1 This was an application for leave to refer questions of law of public interest to the Court of Appeal for its determination, pursuant to s 397 of the Criminal Procedure Code (Cap 68, 2012 Rev Ed) (“the CPC”). The application arose from the prosecution of six leaders of City Harvest Church (“CHC”), the trial of which was concluded in 2015. The accused persons’ appeal was heard in 2016 and decided earlier this year. This application was filed following the High Court’s decision on the appeal. We heard and dismissed it on 3 July 2017, giving brief reasons at the hearing. We now provide the full grounds of our decision.

2 The criteria for the granting of leave to refer questions to the Court of Appeal is strict, and for good reason. The criminal reference mechanism cannot be used as a means by which a dissatisfied litigant institutes a further (and backdoor) appeal against a decision of the High Court which, in the exercise of its appellate jurisdiction, has reviewed the findings of the District Court and reached a decision on the arguments advanced by the parties at the appeal. The schema of the CPC establishes only one tier of appeal – in this case, from the District Court to the High Court – and it is not for the litigant to manufacture a second tier of appeal through the abuse of court processes intended for other purposes. The CPC allows the litigant to have his proverbial day in court, and that day comes to a close when the trial has run its course, the trial judge has rendered his decision and the appellate court has reached its determination as to whether there is any merit in the grounds of appeal.

3 What is also at stake here is the principle of finality in the judicial process. If the court is not careful to guard against applications that amount to nothing more than backdoor appeals, a disingenuous litigant could conceivably keep spinning out applications ad infinitum through the criminal reference mechanism in order to prolong the criminal proceedings indefinitely, thereby delaying the commencement of the sentence lawfully imposed on him. Indeed, even one such application would – in and of itself – constitute an abuse of process if it raises no question of law of public interest and is filed for no other reason than as a delaying tactic, aimed at frustrating the efficient and expeditious conduct of criminal proceedings.

4 In the context of the present application, we also note that the questions which the Applicant sought to refer to the Court of Appeal were closely scrutinised not only in the District Court but also by a specially constituted three-Judge coram of the High Court on appeal. The judgments of both courts, in fact, span a total of approximately 570 pages. In our view, the decision of a three-Judge coram of the High Court should generally represent a final and authoritative determination of the issues arising from the case. Therefore, as a general matter, no leave will – absent exceptional circumstances that we will elaborate upon below – be given for a further reference to be made to the Court of Appeal. We were satisfied that no such exceptional circumstances existed in this case. All the questions raised by the Applicant pertained to elements of the offences of criminal breach of trust (“CBT”) and falsification of accounts, which the High Court unanimously found were satisfied on the facts of this case. In the circumstances, we found that this application should be rejected on this ground alone.

5 In any event, we were satisfied that the questions sought to be referred by the Applicant were either questions of fact, settled questions of law, questions involving the application of settled law to the facts of the case, or questions that simply did not arise from the case before the High Court. These were plainly not proper subjects for a criminal reference as they did not satisfy the requirements under s 397(1) of the CPC.

6 The background to this application is set out in detail in the first instance judgment of the Presiding Judge of the State Courts (“the Judge”), Public Prosecutor v Lam Leng Hung and others [2015] SGDC 326 (“the Conviction GD”), and the appellate judgment of the High Court, Public Prosecutor v Lam Leng Hung and other appeals [2017] 4 SLR 474 (“the MA Judgment”). We briefly recount the facts which are material to the present application, including the relevant findings of fact made by the courts below. It is worth noting from the outset that due to the nature of this application, the findings of fact made by the High Court could not be subjected to challenge and therefore had to be taken as full and accurate, because the criminal reference mechanism does not provide a means for the reopening of factual findings.

7 CHC is a Singapore “mega-church” which was rapidly expanding at the turn of the century. In 2002, it officially embarked on “the Crossover” – a project that involved Ms Ho Yeow Sun (“Sun Ho”), a co-founder of CHC, recording secular pop music albums as a means of evangelical outreach. This was part of CHC’s vision of using popular culture to spread its religious creed. At the same time, the church was actively looking for suitable premises to accommodate its growing congregation and raised large amounts of funds for this purpose through a pledge campaign. These donations were segregated in a Building Fund (“the BF”) and the pledge cards given to the church members explicitly stated that these monies were to be used “for the purchase of land, construction costs, rentals, furniture and fittings”.

8 The six accused persons were leaders of CHC. They are as follows (in order of their position in the church hierarchy):

(a) Kong Hee, the founder and senior pastor of CHC. He was the president of the CHC management board (ie, its board of directors). He is also Sun Ho’s husband.

(b) Tan Ye Peng (“Ye Peng”), a deputy senior pastor of CHC. He was a senior member of the CHC board at all material times.

(c) Lam Leng Hung (“John Lam”), who was either the secretary or treasurer on the CHC board at various times and the chairman of the CHC investment committee in 2007 and 2008.

(d) Chew Eng Han, the Applicant. He was a senior member of the CHC board who held various positions including vice-president from 2006 to July 2007. In July 2007, he resigned from the board so that a company of which he was sole director, AMAC Capital Partners (Pte) Ltd (“AMAC”), could be appointed as CHC’s fund manager.

(e) Serina Wee Gek Yin (“Serina”), who was a member of the CHC board from 2005 to 2007 and the finance manager of the church until 2008. She was the administrator of the Crossover.

(f) Tan Shao Yuen Sharon (“Sharon”), a member of CHC’s accounts department who took over from Serina as finance manager in 2008. She is the only accused person who has never been a member of the CHC board.

9 When the Crossover was launched in 2002, it was focused on the Asian market with Sun Ho releasing Mandarin pop albums. The project had the support of the CHC board and the initial two albums were directly funded by CHC. This arrangement, however, ceased after Roland Poon, an ordinary member of the church, made public allegations that CHC was giving excessive attention to Sun Ho and misusing its funds to promote her career. These allegations generated negative publicity and the CHC board issued a written response, published in The Straits Times, that church funds had not been used to purchase Sun Ho’s albums or to promote her career. The executive members of CHC (“the EMs”) were also told by Kong Hee at an annual general meeting (“AGM”) on 27 April 2003 that no church funds had been used for the Crossover. This was not true. The version of events relayed to the EMs was that the monies used to promote Sun Ho’s albums had in fact come from the family of a church member, Wahju Hanafi (“Wahju”), a wealthy Indonesian businessman who was a member of CHC.

10 After this incident, the accused persons decided that greater distance should be placed between CHC and Sun Ho’s music career to avoid further negative publicity. In particular, they agreed that they had to be “discreet” about the source of the funds used to finance Sun Ho’s music production, including publicity and promotional expenses.

11 To this end, Xtron Productions Pte Ltd (“Xtron”) was incorporated in June 2003 with three shareholders: John Lam, the Applicant and the Applicant’s wife. All three were also its directors. Xtron was, in appearance, an independent firm providing artiste management services to Sun Ho. But the Judge, with whom the High Court agreed, found that Xtron was in substance no more than an extension of CHC and was controlled entirely by the church, and in particular by Kong Hee and Ye Peng, with the directors no more than figureheads.

12 From 2003, Xtron financed Sun Ho’s music career using monies from various sources, including donations and revenue from CHC, for various event management and audio-visual and lighting services which Xtron provided to the church. These funds, however, proved insufficient after Kong Hee resolved that the Crossover, and therefore Sun Ho’s music career, should to be extended to the USA. Specifically, in May 2006, a famous American executive producer, Wyclef Jean, was brought into the project. This significantly increased the amount of money needed to fund the Crossover.

13 Kong Hee, Ye Peng and the Applicant considered ways to raise more funds for Xtron to meet the Crossover’s increased financial needs. Initially, they contemplated Xtron taking a bank loan, but abandoned this option after the interest rates offered by the banks were judged to be too high. The Hong Kong bank Citic Ka Wah, for instance, offered a loan of $9m at an interest rate of 16% per annum.

14 Eventually, upon the Applicant’s suggestion, a plan was hatched for Xtron to take a loan from the BF, notwithstanding that it was a restricted fund meant to be used for building-related expenses. To facilitate this loan, the CHC investment committee, which included Ye Peng, the Applicant, John Lam and Serina, drafted and approved an investment policy in June 2007. The investment policy allowed CHC to invest surplus monies from the BF to generate financial return and thereby maintain the purchasing power of the fund. It also set out the types of permissible investment, such as Singapore dollar-denominated fixed deposits or gold, and the maximum percentage of the overall portfolio that could be allocated to each type of investment.

15 The investment policy was unanimously approved by the CHC board. Shortly thereafter, an extraordinary general meeting (“EGM”) of CHC was held on 7 July 2007 at which Kong Hee told the EMs that it was in the interests of CHC for the monies in the BF to be invested to generate financial returns rather than simply being left untouched, since the church was unlikely to acquire a building anytime soon. The EMs were also informed that AMAC, with the Applicant as its director and major shareholder, would be appointed as CHC’s fund manager to invest the sums in the BF. The EMs were not informed, however, of Xtron’s existence or the plan to use monies from the BF to fund the Crossover.

