1 In the court below, the appellant pleaded guilty to 15 charges of abetment by engaging in a conspiracy with two others to corruptly obtain gratification under s 6(a) read with s 29(a) of the Prevention of Corruption Act (Cap 241, 1993 Rev Ed) (the “PCA”) and four charges of corruptly giving gratification under s 6(b) of the PCA. A total of 40 remaining charges (of which 31 were abetment by conspiracy charges while nine were s 6(b) charges) were taken into consideration for the purposes of sentencing. The District Judge (the “DJ”) imposed a global sentence of 17 months and three weeks’ imprisonment. In this appeal, the appellant challenges the DJ’s decision on sentence on the ground that it is manifestly excessive.

2 In arriving at her decision, the DJ agreed with the parties’ submissions that the sentencing framework set out by the High Court in Takaaki Masui v Public Prosecutor and another appeal and other matters [2021] 4 SLR 160 (“Masui (HC)”) was binding on her and she accordingly relied on it (see Public Prosecutor v Goh Ngak Eng [2021] SGDC 285 (“GD”) at [17]). Masui (HC) was the first case to articulate a sentencing framework for private sector corruption offences under ss 6(a) and (b) of the PCA.

3 After the DJ delivered her decision in August 2021, the Court of Appeal delivered its judgment in December 2021 on ancillary matters arising from Masui (HC) in Public Prosecutor v Takaaki Masui and another and other matters [2022] 1 SLR 1033 (“Masui (CA)”). Although the sentencing framework set out in Masui (HC) was not an issue in Masui (CA), the Court of Appeal declined to endorse it, having observed that the sentencing framework in Masui (HC) “is as complex as it is likely to be of little assistance to sentencing courts” and cautioned that “excessively complex or technical sentencing frameworks are prone to cause confusion and uncertainty, which are the very anthesis of a sound sentencing framework” (see Masui (CA) at [15]).

4 In this appeal, the appellant submits that the Masui (HC) sentencing framework is the appropriate framework to be applied, although the DJ’s application of it had resulted in an “inflated imprisonment sentence”. The Prosecution, on the other hand, urges this court to develop an alternate sentencing framework for offences under ss 5, 6(a) and 6(b) of the PCA, having regard to the Court of Appeal’s observations in Masui (CA). As it stands, the sentencing framework in Masui (HC) is binding on the lower courts. Given that the present case before us also involves private sector corruption, we agree with the Prosecution that this appeal presents an opportunity for this court to consider whether a sentencing framework should be developed for private sector corruption offences under ss 6(a) and (b) of the PCA. Further, if so, it should be considered whether the sentencing framework in Masui (HC) is the appropriate framework and if not, what form it should take. We therefore invited the parties, as well as Mr Elton Tan (“Mr Tan”), who was appointed under the Supreme Court’s young amicus curiae scheme, to address us on these issues, which are central to our decision on whether the appellant’s sentence is manifestly excessive.

5 Having carefully considered the parties’ submissions, we decline to adopt the existing sentencing framework in Masui (HC) for the reasons which we will explain below. Instead, we have developed a sentencing framework for private sector corruption offences under ss 6(a) and (b) of the PCA that is modelled after the two-stage, five-step framework in Logachev Vladislav v Public Prosecutor [2018] 4 SLR 609 (“Logachev”). In developing the sentencing framework, we are of the view that it should not be extended to s 5 of the PCA or to cases of public sector corruption such as those involving public servants and public bodies. This judgment sets out the revised sentencing framework.

6 Also, having carefully considered the materials on the record, we are of the view that, in the proceedings below, the parties had proceeded on a mistaken characterisation of the harm caused by the appellant’s offences. The DJ was led into error by the position taken by the parties and similarly arrived at her decision on sentence on that erroneous basis. Once the harm arising from the appellant’s offences is properly appreciated, it becomes clear that the individual sentences imposed on the appellant (and in turn, his global sentence) were manifestly inadequate. We therefore consider it appropriate to enhance the appellant’s sentences.

7 The appellant is a 55-year-old male Singaporean. He was a director of Megamarine Services Pte Ltd (“Megamarine”), a company in the business of manufacturing and repairing air winches and general trading of equipment. As mentioned earlier, the Prosecution proceeded with 19 charges against the appellant. Fifteen of those were charges of abetment by engaging in a conspiracy with one Rajavikraman s/o Jayapandian (“Raj”) and another Alvin Lim Wee Lun (“Lim”) to corruptly obtain gratification under s 6(a) read with s 29(a) of the PCA (the “Conspiracy Charges”). At the time of those offences, Raj was a project director at Rotating Offshore Solutions Pte Ltd (“Rotating”), which was in the business of building and designing generators and compressors, while Lim was a yard manager in the Facilities Department at Keppel FELS (“KFELS”) shipyard. The other four proceeded charges were of corruptly giving gratification under s 6(b) of the PCA.

