed up for the Dues Membership and the shorter 12-month Term Memberships (which brought in less Cash Sales per month), this is not indicative that this shift in customer preferences are primarily due to the new pricing structures. It may be that after the negative publicity of the closure of True Group (Thailand) and True Group (Malaysia), customers were less enthusiastic about longer Term Memberships.

32 In this connection, I accept that the recovery of the sales of 24-month Term Membership post-September 2017 (when the 2-Year Plan was reintroduced) goes towards establishing that the 2-Year Plan was popular. However, this does not have much bearing on proving that the shift away from the 2-Year Plan caused the plaintiffs’ losses. As will be recalled, it is not disputed that only customers who were paying via IPPs would be affected by the change from the 2-Year Plan to the 1+1 Plan (see [29] above). It is the defendant’s case that IPPs were removed as an option for the plaintiffs’ customers in September 2017. This resulted in total Cash Sales paid via IPPs to be almost non-existent in October 2017. Therefore, the shift back to the 2-Year Plan would not have much positive effect on regaining previously affected customers (that would want to pay via IPPs). Moreover, this data point must be interpreted together with the performance of the plaintiffs as a whole during that period of time. As the plaintiffs had pointed out, the Cash Sales was still significantly less during September to November 2017 (as compared to March to May 2017) even though the popular 2-Year Plan was reintroduced. This should not be the case if the shift away from the 2-Year Plan caused the plaintiffs’ losses. I agree with the plaintiffs that this indicates that the shift from the 2-Year Plan to 1+1 Plan could not have been the cause of the plaintiffs’ loss.

33 Third, even if I accept that the shift away from the 2-Year Plan to the 1+1 Plan caused some of the plaintiffs’ loss, there is no reliable way to estimate the quantum to be deducted. The defendant testified at trial that around 70% of members pay via IPP. However, Ms Lim gave contrary evidence that most of the plaintiffs’ customers “do not use instalment payment plan”. In the absence of adequate evidence about the number of customers who used IPPs to purchase the 2-Year Plan (and to a smaller extent the 1+1 Plan), there remains too much uncertainty as to the number of customers who would have been impacted by the shift away from the 2-Year Plan – and by extension the impact on the plaintiffs’ losses.

34 The defendant says that the plaintiffs’ loss was in part due to its decision to stop offering 36-months Term Memberships after June 2017. Each 36-months Term Membership sold would result in an instant Cash Sales of $2,874 as compared to Cash Sales of $1,999 for a 24-months Term Membership, or Cash Sales of $119 for a Dues Membership. Therefore, even if there was no decrease in total number of memberships sold, a smaller amount of 36-months Term Memberships sold would still mean a significant decrease in Cash Sales.

35 I am of the view that the defendant’s claim here fails because he has not established that the plaintiffs had a policy to stop offering 36-months Term Memberships. The only evidence which the defendant can rely on to establish the existence of such a policy is an email (dated 5 June 2017) sent by Mr Ganeson Bupendra (“Mr Ganeson”), the senior vice president of sales, informing the defendant of the “NEW PREPAID TERM PRICINGS EFFECTIVE 12 JUNE 2017”. The 36-months Term Membership was not one of the options available. Ms Lim’s evidence does not support the defendant’s claim here because she was merely responding to counsel’s question that was “based on the table” found in Mr Ganeson’s email. Read in its proper context, Ms Lim’s evidence was that she did not know if the plaintiffs promoted 36-months Term Memberships after June 2017.

36 Despite the purported existence of a policy where 36-months Term Memberships were not available, sales of such memberships continued. A table summarising the sales of 36 months Term Memberships in 2017 is as follows:

37 In my view, the continuing sale of 36-months Term Memberships in the second half of 2017 indicates that there was no firm policy on the plaintiffs’ part to stop such sales. I do not accept the defendant’s explanation that the sales staff were simply “not encouraged” to sell 36-months Term memberships but that they were “technically speaking, allowed to sell it” by speaking to the management to obtain permission. There is no evidence to support this assertion. The figures above also do not help him. They support an equally good explanation advanced by the plaintiffs that the drop in sales of 36-months Term Memberships after June 2017 was due to the defendant’s breaches. It is sensible that once news of the closures broke, customers would be less inclined to purchase longer Term Memberships from the plaintiffs. This is consistent with the fact that July 2017 and August 2017 are the worst hit months in terms of 36-month Term Membership sales.