16 On 17 August 2007, Xtron and AMAC (as fund manager of CHC) entered into a bond subscription agreement (“the Xtron BSA”). Under the Xtron BSA, AMAC agreed to subscribe to bonds issued by Xtron of up to $13m in value at an interest rate of 7% per annum (“the Xtron bonds”). The Xtron bonds had a maturity period of two years and were due to expire on 16 August 2009. At the time the Xtron BSA was entered into, Xtron was in a net deficit of approximately $3.44m. Nevertheless, $13m was transferred from the BF to Xtron pursuant to the Xtron BSA from August 2007 to March 2008 in four tranches. These monies were used for the Crossover.

17 In mid-2008, under pressure from CHC’s auditors to disclose the true facts surrounding the Xtron bonds, including the uncertainty of repayment given Xtron’s consistently loss-making position and the identity of Sun Ho as a “key player” in Xtron, a decision was made to take Sun Ho out of Xtron. This decision was effectively executed in two stages.

18 First, Sun Ho was transferred from Xtron to another company, Ultimate Assets (“UA”), which was wholly owned by Wahju. To prevent an impairment of the Xtron bonds, CHC and Xtron entered into an amended bond subscription agreement (“ABSA”) on 20 August 2008 to purchase $18.2m of bonds with a maturity date of ten years and an interest rate of 5% per annum. The original Xtron bonds of $13m in value were subsumed under the ABSA, thereby lowering the interest rate from 7% to 5% per annum and extending their maturity period from two to ten years. In this manner, the accused persons were also able to postpone the bond redemption date. Pursuant to the ABSA, fresh funds were disbursed to Xtron which were used as part payment towards the purchase of a commercial building known as “The Riverwalk”. The rest of the purchase price for The Riverwalk was financed by a bank loan secured by a mortgage over the property.

19 Next, two months later, on 7 October 2008, CHC and another company, PT The First National Glassware (“Firna”), entered into a bond subscription agreement (“the Firna BSA”). Under the Firna BSA, CHC was to subscribe to a maximum of $24.5m in bonds from Firna that would mature in three years and yield interest at a rate of 4.5% per annum (“the Firna bonds”). Firna was an Indonesian glassware manufacturing company of which Wahju was the controlling shareholder. But the Firna bonds were never intended to be available for Firna’s glass factory business; nor was Firna’s revenue ever intended to go toward the redemption of the Firna bonds. Instead, the plan was to use the Firna bond proceeds to fund the Crossover. As the courts below found, Wahju was no more than a conduit through whom the funds flowed, and Kong Hee, assisted by Ye Peng, the Applicant and Serina, had complete control over the Firna bond proceeds.

20 The EMs were largely kept in the dark about these transactions. Although Kong Hee informed them at an EGM on 10 August 2008 that $18.2m worth of bonds would be purchased from Xtron, they were led to believe that these monies would be used to acquire The Riverwalk. In fact, the EGM was the first time the EMs were introduced to Xtron. However, they were not told of the earlier Xtron bonds or that Xtron would be taking a bank loan to partially finance the purchase of The Riverwalk. Nor were they were informed that CHC would also be entering into the Firna BSA.

21 From October 2008 to June 2009, a total of $11m was transferred in five tranches from the BF to Firna, pursuant to the Firna BSA. Of this $11m, about $7.56m was used for the Crossover and $2.5m was used by Wahju for his personal expenses.

22 In late 2009, after further questions were raised by CHC’s auditors about the Xtron and Firna bonds, Ye Peng, Sharon, the Applicant and Serina decided that the bonds had to be redeemed. In the same period, continued efforts were made by CHC to secure suitable premises. These two plans overlapped and it was contemplated that: (a) once acquired, Xtron would own the premises for CHC’s benefit; (b) CHC would pay Xtron advance rental to lease the building; and (c) Xtron would thereafter use the monies from the advance rental to redeem the bonds.

23 The accused persons then procured a series of transactions between October and December 2009 to redeem the Xtron and Firna bonds (“the round-tripping transactions”). These transactions are detailed at [45] of the MA Judgment, and can be summarised as follows:

(a) Partial redemption of Firna bonds through Tranche 10 of the SOF. On 2 October 2009, CHC transferred $5.8m from the BF to AMAC as payment for Tranche 10 of a Special Opportunities Fund (“SOF”) administered by AMAC. The SOF was an ongoing fund comprising several tranches under which AMAC guaranteed the principal and a fixed return to the investor. The $5.8m (less a telegraphic transfer fee of $20) was transferred to UA, which then transferred $5.3m to Firna. Firna thereafter transferred $5,228,750 to CHC on 9 October 2009, which was recorded in CHC’s books as a partial redemption of the Firna bonds.

(b) Redemption of remaining Firna bonds through Tranche 11 of the SOF. On 15 October 2009, CHC transferred $5.6m from its General Fund (“the GF”) to AMAC as payment for Tranche 11 of the SOF. AMAC transferred this sum (less a telegraphic transfer fee of $20) to UA. UA transferred $6.1m to Firna, and Firna transferred $6,061,950 to CHC on 22 October 2009, which was recorded in CHC’s books as redemption of the remaining Firna bonds with interest.

(c) Redemption of Xtron bonds through set-off of advanced rental under the ARLA. Sometime after 15 October 2009, CHC signed an Advance Rental License Agreement with Xtron (“the ARLA”). Under the ARLA, CHC had the right to use and occupy the premises provided by Xtron for eight years, in return for the payment of advance rental of $46.27m to Xtron. A further $7m was payable by CHC as a security deposit, making the total sum due from CHC to Xtron under the ARLA approximately $53.27m. From this amount, Xtron set-off $21.5m to fully redeem the bonds it had issued to CHC.

(d) Redemption of Tranches 10 and 11 of the SOF through advanced rental under the ARLA. On 6 November 2009, CHC transferred $15,238,936.61 to Xtron. Of this sum, $12m was for part payment of the advance rental under the ARLA with the remaining sum of $3,238,936.61 being Goods and Services Tax (“GST”) for the advance rental. Xtron transferred $11.455m of this sum to Firna, which then transferred a total of $11.476m to UA. UA transferred the same amount to AMAC. On 16 December and 29 December 2009, AMAC transferred a total of $11,476,625, comprising $11.4m in principal and $76,625 in interest, to CHC in respect of Tranches 10 and 11 of the SOF.

24 To record these transactions, the following account entries were made in CHC’s General Journal on the instructions of Sharon as the church’s finance manager:

(a) an entry on 2 October 2009 describing the payment of $5.8m made to AMAC as “Investment–Special Opportunity Fund” under the accounts name “Investment”;

(b) an entry on 27 October 2009 describing the payment of $5.6m made to AMAC as “Special Opportunity Fund” under the accounts name “Investment”;

(c) an entry on 31 October 2009 describing the set-off amounting to $21.5m in favour of Xtron as “Redemption of Xtron Bonds”; and

(d) an entry on 6 November 2009 describing the payment of $15,238,936.31 made to Xtron as “Advance Rental with Xtron” under the accounts name “Prepayments”.

25 The net result of the round-tripping transactions was that the Xtron and Firna bonds were redeemed. Through the transactions, AMAC’s liability under Tranches 10 and 11 of the SOF was also discharged. As noted by the High Court at [46] of the MA Judgment, the liability owed by Xtron and Firna to CHC under the relevant bond subscription agreements was reconstituted into a liability on Xtron’s part to provide premises to CHC under the ARLA. This obligation was partially met as Xtron subsequently provided CHC with premises at the Singapore Expo for a period of time.

26 In January 2010, CHC acquired a stake in Suntec City. Subsequently, the ARLA was terminated.

27 In May 2010, the Commercial Affairs Department commenced investigations into the transactions. On 1 August 2010, CHC convened an EGM, where the EMs retrospectively approved CHC’s use of the BF to (a) subscribe to the Xtron bonds; (b) subscribe to the Firna bonds; and (c) pay the advance rental and security deposit to Xtron under the ARLA. The EMs also approved the continuation of the Crossover. Effectively, CHC was seeking to retrospectively ratify the transactions that had taken place and which were the subject of the investigations. However, as the courts below found, the EMs were misled as to the true substance of the transactions. At the EGM, it was falsely represented to the EMs that: (a) CHC subscribed to the Xtron bonds because they offered a good interest rate; (b) the Firna bond proceeds were intended as a commercial investment “to help Firna’s business” and Wahju had made an independent decision to use “part” of the funds to support the Crossover; and (c) the ARLA was entered into to provide sufficient funds to Xtron to bid for a property for CHC. The truth was that the Xtron bonds were entered into without any consideration as to their commercial viability for CHC, while the Firna bond proceeds were controlled entirely by Kong Hee and the other accused persons, and the decision to use the Firna bond proceeds to fund the Crossover was made by them. Likewise, the ARLA was not a genuine commercial agreement and its purpose was simply to facilitate the redemption of the bonds and the substitution of debts owed to CHC.