8 Sometime in late-2014, the appellant was approached by Raj, who said that he would be able to refer jobs from KFELS to vendors. Raj explained that he knew someone in KFELS called “Alvin” (referring to Lim), who was in a position to recommend to whom the jobs were to be awarded. According to Raj, Lim wanted 15% of the invoice value of the jobs (before GST) in order to award those jobs, and so the invoices submitted by the vendors to KFELS would have to be marked up. Raj and the appellant then decided that they would seek vendors for jobs in KFELS and would ask for their invoices to be marked up by more than 15%. The mark-up was to be shared between Lim, Raj and the appellant.

9 According to KFELS’s procurement process, it was the Purchasing Department’s responsibility to source for quotations from contractors. However, during the period when Lim oversaw the Facilities Department, he instructed his staff to source for quotations before raising the Electronic Purchase Requisition (“EPR”). The details of the contractor and the price quoted would be included in the description of the EPR. It was a common practice in KFELS shipyard at that time for end-users such as the Facilities Department to source for quotations instead of the Purchasing Department, and the orders raised were still approved even though the Purchasing Manager knew that this should not have been allowed. This administrative lapse in the procurement process meant that Lim effectively decided which contractors would be invited to quote for jobs at KFELS and subsequently, which of these contractors would be recommended to be awarded with jobs.

10 Among the Conspiracy Charges under s 6(a) read with s 29(a) of the PCA:

(a) One charge involved the corrupt obtaining of gratification from one U Keh Choon (“Keh Choon”), the director of Titan Offshore Equipment Pte Ltd (“Titan”), which was wholly owned by him. Titan was in the business of manufacturing and servicing marine equipment, including capstans. Sometime in late-2014, the appellant contacted Keh Choon, asking if he was interested to supply two capstans to KFELS. After Keh Choon indicated his interest and responded with an indicative price, the appellant, on Raj’s advice, responded with a revised marked-up price, which was to be submitted in the quotation to KFELS. Keh Choon agreed that if Titan were to be awarded the job, the difference in these two amounts (the indicative price and the marked-up price) would be paid as commission to the appellant and Raj. Keh Choon was also informed that 15% of that sum would be given to someone in KFELS (Lim). From 2015 to 2017, KFELS purchased two new capstans from Titan and engaged Titan for its servicing and repair services for six existing capstans at the KFELS shipyard. For every job that Titan invoiced KFELS, the appellant would receive a copy of Titan’s marked-up invoice to KFELS, and he would then calculate the amount of mark up and use Megamarine to invoice Titan for the commission (though on four occasions the appellant used another entity, 3W Logistics Services (“3W”) to invoice Titan: see [12(b)] below). In all, Keh Choon made a total of seven commission payments to Megamarine from 2015 to 2017 from Titan’s accounts, amounting to a sum of $196,661.72. Of these payments, one of them, which involved a sum of $107,000 obtained in or around January 2015, was the subject of the amended first charge.

(b) Two charges involved the corrupt obtaining of gratification from one Goh Sheng Li, Stanley (“Goh”), the director of Spectrama Marine & Industrial Supplies Pte Ltd (“Spectrama”), which was in the business of supplying, repairing, servicing and testing industrial lifting equipment. The appellant was previously a long-term customer of Spectrama. Sometime in late-2014, the appellant arranged an introductory meeting for Raj to meet Goh. During that meeting, the appellant and Raj informed Goh about a potential job for KFELS. Goh expressed his interest to take on the job. At a subsequent meeting, the appellant and Raj requested a percentage of the invoiced amounts of the jobs as commission for referring the jobs to Spectrama. It was also made known to Goh that someone in KFELS (Lim) would receive a share of the commissions to ensure that Spectrama was given the business. Goh agreed and he marked up the price in the quotations submitted to KFELS to factor in the commission. From 2015 to 2017, Spectrama performed a total of 49 jobs for the servicing and repair of chain blocks and lever blocks for KFELS. For every job that Spectrama invoiced KFELS, Goh would similarly provide a copy of the marked-up invoice to the appellant, who would then calculate the amount of mark-up and use Megamarine to invoice Spectrama for the commission. In all, Goh made a total of 21 payments to Megamarine from 2015 to 2017 from Spectrama’s accounts, amounting to a sum of $190,917.01. Of these payments, two of them, which involved sums of $21,835.41 and $28,784.36 that were respectively obtained on or around 19 February 2016 and 18 August 2016, were the subject of the amended 16th and 21st charges.