38 Moreover, I agree with the plaintiffs that if the alleged removal of the 36-month Term Memberships in June 2017 had caused such a major negative impact on the plaintiffs, one would reasonably expect this membership option to have been restored to reverse the negative impact. This was especially so since the plaintiffs “would have immediately felt the financial impact of the removal of such memberships” by July 2017. However, this was not done, and the defendant did not offer any explanation for this. There is no evidence to support his claim that he had “discussed [this issue] intensely internally” or that he was “forced to comply with [the direction] of the finance team who reported to Tongfang”. This was in contrast to the shift back to the 2-Year Plan (see [32] above).

39 The defendant says that the removal of IPPs as a payment option for customers after September 2017 had an impact on the plaintiffs’ Cash Sales. At first, he claimed that the withdrawal of these IPP services by United Overseas Bank (“UOB”), Overseas-Chinese Banking Corporation Limited (“OCBC”) and the Development Bank of Singapore (“DBS”) contributed to the plaintiffs’ loss. At trial, he conceded that OCBC had suspended its IPP service with the plaintiffs “because of the adverse news reported in the newspapers about the closures of [True Group (Malaysia) and True Group (Thailand)]” and because the plaintiffs did not agree to the subsequent conditions imposed by OCBC. He also conceded that UOB’s termination of its IPP service with the plaintiffs which happened “back in November 2016 would no longer have an impact on the plaintiffs’ financial performance” for “June 2017 onwards”. Therefore, the only remaining dispute is over whether DBS’s termination of its IPP service with the plaintiffs on 25 September 2017 had an impact on its Cash Sales.

40 The defendant’s assertion here is inconsistent with the objective financial data. As the plaintiffs have pointed out, if the removal of the DBS IPP option was the cause of the plaintiffs’ loss of Cash Sales, it would be reasonable to expect that the plaintiffs’ Cash Sales would decrease from September to October 2017 after DBS terminated its IPP service with the plaintiffs. As it turned out, the financial data reflected that new member sales for both months remained the same — there was no decrease in new member sales despite the removal of the IPP options. The defendant himself emphasises that the percentage of new member sales paid via IPP in October 2017 was negligible (less than 1%), as compared to previous months. Yet, this had no effect on reducing new member sales that month. There is thus no basis for saying that the removal of the IPPs after September 2017 was an intervening cause which contributed to the plaintiffs’ loss.

41 The Dues Plus Membership covers a period of 24 to 36 months. It entails a customer being charged a portion of the membership fee and processing fee upfront with the remainder of the membership fee to be charged on a monthly basis (see [28] above). It is not disputed that the change from Term Memberships to Dues Plus Memberships caused the plaintiffs’ Cash Sales to decrease in the short term (as the full fee is not paid to the plaintiffs’ upfront) in exchange for an overall increase in profitability in the longer run (as the remainder of the fee is paid to the plaintiffs in subsequent months). In other words, the sale of any number of Dues Plus Memberships in a certain month would always lead to a smaller Cash Sales figure as compared to the sale of the same number of Term Memberships (of the same duration) in that same month.

42 The defendant says the shift to Dues Plus Memberships reduced the Cash Sales for each 24-month membership by a little more than 50%. This was a material impact on the plaintiffs’ Cash Sales and displaced any residual effects from the negative publicity stemming from the defendant’s breach. The plaintiffs disagree and frame the issue as one of whether the change to the Dues Plus Membership should be attributed to the defendant’s breach. The plaintiffs say that the Dues Plus Memberships were only adopted because of the poor performance in the second half of 2017 owing to the defendant’s breach. Therefore, it is not an independent cause that broke the chain of causation. The legal burden of proof remained on the defendant to prove that the shift towards the Due Plus Memberships was an independent cause of the plaintiffs’ losses in the first half of 2018.