28 On 4 October 2010, consequent on the termination of the ARLA, Xtron repaid CHC a total of $40.5m. This sum comprised (a) $33,039,117.60 being the unutilised advance rental; (b) $7m being the full amount of the security deposit paid by CHC; and (c) $453,103.02 being the interest accrued from the date of termination of the ARLA until the date of payment. Although it is not exactly clear where Xtron obtained these funds, it appears that a number of loans were granted to the company by various individuals affiliated to the accused persons and CHC so that the repayment could be made to the church.

29 A total of 43 charges were brought against the six accused persons. As the High Court observed at [13] of the MA Judgment, these charges can be broadly characterised into three categories:

(a) Three “sham investment charges” brought against the accused persons except Sharon, arising from the use of the BF to purchase the Xtron and Firna bonds. These charges are for the offence of conspiring to commit CBT as an agent punishable under s 409 read with s 109 of the Penal Code. The first charge is under the 1985 revised edition of the Penal Code (ie, the Penal Code (Cap 224, 1985 Rev Ed)), and the second and third charges are under the 2008 (and current) revised edition of the same Act (ie, the Penal Code (Cap 224, 2008 Rev Ed)). The only difference between the two versions of the Penal Code in relation to s 409 concerns the maximum non-life imprisonment term for the offence. This has no bearing on the present application, which concerns only the elements of the s 409 offence and not the sentences imposed, and all remaining references to “the Penal Code” are to both revised editions of the Act.

(b) Three “round-tripping charges” against the accused persons, except Kong Hee and John Lam, relating to the use of the BF and the GF as part of the round-tripping transactions set out at [23] above. These charges are also for the offence of conspiring to commit CBT by an agent punishable under s 409 read with s 109 of the Penal Code.

(c) Four “account falsification charges” against the accused persons, except Kong Hee and John Lam, arising from the entries recorded in CHC’s General Journal set out at [24] above. These charges are for the offence of falsification of accounts under s 477A read with s 109 of the Penal Code.

30 The questions posed by the Applicant touched only on the convictions of the accused persons and not on the factors taken into account in sentencing them, as we will elaborate shortly (at [39] below). Hence we will focus on the parts of the decisions below that concern conviction.

31 The Judge found the accused persons guilty of all the charges against them. His decision on conviction is succinctly summarised by the High Court in the MA Judgment at [51]–[58]. The High Court allowed in part the appeals against conviction and sentence, with a partial dissent by Chan Seng Onn J. It is worth emphasising, for the reason explained at [6] above, that the High Court’s findings of fact could not be subjected to challenge in this application, in light of the nature and purpose of the application. Given that there was a split decision of the High Court in this case, the factual findings of the majority – which of course determined the outcome of the appeal – were therefore, for all intents and purposes, final and immutable. But, as will become apparent in our analysis of the questions posed by the Applicant, the points of difference between the majority and the minority were in fact not relevant to the present application.

32 The majority of the High Court, comprising Chao Hick Tin JA and Woo Bih Li J, agreed with the Judge that the following five elements had to be proved beyond a reasonable doubt in order to make out the sham investment and round-tripping charges for CBT (collectively, “the CBT charges”) (MA Judgment at [62]):

(a) the accused persons were entrusted with dominion over CHC’s funds;

(b) this entrustment was in the way of the accused persons’ business as agents;

(c) monies from CHC’s funds were misappropriated for various unauthorised purposes in pursuance of a conspiracy to misuse CHC’s funds;

(d) the accused persons abetted each other by engaging in the above conspiracy to misuse CHC’s funds; and

(e) the accused persons acted dishonestly in doing so.

33 In the present application, the questions posed by the Applicant in relation to the CBT charges (see [39] below) pertained only to the third element (ie, misappropriation) and the fifth element (ie, dishonesty). Hence, we will focus on the High Court’s findings on these issues.

34 In relation to the element of misappropriation, the question was whether there was “wrong use” of CHC’s funds. First, the majority rejected the argument that the Crossover could not be a “wrong use” because it was a “church purpose”. The majority agreed with the Judge’s finding that the BF, the monies of which were used to purchase the Xtron and Firna bonds, was a restricted fund meant for specific purposes. It could not simply be used for any “church purpose” and could be utilised only to pay for property and building related expenses or to invest in order to generate returns (MA Judgment at [124]–[135]). Since the transactions which led to the sham investment and round-tripping charges did not fall into the former category, the issue of “wrong use” ultimately turned on whether they constituted genuine investments. On this question, the majority (with Chan J concurring) agreed with the Judge that:

(a) Assessing the Xtron and Firna bonds on the basis of the substance (and not merely the form) of the transactions, they were not genuine investments for which the accused persons were authorised to use the funds in the BF (MA Judgment at [143] and [147]). The Xtron bonds were in effect a means for the accused persons to take out funds from the BF to use for the Crossover. The accused persons were not seriously concerned about whether, and, if so when, CHC would obtain financial return under the Xtron BSA (MA Judgment at [136]). Similarly, the Firna bonds were simply a source of funds for the Crossover and other purposes, and the accused persons were indifferent to the commercial viability or sensibility of the transactions from CHC’s perspective (MA Judgment at [155]).

(b) The round-tripping transactions were also not genuine investments and were nothing less than a perpetuation of fraud, or at the very least, a devious scheme to use the monies in the BF and the GF to create the appearance that Firna, AMAC and Xtron had fulfilled their obligations to CHC (MA Judgment at [161] and [170]).

35 The majority also considered that each accused person had engaged in a conspiracy to put CHC’s funds to wrong use, and was dishonest in doing so. This particular issue was dealt with as follows:

(a) First, it was held that the pertinent question in assessing dishonesty is whether the accused person intended to do an act that would cause wrongful gain or wrongful loss to another in circumstances where he knew that he was not legally entitled to do that act (MA Judgment at [184]). This ruling is at the heart of several of the questions posed by the Applicant, and we will return to it later.

(b) On the facts, the majority rejected the accused persons’ argument that they were not dishonest because they had been open with and relied on the advice of professionals. It agreed with the Judge that the accused persons had withheld and obscured the true relationship between them and Xtron and Firna as well as the real substance of the round-tripping transactions (MA Judgment at [193], [199] and [200]). It also found that all six accused persons had sufficiently engaged in a conspiracy to commit the CBT offences and had acted dishonestly as they all knew that they were not legally entitled to use the funds in the manner in which they did.

(c) Finally, the majority rejected the argument that the accused persons could not have been dishonest because they had acted in what they considered to be the best interests of CHC. This argument related to their motive rather than intention, and there was sufficient evidence that each of the accused persons possessed the requisite dishonest intention. They had acted despite knowing that the transactions were not “above-board” (MA Judgment at [312]–[314]).

Importantly, Chan J likewise concurred that the element of dishonesty was made out (MA Judgment at [437]). This meant that there was no division in the High Court as to the correctness of the Judge’s findings on the elements of misappropriation and dishonesty.

36 The High Court identified three elements to the account falsification charges (MA Judgment at [319]):

(a) the entries made in CHC’s accounts must have been false;

(b) the accused persons must have abetted each other by engaging in a conspiracy to make the false entries; and

(c) in engaging in the conspiracy, the accused persons must have been aware that the entries were false and possessed an intention to defraud.

37 The majority, again with Chan J concurring, held that all three elements were made out:

(a) First, looking to the substance and not merely the legal form of the transactions, the accounting entries were false as: (i) the two payments relating to Tranches 10 and 11 of the SOF were not genuine investments; (ii) the set-off to redeem the Xtron bonds amounted in substance to a writing-off of the bonds from CHC’s books; and (iii) the payment of $15,238,936.31 made to Xtron as “Advance Rental with Xtron” was not a genuine building-related expense but was to enable AMAC to return CHC the money disbursed into Tranches 10 and 11 of the SOF (MA Judgment at [323] and [331]).

(b) Second, the relevant accused persons (the Applicant, Ye Peng, Sharon and Serina) had participated in the plan to use the SOF and the ARLA to create the false impression that the Xtron and Firna bonds been redeemed using funds acquired from genuine commercial transactions. As it was necessary for the accounting entries to be recorded in CHC’s account in order to achieve this plan, they had abetted each other by engaging in a conspiracy to make the false entries (MA Judgment at [333] and [337]).

(c) Third, these four accused persons knew that the relevant transactions were not genuine commercial transactions. This was sufficient to prove that they had an intention to defraud in connection with the false entries (MA Judgment at [332]–[336]). Their argument that they had no intention to defraud as the auditors knew that CHC would be paying Xtron advance rental, and Xtron would then be redeeming the bonds by way of set-off, was also rejected. This disclosure was partial and the accused persons had hidden the true relationship between Xtron and CHC and the true nature of the payments under the ARLA from the auditors (MA Judgment at [338]–[340]).