(c) Twelve charges involved the corrupt obtaining of gratification from one Fatkullah Bin Tiap (“Fatkullah”), the founder and managing director of Growa (F.E.) Pte Ltd (“Growa”), which is in the business of distributing hoists and cranes and providing services such as the repair, servicing and inspections of hoists and cranes. Sometime in late-2014, Raj asked the appellant if he had a contact for a contractor that could provide crane inspection services for KFELS. The appellant then contacted Fatkullah and told him about a job opportunity for Growa to provide crane inspection services for KFELS. Fatkullah was informed that he had to mark up Growa’s quoted prices in order to pay commissions to the appellant and Raj in return for referring Growa for the job at KFELS, and that a portion of these commissions would be paid to someone in KFELS (Lim) so as to ensure that Growa was given the business. Fatkullah agreed to this arrangement. Between 10 January 2015 and 25 October 2016, Growa performed a total of 22 crane inspection jobs for KFELS. Again, for every job that Growa invoiced KFELS, a copy of the marked-up invoice would be provided to the appellant, who would then calculate the amount of the mark-up and use Megamarine to invoice Growa for the commission (though on six occasions the appellant used 3W to invoice Growa: see [12(b)] below). In all, Fatkullah made a total of 18 commission payments to Megamarine from 2015 to 2016 from Growa’s accounts, amounting to a sum of $492,274.90. Of these payments, 12 of them, involving sums between $22,778.16 and $46,170.50 obtained between 30 March 2016 and 28 November 2016, were the subject of the amended 32nd, 33rd, 35th, 36th, 37th, 38th, 39th, 40th, 41st, 42nd, 43rd and 44th charges.

11 After receiving each commission payment from Titan, Spectrama and Growa, the appellant would compute Lim’s, Raj’s and his own share of the commissions. For each commission payment, the appellant would first deduct Lim’s share and another portion for Megamarine’s corporate tax payment, before splitting the remaining amount between himself and Raj evenly. The total amount paid by Keh Choon, Goh and Fatkullah as corrupt gratification (in respect of all the abetment by conspiracy charges, ie, both the Conspiracy Charges and the non-proceeded charges that were taken into consideration for sentencing) was $879,853.63, of which Lim received $293,822.39, Raj received $191,115.89 and the appellant received $191,115.89. The balance sum was paid by Megamarine to the Inland Revenue Authority of Singapore as corporate tax.

12 As for the charges involving the corrupt giving of gratification under s 6(b) of the PCA (the “Non-Conspiracy Charges”):

(a) One charge involved the appellant corruptly giving $3,000 to Raj for furthering the interests of Megamarine with Rotating. Sometime in March 2014, Raj informed the appellant that the purchasing manager of Rotating would request a quotation for the supply of hardware. Raj informed the appellant that, if Megamarine clinched the deal, the appellant would have to give Raj some commission for referring and recommending the works to Megamarine. The appellant agreed; Megamarine was awarded the job and the appellant paid Raj $3,000. This was the subject of the amended 47th charge.

(b) Three charges involved the appellant corruptly giving sums of $2,997.12, $2,308.32 and $2,564.48 to one Ong Tun Chai (“Ong”), the manager of 3W Logistics Services (“3W”), for preparing fictitious invoices on 3W’s letterhead. 3W, a sole proprietorship registered under the name of Ong’s mother, was in the business of transporting cleaning chemicals and laundry for hospitals, prisons and hotels. Sometime in late-2016, the appellant asked Ong to help prepare 3W invoices and delivery orders, which were used to invoice Titan and Growa for the commission that they had agreed to pay for obtaining contracts with KFELS on those occasions when the appellant did not use Megamarine directly to invoice Titan and Growa. In return, the appellant promised to pay Ong 8% of the invoiced amount. Ong agreed and prepared ten such invoices and their corresponding delivery orders. However, the descriptions on these invoices were fictitious, as none of the jobs stated therein were carried out. For these invoices and delivery orders, the appellant paid Ong a total sum of $15,092.80 over ten occasions in 2016. Three of those payments respectively were the subject of the 51st, 52nd and 53rd charges.

13 The DJ applied the two-stage, five-step sentencing framework in Masui (HC) ([2] above) in arriving at her decision. We briefly set out the sentencing framework in Masui (HC) to the extent that it is necessary to place the DJ’s decision in context.

14 At the first stage of the framework in Masui (HC), the court would consider the severity of the offence committed by having regard to all the offence-specific factors present on the facts of the case and arrives at an indicative sentence to reflect it. Steps one to three come under the first stage of the framework, and they are as follows:

(a) At step one, the court would identify and assess the relevant offence-specific factors present on the facts of the case (see Masui (HC) at [238] and [278]).

(b) At step two, the court would broadly determine where the specific level of harm caused by the offender, and the specific level of his culpability, lay along the respective spectrum of harm and culpability, which throws up an indicative sentencing range (see Masui (HC) at [254] and [278]-[279]). This would involve the use of a “modified harm-culpability matrix” and a “contour matrix” (see Masui (HC) at [262]-[277]), the details of which we need not go into for present purposes.

(c) At step three, the court would, in the exercise of its sentencing discretion, make an assessment as to where exactly the offender’s offence lies within the area of the “modified harm-culpability matrix” and the “contour matrix” and thereafter select the most appropriate indicative starting sentence from the range of indicative starting sentences thrown up by the framework (see Masui (HC) at [254] and [281]). The court in Masui (HC) observed that this step provided “specific guidance to a sentencing court as to how the indicative starting sentence is derived” (see Masui (HC) at [283]).