43 The plaintiffs first have to establish that they have suffered a loss before the legal burden shifts to the defendant to show that he did not cause the loss (see [7]–[9] above). In the present case, the reduction in Cash Sales owing to the shift from the Term Memberships to the Due Plus Memberships cannot be taken to be losses. It is not disputed that the plaintiffs earned around the same amount from the sale of a Term Membership and a Due Plus Membership (assuming no difference in length). The only difference is that the former results in greater Cash Sales being realised in the month of sale, whilst the latter results in lesser Cash Sales being realised in the month of sale, with the remainder realised over the term of the membership. There is thus no loss. Since the shift to Due Plus Memberships did not result in loss to the plaintiffs, whether the defendant’s breach was the cause for this shift of strategy is irrelevant.

44 The fact that the difference in Cash Sales due to the shift from Term Memberships to Due Plus Memberships cannot be considered loss is not fatal to the plaintiffs’ attempt to establish its losses in the first half of 2018 using the Historical Benchmark. As long as the plaintiffs were able to reasonably control for this factor, any other differences in Cash Sales would still be loss to the plaintiffs which the defendant has to explain.

45 I am of the view that Mr CC Tam’s approach towards accounting for this change in the plaintiffs’ business model from Term Memberships to Dues Plus Memberships is a sensible and reasonable one and should be adopted. In his expert report, Mr CC Tam adjusted the Cash Sales of December 2017 and the first half of 2018 to a full prepaid Term Membership business model. This was done by using the revenue from Term Memberships from October 2017 as a proxy (the “Proxy Term Revenue”) for December 2017 and the first half of 2018. All revenue from Dues Plus Memberships were removed in these months. I find Mr CC Tam’s approach to be reasonable because using the revenue from Term Memberships in October 2017 is fair — this revenue did not differ too greatly from the other months after revenues had stabilised from the closures. Mr CC Tam’s approach also includes changes that reduce the plaintiffs’ loss. He moderated the December 2017 revenue from Term Memberships significantly. He removed the revenue from Dues Plus Memberships in January 2018 without adding to its Term Memberships revenue. Pertinently, in keeping the defendant’s assertion that the shift to Dues Plus Memberships would effectively reduce the Cash Sales for each 24-month membership by a little more than 50% (see [43] above), the actual revenue from Dues Plus Memberships were replaced with Proxy Term Revenues which are around twice as high.

46 Mr CS Tham does not have an alternative way to account for the change in the plaintiffs’ business model from Term Memberships to Dues Plus Memberships in 2018. He simply criticises Mr CC Tam’s approach on the basis that it is not suitable to use the month of October 2017 as a proxy for actual Cash Sales for every month in the first half of 2018. This is because negative publicity may not have had the same impact in June 2018 as it did in October 2017 and there is no evidence that the negative impact was close. Mr CS Tham’s criticism is valid but there is no requirement that Mr CC Tam’s approach has to be perfect. For the reasons above (at [45]), I find that Mr CC Tam’s approach remains reasonable and sensible to adopt. The defendant’s position of simply not attempting to measure the plaintiffs’ losses in the first half of 2018 is untenable in the light of the objective financial data showing that the average number of memberships sold every month had fallen from July 2017 to June 2018 (see [23] above).

47 In conclusion, my decision on the quantification of the plaintiffs’ losses is as follows:

(a) The Historical Benchmark is to be adopted (see [11]–[18] above).

(b) The plaintiffs’ approach towards utilising the Historical Benchmark is to be adopted, inclusive of the additional downward adjustment made by Mr CC Tam to the Cash Sales of July and August 2016 to account for the closure of California Fitness (see [19]–[20] above).

(c) EFT Income is to be included in the quantification exercise (see [21] above).

(d) The time period of 12 months from July 2017 to June 2018 is to be adopted for the quantification exercise (see [22]–[26] above).

(e) The plaintiffs’ approach towards subtracting expenses is to be adopted (see [27] above).

(f) Mr CC Tam’s approach towards accounting for the change in the plaintiffs’ business model from Term Memberships to Dues Plus Memberships is to be adopted (see [41]–[46] above).

48 Parties are to instruct their experts to reach an agreement on the value of the plaintiffs’ losses based on the above parameters. Parties have liberty to apply if an agreement cannot be reached. I will hear parties on costs.