38 While we were cognisant that the Applicant was a litigant-in-person, we nevertheless took the view that the questions that he sought to refer were formulated in an unsatisfactory way; they were broadly worded, overlapping and frequently lacked clarity and focus. In addition, there were no fewer than ten “sections” in his supporting affidavit, each comprising of multiple sub-questions that were expressed in a verbose and overly intricate manner. Indeed, we found that some questions were so unintelligible as to be incapable of being understood, let alone answered. In light of the fact that the Applicant was a litigant-in-person, we were minded to – as we were permitted to do (see, for instance, the decision of this court in Bachoo Mohan Singh v Public Prosecutor and other applications [2010] 1 SLR 966 at [86]) – reframe his questions so that they might be put at their highest and that no prejudice might result from the fact that the Applicant was not legally trained. We restated his ten “sections” as constituting the following ten questions:

(a) Question 1 – What is the meaning of “misappropriation”: does it refer to any “wrong use” of money or property which is unauthorised or does it require “the taking of someone’s money or property and using it for oneself”?

(b) Question 2 – In determining dishonesty, what constitutes “an intention to cause wrongful loss”? In particular, is there necessarily an intention to cause wrongful loss whenever an accused person has the intention to use the entrusted funds for an unauthorised purpose with the knowledge that he has no legal entitlement to do so, even if: (i) the property will be ultimately returned to the owner, (ii) the unauthorised use was for the purpose and for the benefit of the owner, and (iii) there may be a potential gain to the owner?

(c) Question 3 – The High Court used an objective test in characterising the relevant transactions and then wrongly inferred dishonesty from the fact that they could not be characterised as genuine investments, thereby making the offence of CBT one of strict liability. How does the court determine whether the use of the entrusted property was within the authorised aims for which it was entrusted to the accused person (eg, in this case, whether the Xtron and Firna bonds and the round-tripping transactions were genuine investments)? In particular, does the court characterise the relevant transaction objectively or subjectively based on what the accused person himself believed to be the nature of the transaction?

(d) Question 4 – When the entrusted funds are owned by a company or society, can the use of the funds to achieve the objectives of the company or society be said to be “wrongful”? Also, what is the impact of a retrospective ratification of the unauthorised transactions by the company or society, and can the use of the funds then still be considered “wrongful”?

(e) Question 5 – Is there a conflict of judicial authority between the High Court’s decision in the present case and two other cases: (i) Raffles Town Club Pte Ltd v Lim Eng Hock Peter and others (Tung Yu-Lien Margaret and others, third parties) [2010] SGHC 163 (“Raffles Town Club (HC)”) (together with the appellate decision in Raffles Town Club Pte Ltd v Lim Eng Hock Peter and others and other appeals [2013] 1 SLR 374 (“Raffles Town Club (CA)”)); and (ii) Periasamy s/o Sinnappan and another v Public Prosecutor [1996] 2 MLJ 557 (“Periasamy”)?

(f) Question 6 – Can an accused person be found to be dishonest even if he had an honest belief that his actions were proper and legal? In particular, is there a conflict in judicial authority between the High Court’s decision and the earlier authorities which have held that a person who honestly believes that he was legally entitled to carry on certain actions cannot be said to have acted dishonestly?

(g) Question 7 – In order to show that there was an intent to defraud in the context of the account falsification charges, must it be proved that there was intent to either make a gain or cause injury to another, or is it sufficient that the accused person intended to deceive? Does it matter that the person allegedly being deceived would not have acted differently even if he had known the true facts?

(h) Question 8 – Can an account entry be considered to be false if the purpose for which the funds were to be used by the payee was not disclosed in the accounts even though the entry reflected the existence of actual contractual obligations and liabilities?

(i) Question 9 – Can the round-tripping and account falsification charges be maintained if it is proved that the Applicant was not dishonest in initially using the BF for the Crossover (ie, if the sham investment charges are not made out)?

(j) Question 10 – The Applicant contended that the advance GST paid under the ARLA was prepayment of an unavoidable expense. In such circumstances, can the payment of the tax pursuant to the ARLA be considered “unlawful” and as causing a “wrongful loss”?

39 Two further observations can be made, each of which has already been alluded to. First, the questions that the Applicant posed related exclusively to his convictions under the charges against him and did not concern the sentence imposed by the High Court. Second, the questions touching on the CBT charges only pertained to the High Court’s findings on the elements of misappropriation and dishonesty.

40 Finally, we noted that the Applicant included multiple references in the application to s 9A of the Interpretation Act (Cap 1, 2002 Rev Ed) (“the IA”). Section 9A(1) of the IA states that in interpreting a provision of a written law, an interpretation that would promote the purpose or object underlying the written law should be preferred to an interpretation that does not promote that purpose or object. Section 9A(2) identifies the circumstances in which consideration may be given to extrinsic material in construing a statutory provision, including a situation where there is a need “to confirm that the meaning of the provision is the ordinary meaning conveyed by the text of the provision” (s 9A(2)(a)). It is trite that the courts have to apply s 9A of the IA in construing any legislation. So the references to the provision in the application were entirely superfluous and did not add anything of substance to the questions as framed above. Of course, it would have been open to the Applicant to argue, if leave had been granted, that the High Court’s rulings of law on the above questions were wrong and ought to be overruled by the Court of Appeal because they do not promote the purpose or object of the Penal Code; but that presupposed that these were questions of law of public interest which satisfied the threshold conditions for bringing a criminal reference. It is to these conditions which we now turn.

41 The law with respect to the granting of leave under s 397(1) of the CPC is uncontroversial and was recently affirmed by this court in Lee Siew Boon Winston v Public Prosecutor [2015] SGCA 67 at [6] as well as Huang Liping v Public Prosecutor [2016] 4 SLR 716 at [8]. Four conditions must be satisfied before leave can be granted:

(a) First, the reference to the Court of Appeal can be made only in relation to a criminal matter decided by the High Court in the exercise of its appellate or revisionary jurisdiction.

(b) Second, the reference must relate to a question of law and that question of law must be a question of law of public interest.

(c) Third, the question of law must have arisen from the case which was before the High Court.

(d) Fourth, the determination of that question of law by the High Court must have affected the outcome of the case.

42 In so far as the second condition is concerned, it is clear that there must be a question of law involved (as opposed to a mere question of fact). The former is necessarily normative in nature given that it would apply – generally or universally – to other (similar) situations. The latter, on the other hand, is necessarily confined or limited to the case at hand. As this court put it in Public Prosecutor v Teo Chu Ha [2014] 4 SLR 600 at [31]:

43 The following approach articulated by the Malaysian Federal Court in A Ragunathan v Pendakwa Raya [1982] 1 MLJ 139 (“Ragunathan”) at 141, which this court approved in Mohammad Faizal bin Sabtu and another v Public Prosecutor and another matter [2013] 2 SLR 141 at [19], is instructive:

44 It is not the function of the criminal courts to answer theoretical or abstract legal questions. That is the reason why the question of law to be referred has to arise from the case which was before the High Court, and the determination thereof must have affected the outcome of the case. In particular, the fourth condition set out at [41(d)] above requires the answer to that question of law to have been one of the grounds or bases upon which the High Court had decided the matter or issue before it: see Public Prosecutor v Li Weiming and others [2014] 2 SLR 393 (“Li Weiming”) at [20].

45 Before we turn to the substance of the questions raised by the Appellant, there is an important preliminary point of principle that should be considered. That point can be put as follows: what are the circumstances in which leave to bring a criminal reference will be granted when the Magistrate’s Appeal was heard by a specially convened coram of three Judges of the High Court?

46 In our judgment, when a three-Judge coram of the High Court has ruled, its decision should generally represent a final and authoritative determination of the issues arising from the case. Therefore no leave would (absent exceptional circumstances) be given for a further reference to the Court of Appeal.

47 This approach is justified because a three-Judge coram is a de facto Court of Appeal – comprising Justices of the Supreme Court, Judges of Appeal and perhaps even the Chief Justice – and is convened precisely to deal with important questions affecting the public interest which require detailed examination. Judith Prakash JA made the same point in a recent case, TUC v TUD [2017] SGHCF 15 (“TUC”), albeit in the context of an application for leave to appeal against a decision of a three-Judge coram of the Family Division of the High Court. The High Court coram in that case comprised the Chief Justice and two Judges of Appeal. As Prakash JA explained at [10], “[i]t was precisely because there were questions of general principle to be decided for the first time, on which a decision by a higher tribunal would be to the public advantage, that three members of the Court of Appeal sat in the High Court to hear this appeal” [emphasis added]. She proceeded to observe as follows (at [12]):

48 In our view, Prakash JA’s observations are entirely applicable to an application to bring a criminal reference from the appellate judgment of a three-Judge coram of the High Court. When a party to a criminal matter seeks to have the Court of Appeal reconsider a question which has already been determined by a three-Judge coram, this should only be allowed in exceptional situations. Otherwise, unnecessary duplication of efforts would result. More importantly, this would undermine the very reason why the three-Judge coram was specially convened in the first place.