15 At the second stage, the court would consider all the offender-specific factors to derive a sentence for each individual charge. Steps four to five come under the second stage of the framework, and they are as follows:

(a) At step four, the court would consider the offender-specific factors which did not directly relate to the commission of the offence (see Masui (HC) at [284]).

(b) Finally, at step five, the court would have regard to the totality principle in determining the final global sentence of the offender (see Masui (HC) at [286]).

16 Applying step one of the framework in Masui (HC), the DJ agreed with the harm and culpability factors identified by the Prosecution (see GD at [45]):

17 As for the factors going towards harm, the DJ agreed with the parties that the corrupt transactions in the Conspiracy Charges posed “potential detriment to KFELS” because Lim, who had the power to select and recommend vendors for KFELS, was corrupted into ensuring that the jobs were only awarded to vendors who had given or promised him money in return (see GD at [47]). The DJ considered that the offences thus compromised a fair and safe procurement process at KFELS, and an unsuitable vendor “could lead to disastrous consequences given the nature of its business” (see GD at [47]). The DJ assessed the case as one of slight harm given that there was only one harm factor (potential detriment to KFELS). However, the DJ was of the view that, since KFELS was in a strategic industry, the potential harm could be significant and therefore she assessed it to be on the upper end of slight harm (see GD at [48]).

18 On the factors going towards culpability, the appellant accepted the factors (a) to (c) as identified by the Prosecution but disputed factors (d) and (e). On factor (d), the DJ observed that, since the appellant had pleaded guilty to the charges relating to Ong and admitted unreservedly to the Statement of Facts, which stated that he had enlisted Ong’s help to issue fictitious invoices to Titan and Growa for jobs which were not performed, he was making use of the falsified invoices issued by 3W to avoid detection and cloak the corrupt transaction (see GD at [56]). On factor (e), the DJ found that, given the appellant’s admission in the Statement of Facts that he had agreed to participate in the conspiracy because he needed money and that he had secured the sizeable bribes by seizing the “business opportunity”, the appellant had clearly committed the offences for financial gain and profit (see GD at [58]). The DJ therefore assessed culpability to be on the upper end of medium culpability (see GD at [59]).

19 For the purposes of step three, the DJ, in the exercise of her sentencing discretion, considered it unnecessary to identify a discrete indicative starting sentence at that juncture (see GD at [64]). Accordingly, she arrived at a range of indicative starting sentences of between seven to ten months’ imprisonment and a fine of up to $100,000 per charge based on the inputs of harm and culpability levels identified at step two (see GD at [61]).

20 However, the DJ decided that additional fines were not warranted in this case. She noted that fines were discretionary under s 6 of the PCA, with the consideration being whether there was a need to disgorge or confiscate any ascertainable profits which the appellant might have made from his illegal behaviour (see GD at [75]). Here, the illegal profits of $191,115.89 which the appellant made from his offences were already the subject of a penalty order under s 13 of the PCA, and the appellant had voluntarily surrendered that amount to the Corrupt Practices Investigation Bureau (“CPIB”). There was thus no necessity to impose an additional fine to disgorge or confiscate the appellant’s illegal profits, and the imprisonment terms imposed in the case would serve as adequate deterrence and punishment (see GD at [75]-[78]).

21 At step four of the framework, the DJ found the following offender-specific factors to be relevant:

22 The DJ agreed with the Prosecution that the main aggravating factor was the 40 charges taken into consideration for the purpose of sentencing. The amount of gratification involved in those 40 charges was $324,787.04. The DJ considered that an uplift in the sentence was warranted, given that both the number of charges taken into consideration as well as the bribe amounts involved in those charges were significant (see GD at [81]-[82]).

23 The DJ also found that the significant mitigating factors in this case were the appellant’s full cooperation with the CPIB and his early plea of guilt. The DJ accepted that the appellant had demonstrated genuine remorse by coming clean on his wrongdoings, rendering every assistance he could by furnishing the evidence in his possession and expressing his willingness to be a Prosecution witness. The DJ viewed the appellant’s voluntary and proactive disgorgement of the bribes as being consistent with his remorse, notwithstanding that it would have been the subject of a penalty order under s 13 of the PCA and that it came about only after the appellant had been charged (see GD at [84]). The DJ observed that while the appellant had confessed to the CPIB about offences which were not originally part of the investigations, that did not make the appellant a “whistleblower” for mitigation purposes. That was because the appellant “owned up” only after he had been summoned by the CPIB for investigations (see GD at [85]-[87]). As the offender-specific mitigating factors outweighed the aggravating ones in the case, the DJ adjusted the indicative starting sentence of seven to ten months’ imprisonment per charge downwards to a final sentencing range of four to nine months’ imprisonment per charge (see GD at [90]).