49 The above analysis is, however, subject to an important caveat. That caveat stems from the unalterable fact that the three-Judge coram would be sitting as a bench of the High Court and therefore can only exercise the powers that the High Court has. As Wee Chong Jin CJ (sitting in the High Court with FA Chua J and AV Winslow J) noted in Mah Kah Yew v Public Prosecutor [1968-1970] SLR(R) 851 at [1], a coram comprising three Judges hearing a Magistrate’s Appeal is “the High Court and its powers, although it consists of three judges, are no greater and no less than the powers of a single judge when both are exercising the same appellate jurisdiction”. In other words, even if the three-Judge coram may be a de facto Court of Appeal, it is not one de jure. Hence, while the High Court can depart from previous High Court precedents, it does not have the powers, unlike the Court of Appeal, to (a) overturn or overrule other decisions of the High Court; or (b) depart from decisions of the Court of Appeal. These are established principles of the doctrine of stare decisis.

50 Therefore, the central question, in the context of determining whether leave should be granted to bring a criminal reference arising from the decision of a three-Judge coram hearing a Magistrate’s Appeal, is whether the question of law of public interest posed is one that only the Court of Appeal can properly deal with by virtue of the position and powers that it has as the apex court of the land. In our view, when a three-Judge coram has been convened in the High Court to hear the Magistrate’s Appeal, this additional consideration ought to be borne in mind on top of the four conditions set out above at [41]. This additional hurdle would only be surmounted in exceptional cases such as where there is a need to reconsider and possibly overturn an established line of High Court authority or depart from a decision of the Court of Appeal.

51 Bearing these principles in mind, we now turn to the questions raised by the Applicant in the present proceedings.

52 We found that even a cursory reading of the questions raised by the Applicant revealed that the questions he sought to refer were either (impermissible) attempts to reopen and/or change established principles of law in order to escape personal liability for his actions, or were simply questions of fact which could not, by any stretch of the imagination, be characterised as questions of law. Hence, the second condition set out above at [41] was not satisfied. Some of the questions also pertained to issues on which the High Court did not make a decision and were therefore purely hypothetical in nature. Those questions therefore (also) failed the third and fourth conditions.

53 In the circumstances, it was not in the least surprising that the questions themselves were (as we alluded to above) phrased in an awkward and over-elaborate manner – this was because what the Applicant did in this application was (in the main) to “dress up” challenges to established principles of law and findings of fact as novel questions of public interest arising from the High Court’s decision. This was precisely what we stated at the outset of this judgment could not be done – the instituting of what was, in substance and effect, a further (and backdoor) appeal on the substantive merits, seeking to controvert findings of fact that were made by the trial court and that had been reviewed by the High Court on appeal. In our view, this was nothing more than a blatant abuse of the process of the court. Whatever the Applicant’s personal dissatisfaction with the result, there had to be finality in the judicial process once that process had run its course – a process which, in this case, had been marked by careful and objective analysis of both the facts and the law by a trial court as well as an appellate court. In so far as the Applicant sought to reopen and/or change well-settled principles of law, this too was simply not the purpose, and could not be done by way, of a criminal reference (see the extract from Ragunathan at [43] above).

54 More importantly – and for the reasons explained at [46]–[50] above – we were mindful that the Magistrate’s Appeal in this case was heard by a specially convened coram of three Judges of the High Court. While we do not rule out the possibility that leave to bring a criminal reference may be granted in such a situation if there are exceptional circumstances, we were satisfied that this was far from such a case. None of the questions which arose from this application was dealt with by the three-Judge coram in a manner that required any overturning or overruling of a line of High Court authority, or a departure from a decision of the Court of Appeal. The High Court’s findings in relation to these questions (where they actually arose from the High Court’s decision) were unanimous. This was not a case in which only the Court of Appeal, by virtue of its powers and position, could deal with the issues raised. Indeed, we were of the view that the High Court (as well as the Judge) had provided a careful and comprehensive consideration of the questions of law that arose in the appeal, including the questions of law that the Applicant raised. On this basis alone, we would have dismissed the application.

55 In any event, we set out our specific findings on each of the ten questions raised by the Applicant. We were satisfied that none of these questions came close to meeting the threshold conditions for leave to be granted to bring a criminal reference set out at [41] above.

56 We begin with Question 1, which concerned the meaning of “misappropriation” in the offence of CBT. It involved a settled question of law and therefore failed the second condition for leave to be granted (ie, it was not a question of law of public interest) (see [43] above).

57 It is a settled principle of law that the actus reus, or physical element, of misappropriation in the offence of CBT is “to set apart or assign to the wrong person or wrong use” (see Tan Tze Chye v Public Prosecutor [1997] 1 SLR(R) 876 at [37], applied at [95] of the Conviction GD; and Phang Wah and others v Public Prosecutor [2012] 1 SLR 646 at [48], applied at [123] of the MA Judgment). Whether a particular use of the entrusted funds amounts to a “wrong use” is a question of fact to be answered by reference to the scope of the authority and consent given by the owner in entrusting the funds to the accused person.

58 In the present case, both the Judge and the High Court were unpersuaded that the BF could be used for every so-called “church purpose”, because it was a restricted fund intended only for specific purposes (Conviction GD at [125]; MA Judgment at [125]). The Applicant in effect sought to overturn this finding by arguing that the focus of the courts below on “wrong use” was erroneous. He submitted that the “dictionary definition” of misappropriation should have been applied instead. This definition, according to him, requires “the taking of someone’s money or property and using it for oneself” and does not include a situation where the funds were used for the owner’s purpose and benefit. In addition, he sought to draw a common thread through the various property offences in the Penal Code, which, in his submission, all require the taking away of property for oneself.

59 We found that these submissions did not give rise to any question of law of public interest. First, they were unsupported by any legal authority and there was no indication that the settled and longstanding common law definition of “misappropriation” applied by the courts below has given rise to any controversy or confusion. This is unsurprising since whether or not an act amounts to “misappropriation” is a fact-specific enquiry which invariably turns on the circumstances of each case.

60 Second, the “dictionary definition” put forward by the Applicant was flawed as that definition focuses on the intention of the accused person and therefore conflates the objective physical element of “misappropriation” with the subjective mens rea, or mental element (which is that the misappropriation must have been “done dishonestly” (see the Conviction GD at [95])). We will examine the element of dishonesty as part of our analysis of Question 2. For present purposes, it suffices to note that even if we turn our attention to whether the misappropriation was “dishonest”, the definition of “dishonesty” in s 24 of the Penal Code – which refers to an intention of causing wrongful gain or wrongful loss – makes it clear that there is no requirement that the accused person must have taken the property to “use it for himself”. An intention to cause wrongful loss, without more, is sufficient. In addition, the common thread which the Applicant sought to draw through the property offences in the Penal Code simply does not exist. As pointed out by the Prosecution, even in relation to the simple offence of theft under s 378 of the Penal Code, there is no requirement that the property must have been taken for the offender’s own use. The same is true for the offences of cheating and extortion under ss 383 and 415 of the Penal Code, respectively.

61 Thus the Applicant’s submissions on Question 1 were unsupported by authority and were undermined by the language and provisions of the Penal Code. For these reasons, we found that the Applicant had utterly failed to provide any reason for the Court of Appeal to reconsider the settled legal principles governing the element of “misappropriation”.

62 Question 2, which concerned the element of “dishonesty” in the offence of CBT, appeared on a superficial examination to be worthy of some examination. However, upon a closer reading of the High Court’s decision, it was clear that the question arose from a misapprehension and mischaracterisation of the High Court’s analysis of dishonesty. It therefore failed the fourth condition, ie, that the determination of the question by the High Court must have affected the outcome of the case. As the High Court’s analysis of this issue was based on established legal principles, Question 2 also failed the second condition (ie, it was not a question of law of public interest).

63 The question was essentially based on the argument made below by the accused persons that even if they had used CHC’s funds for unauthorised purposes with the knowledge that they had no legal entitlement to do, this did not necessarily mean that they had acted with the intention to cause wrongful loss (ie, dishonesty) (see the MA Judgment at [173]). The Applicant similarly argued in the present application that the High Court erred by equating an intention to cause wrongful loss, on one hand, with the use of the funds for an unauthorised purpose with the knowledge that there was no legal entitlement to do so, on the other.

64 As noted above, this submission was premised on a misapprehension and mischaracterisation of the High Court’s analysis of the requirement of dishonesty. What the High Court found was that “where an accused knows that an action is unauthorised but nonetheless proceeds to execute it voluntarily, this would strongly support a finding of dishonesty” [emphasis added] (MA Judgment at [179]). It did not rule that an intention to cause wrongful loss will necessarily be present whenever an accused person intended to use the entrusted funds for an unauthorised purpose with the knowledge that he had no legal entitlement to do so. This analysis was in line with the submission of the Prosecution that knowledge of unauthorised use itself does not satisfy the mens rea requirement of a CBT charge, although it will be a substantial component of proof of mens rea (see the MA Judgment at [175]).

65 The High Court’s treatment of illustration (d) of s 405 of the Penal Code (“illus (d)”) also makes it clear that the majority (with whom Chan J agreed) was cognisant that there may be cases where an intention to cause wrongful loss will not be present even if the accused person knows that an action is unauthorised but nonetheless proceeds to execute it. Section 405 defines the offence of CBT, and the relevant illustrations to the provision are as follows:

66 While the Judge, at first instance, confined illus (d) to “a situation where a person is authorised to make a specified investment for purposes of financial profit, and instead makes a different investment honestly believing that this would bring in greater financial profit” (Conviction GD at [189]), the majority of the High Court preferred a broader approach. It held that the scenario in illus (d) is one where the agent “did not intend to wrongfully deprive the principal of the principal’s funds” (MA Judgment at [183]). In other words, illus (d) is an example of a situation where there is no intention to cause wrongful loss (ie, no dishonesty) despite knowledge of unauthorised use. The majority identified the various factors which indicate that there is no such dishonest intention in illus (d), including the fact that the funds were still invested by the agent for the principal’s financial benefit, with any financial gains intended to accrue to the principal, and the comparison made between what the agent was instructed to do, and what he eventually did based on the honest belief that what he did would be more to his principal’s benefit than what his principal had originally instructed him to do. The majority also noted that illus (d) states that the agent acted “in good faith” and “not dishonestly”, which suggests that he “did not believe that his disobedience of his principal’s direction was wrongful in the circumstances” (MA Judgment at [183]). The High Court then concluded as follows (MA Judgment at [184]):

67 It is thus clear that, contrary to the Applicant’s submission, the High Court did not simply equate an intention to cause wrongful loss (ie, dishonesty) with knowledge of unauthorised use. Its finding was that such knowledge strongly supports, but will not necessarily lead to, a finding of dishonesty. Thus the premise of Question 2 simply falls away.

68 On the facts, the High Court found that the accused persons did have the intention to cause wrongful loss to CHC in procuring the relevant transactions (MA Judgment at [313]–[314]). Focusing on the Applicant, the High Court was satisfied that he:

(a) knew that the Xtron bonds were not genuine investments and did not really at any time consider their purchase as a commercially sensible investment for CHC, which investment would generate any financial returns for the church (MA Judgment at [264]–[269]);

(b) knew that the Firna bonds were not a commercial investment but a temporary means of obtaining funds from CHC for the Crossover (MA Judgment at [273]–[275]);

(c) had participated in conveying misleading information about the bonds to CHC’s auditors, legal counsel, and a member of CHC’s investment committee (MA Judgment at [270] and [276]); and

(d) was the main architect of the round-tripping transactions, which he knew were not legally above-board (MA Judgment at [278] and [281]).

In such circumstances, there was ample evidence that the Applicant had acted dishonestly with an intention to cause wrongful loss to CHC.

69 In so far as the Applicant contended that there cannot be an intention to cause wrongful loss where (a) the property will ultimately be returned to the owner, (b) the unauthorised use was for the purpose and for the benefit of the owner, and (c) there may be a potential gain to the owner, these arguments had already been dealt with by the High Court. In doing so, the High Court considered the language of the Penal Code and applied established principles of law. Hence these submissions did not amount to any questions of law of public interest.

70 As the High Court noted at [177] of the MA Judgment, it is well-established that the requirement of “loss” may be made out even if the loss is only temporary (see Explanation 1 to s 403 of the Penal Code). Thus it does not matter whether the property will be ultimately returned to its owner. For the same reason, the fact that there may be a potential gain cannot exonerate an accused person who acts intending to wrongfully keep out or wrongfully deprive the owner of his property (see s 23 of the Penal Code). In any event, this question was entirely hypothetical given the High Court’s finding that the transfers of funds to Xtron, Firna and AMAC were not genuine investments because the accused persons did not genuinely expect financial gain from those transfers. Finally, it is hornbook law that motive is distinct from intention. Thus, even if the unauthorised use by the accused person was for admirable motives and was for what he considered to be “for the purpose and for the benefit” of the owner, this does not preclude a finding that there was dishonesty if the circumstances indicate that there was nevertheless an intention to cause wrongful loss, such as in the present case (see MA Judgment at [313]). Again, as illus (d) indicates, there may be situations where an accused person with knowledge of unauthorised use may be found to have acted in good faith and without an intention to cause wrongful loss. But this was far from such a case given the findings of fact that the High Court made in relation to the Applicant (see [68] above). These findings indicated that the Applicant had acted dishonestly and without regard to the legality of his actions even if he, along with the other accused persons, had acted in what they considered to be the best interests of CHC.

How does the court determine whether the use of the entrusted property was within the authorised aims for which it was entrusted to the accused person (eg, in this case, whether the Xtron and Firna bonds and the round-tripping transactions were genuine investments)? In particular, does the court characterise the relevant transaction objectively or subjectively based on what the accused person himself believed to be the nature of the transaction?

71 Question 3 was, once again, based on a mischaracterisation of the High Court’s analysis of the element of dishonesty for the CBT charges. It therefore failed the fourth condition for leave to refer a question to the Court of Appeal. It also failed the second condition because it could be answered by applying the established principles on the distinction between the actus reus and mens rea of the offence of CBT.

72 The question rested on the premise that the High Court used a set of independent criteria to determine whether the relevant transactions were “investments”, and then wrongly inferred subjective dishonesty from the fact that the transactions could not be characterised as genuine investments from an objective viewpoint. The Applicant submitted that the court should have instead characterised the relevant transactions based on what the accused persons themselves believed would qualify as “investments”. He also contends that the High Court’s approach made the offence of CBT “one of strict liability”.

73 This question was flawed essentially for the same reason as Question 1 – it conflated the objective physical element of “misappropriation” with the subjective fault element that the misappropriation must have been “done dishonestly” (see [60] above). The element of “misappropriation”, as noted above at [56], turns on whether there was “wrong use”. And the court’s determination as to whether there was “wrong use” is an objective inquiry that must be undertaken by reference to the scope of the authority and consent given by the owner in entrusting the funds to the accused person. What criteria a court ought to use in deciding whether a particular transaction was authorised or consented to inescapably depends on the court’s findings as to the ambit and limits of that authority. In addition, the fact that “wrong use” is to be determined objectively does not mean that the offence of CBT is one of strict liability. The court will still have to separately determine if there was dishonesty – which is the mens rea requirement for the offence of CBT.

74 It is clear from the judgments of the Judge and the High Court that the courts below applied these established principles correctly, and were careful to maintain an analytical distinction between the actus reus and the mens rea of the offence of CBT. The Judge expressly noted at [193] of the Conviction GD that “[w]hether or not the accused persons acted ‘dishonestly’ is of course a subjective enquiry that must be answered by reference to their actual state of mind at the time of the alleged offences” [emphasis added]. He went on to emphasise that “the mere fact that the purported investments in the Xtron and Firna bonds and Tranches 10 and 11 of the SOF were not actually investments … is not determinative of that subjective question of what the accused persons believed” [emphasis added]. This was the same approach adopted by the High Court (see the MA Judgment at [62]).

75 This point is put beyond any doubt when one observes that the courts’ finding that the transfers of funds to Xtron, Firna and AMAC did not constitute genuine investments did not in fact furnish the basis for their subsequent finding that the accused persons were dishonest. That finding of dishonesty was inferred from the fact that the accused persons had omitted to disclose, or actively obscured or hid, important facts material to the transactions from the EMs of CHC, the CHC Board, the auditors and/or the lawyers (see the Conviction GD at [194]; and the MA Judgment at [172] and [206]). Thus the courts’ decisions that, on the one hand, the monies from the BF were not transferred for the purposes of genuine investments and, on the other hand, that the accused persons acted dishonestly, were reached on entirely separate factual bases and for entirely separate reasons. Question 3 therefore arose from a fundamental misreading or misunderstanding of the reasoning of the courts below. It did not merit further consideration.

76 There were two parts to Question 4. The first part, which concerned whether certain uses amounted to “wrong use”, failed the second condition as it concerned a factual finding that did not give rise to any question of law. The second part of Question 4 did not meet the fourth condition as ratification was not an issue which affected the outcome of the case.

77 The first part of Question 4 was based on the argument made below that the use of the funds for the Crossover could not be said to be “wrongful” as the Crossover was a “church purpose” supported by CHC’s congregation. It therefore overlapped with Question 1. For the reasons given at [56]–[60] above, this argument did not give rise to any question of law of public interest. It was rejected by the courts below based on the fact-specific finding that the BF was a restricted fund meant for specific purposes and could not be used for any “church purpose”. Indeed, whether the use of a company’s or society’s funds amounts to “wrong use” was at best a question of mixed fact and law, and at worst a pure question of fact. The legal aspect to the question – concerning the proper definition of “wrong use” – is well-settled, for the reasons set out at [57] above. The remaining aspect was simply a factual inquiry as to the scope of the authority given by the owner of the funds and the use that the accused person put those funds to. That factual inquiry will obviously turn on the particular circumstances of each case. It is not a question that can be answered through a misdirected and ill-defined inquiry as to whether the use of the funds was in line with “the objectives” of the company or society.