24 As for those Non-Conspiracy Charges relating to Ong (see [12(b)] above), the appellant accepted that imprisonment terms were warranted as those offences were connected to the scheme that was the subject of the Conspiracy Charges (see GD at [97]). In respect of those charges, the DJ agreed with the appellant that a sentence of two weeks’ imprisonment per charge would be fair (see GD at [100]). As for the Non-Conspiracy Charge relating to Raj (see [12(a)] above), the appellant submitted that a fine would be appropriate because it had arisen from a different factual matrix (see GD at [97]). The DJ disagreed with the appellant and was of the view that the custodial threshold had been crossed for this charge, given the appellant’s overall criminality (see GD at [99]). Accordingly, she concluded that a sentence of one week’s imprisonment for the Non-Conspiracy Charge relating to Raj was appropriate (see GD at [99]).

25 After adjusting for the offender-specific factors at step four of the framework, the DJ arrived at the following final sentences (see GD at [101]):

26 As for step five of the framework in Masui (HC), the DJ ordered five sentences to run consecutively to reflect the totality of the appellant’s corrupt criminal conduct: one sentence from each vendor (Titan, Spectrama and Growa) in respect of the Conspiracy Charges, one sentence from the Non-Conspiracy Charges relating to Ong and 3W, and the sentence imposed for the Non-Conspiracy Charge relating to Raj (see GD at [105]). This resulted in a global sentence of 17 months and three weeks’ imprisonment (see GD at [106]). She further considered that the sentences imposed in the present case were in line with the precedents tendered by the parties, in particular Public Prosecutor v Geow Chwee Hiam [2016] SGDC 139 (which sentence was affirmed on appeal in HC/MA 19/2016/01) and Public Prosecutor v Li Chaoyun and another [2020] SGDC 215 (which sentences were varied on appeal in HC/MA 9042/2020/01 and HC/MA 9043/2020/01) (see GD at [111]).

27 The appellant contends that the sentence which the DJ imposed was manifestly excessive. In the main, he argues that the DJ erred in finding that this was a case of slight harm/medium culpability. The appellant says that this was a result of the DJ erroneously finding that KFELS was in a strategic industry, which resulted in her concluding that the potential harm caused by the offending conduct was significant. The appellant also contends that the DJ failed to give sufficient weight to the fact that this was a “victimless crime” and that any danger of harm arising from his offending conduct by the hiring of an unsuitable vendor was merely hypothetical.

28 In response, the Prosecution submits that the DJ’s assessment of the case was correct, and that she correctly concluded that KFELS was in a strategic industry and that any potential harm to KFELS was a harm-related aggravating factor. In fact, the Prosecution submits that the present case ought to have been categorised as one of moderate harm/medium culpability. On this point, the Prosecution relies on a few other offence-specific factors which were not canvassed in the proceedings below. However, the Prosecution did not file a cross-appeal and at the hearing before us, it confirmed that it was not seeking to disturb the DJ’s decision on sentence and would leave the matter to the court. We will return to this point later (see [129] below).

29 The parties including Mr Tan agree that this court should develop a sentencing framework. The Prosecution and Mr Tan agree that such a framework should be modelled after the two-stage, five-step framework in Logachev ([5] above), although they have different views about its design. Specifically, they differ on some of the offence-specific factors which are to be included under step one of the framework. They also differ on some of the indicative starting sentence ranges in the sentencing matrix under step two of the framework. On the other hand, Mr Rajwin Singh Sandhu (“Mr Singh”), counsel for the appellant, stated in his written submissions that such a framework can be modelled after the framework in Masui (HC) ([2] above), although he did not pursue that point before us during oral arguments.

30 As for the scope of the proposed sentencing framework, the Prosecution submits that it should be extended to apply to offences under s 5 of the PCA. The Prosecution also takes the position that any such sentencing framework can be extended to public sector corruption offences that do not involve “a contract or a proposal for a contract with the Government or any department thereof or with any public body or a subcontract to execute any work comprised in such a contract”, ie, public sector corruption offences under ss 6(a) and (b) to which s 7 of the PCA does not apply. On the other hand, both the appellant and Mr Tan agree that it should be limited to offences under ss 6(a) and (b) of the PCA and to cases of private sector corruption.

31 As we had indicated to counsel during the hearing before us, having considered the written submissions, a sentencing framework should be developed for private sector corruption offences under ss 6(a) and (b) of the PCA. In our judgment, a sentencing framework, which can be developed taking into consideration the substantial body of jurisprudence for offences under ss 6(a) and (b) of the PCA, will provide guidance to sentencing courts and in particular the lower courts, the Prosecution and the Defence in approaching sentencing in a broadly consistent manner, with due regard to the salient factors. Also, given that the need for deterrence has resulted in a recent upward trend in custodial sentences for serious private sector corruption offences, sentences imposed in similar or analogous cases from several years ago may no longer constitute appropriate reference points. The consistency provided by a sentencing framework will assist in consolidating the approach going forward. Although the ways in which private sector corruption can manifest itself are diverse (see Public Prosecutor v Syed Mostafa Romel [2015] 3 SLR 1166 (“Romel”) at [26]), that does not, in and of itself, preclude the adoption of a sentencing framework provided that the court can develop a methodology that is workable, and which assists sentencing courts in arriving at outcomes in a “broadly consistent way” (see Public Prosecutor v Wong Chee Meng and another appeal [2020] 5 SLR 807 (“Wong Chee Meng”) at [57]).