78 The second part of Question 4 concerned the impact, if any, of retrospective ratification of unauthorised transactions. This point was also raised by the Applicant below, but the High Court took the view that there was no need to address it specifically in order to determine the outcome of the appeals. In other words, this question did not satisfy the fourth condition as the point was not determined and therefore did not affect the outcome of the case. In any event, we found that the only authorities cited by the Applicant for the broad proposition that retrospective ratification may operate as a defence to a CBT charge were readily distinguishable. The first case was Raffles Town Club (HC), where the High Court found that the former directors of the club had breached their directors’ duties by charging expenses to the club which were not for its benefit or reasonably incidental to its business. The court held that the directors were not liable because the charging of these expenses had been ratified by the members of the company (Raffles Town Club (HC) at [182]; upheld in Raffles Town Club (CA) at [30]–[31]). Crucially, that was a case concerning a director’s civil liability to his or her own company and it did not deal with the entirely separate question of whether an accused person’s criminal liability to the state can be affected by retrospective ratification. On this basis alone, it was clear that the case did not assist the Applicant.

79 The second authority was Periasamy, where the Malaysian Court of Appeal held that valid consent could be given subsequent to the impugned use or disposal. It is not clear if this dicta forms part of Singapore law, and there is at least one local authority decided by Wee CJ, Yeow Fook Yuen v R [1965] 2 MLJ 80 (“Yeow Fook Yuen”), which suggests that a criminal act cannot be decriminalised by subsequent ratification, particularly after the commencement of police investigations (at 82–83).

80 However, even assuming that Periasamy is a part of our law, a closer examination of the decision revealed that it did not support the Applicant’s position. The Malaysian Court of Appeal held that retrospective ratification will only operate as a defence “if the facts emerge to show that there was true consent in its legal sense” [emphasis added] and does not apply if “the ratification is but a cloak to cover up a dishonest act” (at 587). On the present facts, the High Court, in agreement with the Judge, found that at the EGM held in August 2010, Kong Hee and Ye Peng had misled the EMs as to the true nature of the Xtron and Firna bonds, as well as the payments under the ARLA, in an effort to obtain ex post facto ratification of the transactions (see [27] above; and the MA Judgment at [245] and [262]). This was an attempt to cover up the accused persons’ dishonest actions after police investigations had commenced, just as in the case of Yeow Fook Yuen. In other words, there was no true consent on the part of the EMs because Kong Hee and Ye Peng simply had not provided them with a full and faithful account of the facts, even after their misdemeanours had been uncovered and investigations had commenced into the transactions (see [27] above). The absence of this factor in Periasamy was one of the principal bases on which the Malaysian High Court distinguished Yeow Fook Yuen (at 587). Thus, the Applicant’s reliance on Periasamy was doomed to fail.

81 Another key point of distinction between Periasamy and the present case was that in the former case, the Malaysian High Court’s decision turned on the fact that the letter of offer (which was the document setting out the terms of the bank’s agreement to the loan in question and which also governed the scope of entrustment to the relevant accused person) expressly provided that the terms could be altered or amended or even withdrawn at the bank’s discretion and was accordingly “not cast in stone”. In other words, the bank expressly left open the possibility that it could retrospectively change the terms of its loan at its discretion, even for the purposes of authorising in an ex post facto manner any violation of the terms of the entrustment to the accused person. There can be little dispute that no such broad and permissive terms of reference existed in the present case, which simply involved a situation where unauthorised uses were made of the church’s funds, attempts were made to conceal the wrongdoing, and a belated effort to obtain approval of the conduct sought (albeit on the back of further misrepresentations and deliberate omissions).

82 Consequently, this was not an appropriate case for the Court of Appeal to examine the impact of retrospective ratification on the offence of CBT. It was not a legal issue which was determined by the High Court; nor were the authorities cited by the Applicant applicable on the facts of this case.

Whether there is a conflict of judicial authority between the High Court’s decision in the present case and the decisions of Raffles Town Club (HC) and Periasamy?

83 Question 5 was not, in fact, a freestanding question. The alleged conflict of judicial authority between, on the one hand, the decisions of the Singapore High Court (in Raffles Town Club (HC)) and the Malaysian Court of Appeal (in Periasamy) and, on the other hand, the decision of the High Court in the present case, concerned issues that fell within the scope of Questions 2 and 4 above – ie, the definition of “an intention to cause wrongful loss” (which constitutes dishonesty) and the effect of retrospective consent and ratification on the offence of CBT. For the reasons given in our analysis of Questions 2 and 4 as set out above, we likewise find that Question 5 did not amount to a question of law of public interest. We will briefly elaborate.

84 First, the Applicant submitted that the High Court’s analysis of the element of dishonesty contradicted that of the Malaysian Court of Appeal in Periasamy. In particular, the Applicant relied on Periasamy for the uncontroversial proposition that the offence of CBT is not an offence of strict liability and requires, besides the doing of an unauthorised act, a dishonest intention (ie, an intention to cause wrongful loss or gain). As is made evident in our analysis of Questions 2 and 3 above, the High Court’s reasoning was entirely in line with this proposition (see [63]–[68] and [74]–[75] above).

85 Second, the Applicant contended that the High Court’s disregard of the impact of consent and ratification on the CBT charges was contrary to both Raffles Town Club (HC) and Periasamy. This was simply not the case as we have made clear in our examination of Question 4 (see [78]–[81] above).

86 Question 6 was a purely hypothetical question and therefore failed the third and fourth conditions.

87 One of the key findings of fact made by the Judge, which was upheld by the High Court, was that the accused persons, including the Applicant, did not have an honest belief that the transactions were proper and legal. The Judge made this finding in clear and unequivocal terms at [477] of the Conviction GD:

88 Specifically, in relation to the Applicant, the Judge found that, although he trusted his “own thinking” about the propriety and legality of the transactions, this was due to the “extravagant overconfidence that characterised his conduct and mindset”. This mindset led him to see no need to rely on lawyers for legal advice and emboldened him to conceive of various “dishonest and expedient means to an end, knowing that the BF would be used for an unauthorised purpose” (Conviction GD at [386]). Put simply, the Applicant did not honestly believe that he was legally entitled to act as he did. Hence the authorities cited by the Applicant, in support of the principle that a person who honestly believes that he was legally entitled to carry on certain actions cannot be said to have acted dishonestly, were not relevant to the present case.

89 Question 7 was, in substance, a factual challenge to the High Court’s finding that there was an intent to defraud in respect of the account falsification charges. It failed the second condition as it did not give rise to any question of law.

90 The Applicant’s argument can be briefly summarised. The requirement of an intent to defraud under s 477A of the Penal Code necessitates an intent to deceive, and through that deception, to gain a benefit or cause an injury. The High Court and the Judge erred by focusing purely on deception without considering if there was an intention to cause a benefit or an injury. The Applicant also argued that the auditor Sim Guan Seng (“Sim”) testified that he would have raised questions about Xtron’s ability to repay its debts to CHC even had he known that the advance rental to be paid by CHC to Xtron under the ARLA was really for the purpose of allowing Xtron to redeem its bonds. Thus neither benefit to the Applicant nor injury to the auditors accrued as a result of the false entry in CHC’s books.

91 Although the Applicant attempted to cast the issue as a question of law, there was, in reality, little controversy about the legal aspect of the issue. In Li Weiming, the Court of Appeal held at [85] that the mens rea requirement of an intent to defraud in s 477A of the Penal Code is “an intent to defraud directed at an object, which may be proven by adducing evidence that supports a finding or inference of fact of an intention to either defraud persons generally or a named individual or entity” [emphasis in original]. This definition was cited and applied by the High Court (see the MA Judgment at [332]). The focus of the court’s analysis in Li Weiming was on whether the Prosecution was required to prove that the accused’s intent to defraud was directed at particular persons, or if it sufficed for the Prosecution to show a general intent to defraud. There was no dispute in Li Weiming about the existence of the requirement that the accused must have intended, through his deception, to cause injury or create an advantage; indeed, the court in Li Weiming accepted that this was part of the mens rea of an intent to defraud, when it held at [84] that “[i]t is clearly possible that a person may carry out an act with an intent to defraud by practicing a deception with the aim of causing an injury, loss or detriment or obtaining an advantage, even if he is indifferent as to who the object of his fraudulent intent is” [emphasis added].

92 In the present case, it is clear from the findings of the Judge, which were upheld by the High Court, that there was such an intention to cause injury and loss, as well as an intention to obtain an advantage. The account falsification offences were inextricably tied to the round-tripping transactions in that it was necessary for false accounting entries to be recorded in CHC’s books in order to perpetrate the false impression, generated through the round-tripping transactions, that the Xtron and Firna bonds had true value and had been redeemed using funds acquired from genuine commercial transactions (see the Conviction GD at [448] and [452]; and the MA Judgment at [333]). It therefore hardly needs saying that in so far as the round-tripping transactions were meant to allow the accused persons to conceal from the auditors the true nature and purpose of the Xtron and Firna bond purchases, so was the falsification of accounts intended to facilitate the accused persons’ achievement of these goals. This was the advantage or benefit that the accused persons sought to obtain through their falsification of accounts (and, conversely, the injury to the auditors who therefore failed to detect the round-tripping).