32 Thus, the issues which remain before us are as follows:

(a) First, should the existing sentencing framework in Masui (HC) be followed, or should the proposed sentencing framework be based on the two-stage five-step framework in Logachev?

(b) Second, should the proposed sentencing framework:

(i) extend to private sector corruption offences under s 5 of the PCA; and

(ii) extend beyond cases of private sector corruption to also include cases of public sector corruption?

(c) Third, if the proposed sentencing framework is to be based on the two-stage, five-step framework in Logachev:

(i) what are the offence-specific factors that should be included under step one of that framework; and

(ii) what indicative sentencing ranges should be provided in the sentencing matrix under step two of that framework?

(d) Fourth, is there is any merit in the appellant’s contention that the sentences imposed by the DJ were manifestly excessive and do they, in any case, warrant reconsideration?

33 In our judgment, the sentencing framework in Masui (HC) ([2] above) should not be followed and the proposed sentencing framework should be based on the two-stage, five-step framework in Logachev ([5] above), which the High Court in Wong Chee Meng has already adopted as a sentencing framework for the aggravated offence of participating in a corrupt transaction with agents under s 6 read with s 7 of the PCA. We will first state our reasons for not following Masui (HC) before turning to the Logachev framework and explain why we consider it an appropriate sentencing framework for offences under ss 6(a) and (b) of the PCA.

34  In our view, the sentencing framework in Masui (HC) is excessively complex and technical, making it prone to confusion and uncertainty and thus rendering it of limited assistance to sentencing courts (see Masui (CA) ([3] above) at [15]). More importantly, the reasons which the judge in Masui (HC) (the “Judge”) thought justified the complexity of that framework are, with respect, unfounded. Let us explain.

35 Earlier, in referring to the DJ’s decision, we have set out briefly the five steps of the sentencing framework in Masui (HC) (see [14]-[15] above). The complexity of the framework in Masui (HC) lay in the use, at steps two and three, of a “modified harm-culpability matrix” and a “contour matrix” in which “every single point in the matrix” corresponded to the specific harm and culpability levels associated with the offence (see Masui (HC) at [254]-[281] and [283]). The Judge considered it desirable that “specific guidance” is provided to a sentencing court as to how an indicative starting sentence is derived (see Masui (HC) at [283]).

36 However, the Judge’s approach is founded on various erroneous assumptions. The Judge considered it necessary that “every single point in the matrix” correspond to a specific indicative starting sentence because of what he described as the “continuity principle” and the “single point principle”. The “continuity principle” requires that (see Masui (HC) at [148]-[151]):

37 With respect, we disagree. Although it is correct as a matter of principle that a sentencing court should be able to utilise the full sentencing spectrum prescribed by Parliament for the particular offence, that does not mean that the indicative starting sentences prescribed by the sentencing framework must also progress smoothly and proportionately with the severity of the offence. The indicative starting sentence is more than just a function of the severity of the offending conduct, and it depends on a range of factors, such as where the bulk of the offences occur within the spectrum of offending conduct and the relevant sentencing considerations engaged by the offending conduct.

38 On the other hand, the “single point principle” was a requirement that (see Masui (HC) at [155] and [171]):

39 We respectfully disagree with the above for two reasons. First, the “single point principle”, as conceived by the Judge, is founded on the erroneous premise that there is a simple and direct linear relationship between the severity of offending conduct (measured in terms of its harm and culpability levels) and the length of the indicative starting sentence (see, eg, Public Prosecutor v Raveen Balakrishnan [2018] 5 SLR 799 (“Raveen Balakrishnan”) at [77]). As we explained earlier (at [37]), the severity of offending conduct is but one of the various factors that the sentencing court takes into account in determining what the indicative starting sentence should be.

40 Second, we are doubtful if it is at all possible for a sentencing court to undertake the exercise of identifying the specific harm/culpability levels associated with the offending conduct in each case. Every sentence reflects a complex amalgam of numerous and various factors and imponderables and requires the very careful evaluation by the court of matters such as public interest, the nature of the offence and the circumstances in which it had been committed (see Dinesh Singh Bhatia s/o Amarjeet Singh v Public Prosecutor [2005] 3 SLR(R) 1 at [24]). We therefore agree with Mr Tan’s submission that the factors relevant in assessing the severity of the offence are inherently incapable of being weighed or evaluated with exactness. A sentencing court can readily identify the relevant harm-related or culpability-related factors that are engaged by the facts, but it will be difficult (if at all possible) to pinpoint the exact levels of harm and culpability that were caused, since the weight or significance of each of these factors cannot be determined with exactitude in each case.