93 The second aspect of the Applicant’s argument – ie, that the auditors would still have entertained doubts about the ability of Xtron to pay its debts to CHC even if they had known that the purpose of the payment of advance rental under the ARLA was to enable Xtron to redeem the bonds, and that therefore no injury had been caused to the auditors or benefit gained by the accused persons flowing from the insertion of the false entry – was plainly an argument concerning the facts of the case. It was not a question of law, and certainly not one of public interest. In any event, the focus of the fault element of an intent to defraud is on the subjective mental state of the accused person; thus it is difficult to see how the reaction or response of the person allegedly being deceived, and the question of whether he might have acted differently even if he had known the true facts, is relevant to determining the existence of an intent to defraud.

94 Even leaving those difficulties aside, the Applicant’s argument simply does not accord with the evidence. As the High Court described (see the MA Judgment at [339]), what Sim told the court was that he had not been informed that the real purpose of the ARLA was to facilitate the redemption of the Xtron bonds, and that if he had known that the true purpose of the ARLA was to facilitate the redemption of the Xtron bonds then he would have had to consider more carefully whether Xtron really had the ability to redeem the bonds. The High Court surmised at [340] that “[i]t is therefore apparent that Sim was not privy to the full facts concerning the ARLA [and it was] his evidence that he would have inquired further if he knew that the whole purpose of the ARLA was to facilitate the bond redemption”. In other words, Sim did not say, as the Applicant suggests, that “even with the replacement of the bonds by the ARLA … he would still have questioned the ability of Xtron to repay the underlying debts owing”. What Sim had said was quite the opposite – it was because he did not know that the purpose of the ARLA was to enable redemption of the Xtron bonds that he did not further question Xtron’s ability to redeem the bonds. The Applicant’s question was therefore based on a misreading of the evidence and was, for this reason, divorced from the facts of the case and was entirely hypothetical in nature.

95 Question 8 was another question of fact, thinly disguised as one of law. It failed the second condition.

96 The Applicant took issue with the High Court’s finding (see the MA Judgment at [322]) that evidence of normal accounting practice was relevant in determining what the correct accounting entry should be, and that normal accounting practice, as represented by the Financial Reporting Standard 24 (2006) (Related Party Disclosures) issued by the Council on Corporate Disclosure and Governance, required that “[i]n considering each possible related party relationship, attention is directed to the substance of the relationship and not merely the legal form”. The accounting entries recorded Tranches 10 and 11 of the SOF as an “Investment”, the payment of $15.2m to Xtron as “Advance Rental with Xtron”; and a purported set-off of advance rental amounting to $21.5m for “Redemption of Xtron Bonds”. But these were not, “in truth and in substance”, respectively investments, advance rental or a set-off. Therefore, the High Court reasoned that the accounting entries had to be false.

97 In his affidavit, the Applicant referred to a document called “The Conceptual Framework For Financial Reporting”, issued in February 2011 by the Accounting Standards Council, and the Financial Reporting Standard 39 (2013) (Financial Instruments: Recognition and Measurement) (“FRS39”) which stated that financial assets or liabilities should be recognised in an entity’s statement of financial position when the entity becomes a party to the contractual provisions, and that the entity should “derecognise” such assets when the contractual rights to the cash flow from the financial assets expire. Relying on the FRS39, the Applicant submitted that the entering of accounting entries ought to be based on the existence of contractual obligations, and not on an assessment of whether there had been sufficient disclosure.

98 It must be noted that there was no dispute between the parties on what might be regarded as a clear question of law – that is, whether normal accounting practice was relevant in determining whether an entry was a “false entry” within the meaning of s 477A of the Penal Code. On the contrary, it was accepted by all the parties (naturally including the Applicant) that normal accounting practice was relevant and should be examined. Hence, all that the Applicant was really disputing was the High Court’s finding as to what normal accounting practice demanded. Did normal accounting practice require that only entries that reflected the substance of the transaction be included in the accounts, or did it suffice that the entries reflected what was stated in contractual documents? The High Court, based on various sources of evidence put before it, decided that normal accounting practice looked to the substance rather than the legal form of the underlying transaction in determining whether an accounting entry was false (see the MA Judgment at [321]–[322]). The High Court’s findings and the Applicant’s argument on this matter were respectively based on different statements of accounting practice. The issue was accordingly one of fact and evidence rather than law, and there was consequently no basis for it to be reconsidered by the Court of Appeal.

Can the round-tripping and account falsification charges be upheld if it is proved that the Applicant was not dishonest in initially using the BF for the Crossover (ie, if the sham investment charges are not made out)?

99 Question 9 was plainly fact-specific and did not give rise to any question of law. It therefore did not satisfy the second condition. It was also entirely hypothetical and failed the third and fourth conditions.

100 The Applicant queried whether, given the High Court’s finding that the accused persons’ conspiracy to round-trip monies and falsify CHC’s accounting entries was for the purpose of covering up the misappropriation reflected in the sham investment charges, the account falsification charges could still stand in the event that the sham investment charges could not be proven.

101 It was difficult to see how this was a question of law, much less one of public interest. Whether the accused persons’ convictions for the round-tripping and account falsification charges could still stand if their convictions for the sham investment charges were set aside was evidently a question which turned on the circumstances of the case. It depended on whether the elements of the ss 409 (for the round-tripping) and 477A (for the account falsification) charges remained satisfied even if the sham investment charges were not made out, based on the specific findings of fact made in relation to the accused persons. More importantly, this question was entirely hypothetical given the High Court’s decision that the accused persons had committed CBT in relation to the sham investment charges. There was therefore completely no reason for the High Court to engage in the wholly academic exercise of considering whether the round-tripping and account falsification charges could stand if the sham investment charges did not. In the circumstances, Question 9 was entirely moot and was not the proper subject for a criminal reference.

102 Question 10 was clearly a question of fact and therefore failed the second condition.

103 As described at [23(d)] above, as part of the round-tripping transactions, a sum of $15,238,936.61 was transferred from the BF to Xtron on 6 November 2009, of which $12m was stated to be part payment of the advance rental due to Xtron under the ARLA and the remaining $3,238,936.61 allegedly comprising GST. These sums formed the subject of the sixth charge against the Applicant (as part of the round-tripping charges). The Applicant submitted that the GST payment was “made in pursuance of a contractual rental agreement between CHC and Xtron” and therefore the payment of GST was an “unavoidable expense” that was done “for the benefit of the church”.

104 This was manifestly a factual rather than a legal question. Whether or not the GST payment was an “unavoidable expense” was to be determined on the facts of the case. On this basis alone, the question could not be regarded as a proper subject for a criminal reference. In any event, we found that Question 10 was nothing more than an indirect means of reopening the question as to whether the ARLA was really a building-related agreement and whether the expenses associated with it were therefore really building-related expenses. The High Court soundly rejected such a characterisation of the ARLA (see the MA Judgment at [165]–[168]), finding that it was not a a genuine commercial agreement because (a) there was little evidence as to how the rental rate and rental period were arrived at; (b) the amount to be transferred under the ARLA was based purely on the sums needed to redeem the Xtron and Firna bonds; and (c) the terms of the ARLA were not commercially justifiable.

105 Crucially, the High Court then proceeded to consider CHC’s payment of GST pursuant to the ARLA. It found that there was no basis for the GST payment; indeed it was “egregious that the appellants were willing to allow CHC to incur a GST expense of $3.2m on the ARLA for the purpose of conveying the impression that the ARLA was a genuine agreement” (MA Judgment at [169]). In its subsequent consideration of the appropriate sentences to be imposed, the High Court found at [407] that the GST was incurred “in order to create the false impression that the ARLA was a genuine agreement for advance rental” and took the view that this was an aggravating factor, because the $3,238,936.61 would have represented actual loss to CHC had the ARLA not later been rescinded. For this reason, we found that it simply did not lie in the mouth of the Applicant (or indeed any of the accused persons) to argue that the GST payment – which was simply a means by which the accused persons sought to create a false appearance of the nature of the ARLA – was an “unavoidable expense”.

106 For the above reasons, we dismissed the application in its entirety. As the courts below had found, this was a case where the accused persons clandestinely applied church donations, which were collected from CHC’s members and were designated for clear and specific purposes, to advance an aim that was entirely outside the scope of the authorised uses of the donations. They did so through the purchase of bonds, superficially branded as investments, so as to obscure what they were really doing from CHC’s lawyers, auditors and the church’s own members. They then exacerbated the situation by concocting a series of fraudulent transactions aimed at removing these bonds from CHC’s accounts, in order to mask their dishonest conduct.

107 Both the Judge and the High Court had no doubt that the Applicant, as the primary financial architect of these transactions, was heavily involved and indeed instrumental in this illegal enterprise. The present application, which was ill-considered and wholly unmeritorious, provided no basis for the Court of Appeal to re-examine the detailed findings which led to the Applicant’s conviction.