41 Indeed, the Judge himself recognised that the harm and culpability levels associated with the offending conduct in each case can seldom be specifically determined. As he said in Masui (HC) (at [254]):

42 Thus, the Judge accepted, more often than not, a sentencing court can only identify the harm and culpability levels in each case as a “range”, which in turn will throw up a range of indicative starting sentences for the court to consider (see Masui (HC) at [254]). It is then for the sentencing court to exercise its discretion to select the most appropriate indicative starting sentence from within that range (see Masui (HC) at [254]). That much is unobjectionable. However, the Judge went a step further and considered it necessary that guidance be furnished to a sentencing court as to how that discretion is exercised. This required the sentencing court to make an assessment under step three of the framework as to where exactly the offending conduct lies on the continuums of harm and culpability, and then interpolate with the assistance of points and lines plotted on the “modified harm-culpability matrix” and “contour matrix” (see [35] above) to arrive at the appropriate indicative starting sentence (see Masui (HC) at [280]-[281]).

43 With respect, we disagree with this approach. The identification of an indicative starting sentence within the applicable indicative sentencing range is one of granulating the case with due appreciation of the harm- and culpability-related factors engaged by the offending conduct (see Logachev at [79]). We reiterate our earlier observation that the applicable indicative starting sentence is more than just a function of the severity of the offending conduct (see [37] above). It is therefore not a mathematical exercise. It is also not the objective of a sentencing framework to produce a mathematically perfect graph that identifies a precise point for the sentencing court to arrive at in each case. They are meant to only guide the court towards the appropriate sentence in each case using a methodology that is broadly consistent (see Mohd Akebal s/o Ghulam Jilani v Public Prosecutor and another appeal [2020] 1 SLR 266 at [20(b)]). An approach aimed at achieving mathematical precision is antithetical to the exercise of sentencing, which is largely a matter of judicial discretion and requires a balanced judgment and assessment of myriad considerations (see Angliss Singapore Pte Ltd v Public Prosecutor [2006] 4 SLR(R) 653 at [13]). Requiring a court to engage in the sentencing exercise against the backdrop of specific guidance and in the aim of achieving mathematical precision will have the untoward effect of fettering the court’s discretion.

44 Therefore, with respect to the Judge, the reasons of principle which he considered justified the complexity of the sentencing framework in Masui (HC), and which rendered it excessively complex and unworkable for sentencing courts to apply, are founded on a false premise. We therefore do not follow it.

45 Given the above, all that remains is for this court to consider how the proposed sentencing framework for ss 6(a) and (b) offences under the PCA can be modelled after the two-stage, five-step framework in Logachev ([5] above). The Logachev framework separates the sentencing exercise into its two underlying component parts, corresponding to the first and second stages of that framework. At the first stage, the court arrives at an indicative starting point sentence for the offender which reflects the intrinsic seriousness of the offending act. It involves the following three steps (see Logachev at [76]-[79]):

(a) At step one, the court identifies, by reference to factors specific to the particular offence under consideration, (i) the level of harm caused by the offence and (ii) the level of the offender’s culpability. Harm is a measure of the injury caused to society by the commission of the offence, while culpability is a measure of the degree of relative blameworthiness disclosed by the offender’s actions and is assessed chiefly in relation to the extent and manner of the offender’s involvement in the criminal act. The harm caused by the offence may be categorised as either “slight”, “moderate” or “severe”; while the offender’s culpability may be categorised as either “low”, “medium” or “high”.

(b) At step two, the court identifies the applicable indicative sentencing range, by reference to the level of harm caused by the offence (in terms of slight, moderate or severe) and the level of the offender’s culpability (in terms of low, medium or high).

(c) At step three, the court identifies the appropriate starting point within the indicative sentencing range that has been identified pursuant to step two. In doing so, the court has regard to the offence-specific factors and considers the harm and culpability levels associated with the offending conduct. Although this step will engage the same offence-specific factors as those considered at the first step, it is not an instance of double-counting any factors. Rather, it is an exercise of granulating the case to arrive at a sense of what the starting point in that case should be, for sentencing purposes.

46 At the second stage, the court makes adjustments to the starting point sentence identified under the first stage and arrives at a sentence that reflects the personal circumstances of the offender, by taking into account the relevant aggravating and/or mitigating circumstances unique to the offender and considering if the overall sentence arrived at is proportionate and consistent with the overall criminality of the offender (see Logachev at [80]–[81]). It involves the following two steps:

(a) Step four requires the court to make adjustments to the identified starting point as may be necessary to take into account factors personal to the offender’s particular circumstances. These offender-specific factors include aggravating factors (like offences taken into consideration for sentencing purposes, relevant antecedents and lack of remorse) and mitigating factors (like a guilty plea, voluntary restitution in the case of property-related offences and cooperation with the authorities), although the foregoing is not an exhaustive list. It is possible that the court may find at this stage that an adjustment of the sentence beyond the indicative sentencing range identified pursuant to the second step may be necessary.

(b) In a case where an offender has been convicted of multiple charges, step five requires the court to consider if further adjustments should be made to the sentence for the individual charges to take into account the totality principle. As the Court of Appeal recently emphasised in Public Prosecutor v Azlin bte Arujunah and other appeals [2022] SGCA 52 (“Azlin”) (at [199(b)] and [205]), the totality principle not only has a limiting function, in guarding against an excessive overall sentence arising from the sentences ordered to be run consecutively under s 307(1) of the Criminal Procedure Code 2010 (2020 Rev Ed), but also a boosting effect on individual sentences where they would otherwise result in a manifestly inadequate overall sentence (see also Gan Chai Bee Anne v Public Prosecutor [2019] 4 SLR 838 (“Anne Gan”) at [20]). The key is to ensure that the aggregate sentence is sufficient and proportionate to the offender’s overall criminality.

47 We consider the two-stage, five-step framework in Logachev an appropriate sentencing framework for ss 6(a) and (b) offences under the PCA. We say this for two reasons. First, at the first stage of the Logachev framework, the court classifies the severity of the offending conduct not by reference to the particular facts of each case, but based on the salient features of offending conduct (manifesting themselves as offence-specific factors). This makes it apt for encapsulating the diverse circumstances in which private sector corruption can occur. Second, given the established body of jurisprudence on offences under ss 6(a) and (b) of the PCA (see, eg, the discussion by the High Court in Public Prosecutor v Tan Kok Ming Michael and other appeals [2019] 5 SLR 926 (“Tan Kok Ming Michael”) at [99]), we find ourselves assisted in identifying the salient features of offending conduct from which the relevant offence-specific factors can be derived, and which will provide the court with a sense of how the sentencing spectrum under ss 6(a) and (b) of the PCA (up to a fine of $100,000 or an imprisonment term of five years or both) should be spread across the harm/culpability categories in the sentencing matrix at step two of the framework. From this point onwards in the judgment, we will refer to the proposed sentencing framework as the “revised sentencing framework”.

48 Having determined that a sentencing framework modelled after that in Logachev ([5] above) should be developed, we now turn to consider its scope. The Prosecution submits that the revised sentencing framework for offences under ss 6(a) and (b) of the PCA “can and should encompass offences under s 5 of the PCA”. For ease of reference, we set out these provisions in full:

49 As we had indicated during the hearing, we are not minded to extend the revised sentencing framework to offences under s 5 of the PCA and to cases of public sector corruption as it would be contrary to principle, for these were not the offences before us for the purposes of the appeal. Given that sentencing is often an intensely fact-sensitive exercise, we consider it inappropriate for the court to expound on a sentencing framework for such offences without the benefit of facts to which such a framework can be applied and tested. In addition, we set out brief conceptual reasons why we are not persuaded that the revised sentencing framework should be so extended.

50 In our judgment, the revised sentencing framework for offences under ss 6(a) and (b) of the PCA should not be extended to offences under s 5 of the PCA because both provisions are directed at distinct mischiefs and so will engage different considerations in the sentencing exercise. While s 5 of the PCA targets corrupt transactions more generally, s 6 is specifically directed at a situation where the corrupt procurement of influence involves the agent subordinating his loyalty to his principal in furtherance of his own interests. For instance, sub-subsection (ii) of s 5 of the PCA is directed at the corrupt procurement of influence within the public service, whether or not that involves an agent subordinating his principal’s interests over his own. As for sub-subsection (i) of s 5, it is a wide provision that was introduced following amendments to the Prevention of Corruption Ordinance 1960 (No 39 of 1960) by the Prevention of Corruption (Amendment) Act 1966 (Act 10 of 1966), which extended the scope of corrupt transactions coming within the fold of s 5 so that any form of corrupt giving is now criminalised under the PCA, even if it occurred outside of transactions involving public bodies or agents. Its purpose was to augment the State’s powers to address corruption and so it stood to reason that Parliament would not have had a distinct mischief in mind when legislating for sub-subsection (i) of s 5, because it was simply introduced to provide the State with even greater powers to tackle bribery and corruption (see Song Meng Choon Andrew v Public Prosecutor [2015] 4 SLR 1090 at [34]).

51 The different mischiefs at which each provision is directed give rise to the possibility that different sentencing considerations may be relevant for offences under s 5, as compared to offences under ss 6(a) and (b) of the PCA (see Wong Chee Meng ([31] above) at [59]). Further, the absence of a common mischief at which both provisions are directed means that the salient features attaching to offending conduct under each provision will likely differ. As such, they may not share a common pool of potentially relevant offence-specific factors for the purposes of step one of the revised sentencing framework. This means that particular offence-specific factors might come to be excluded simply because of the offence in question (whether it is one under s 5, or one under ss 6(a) or (b) of the PCA) and not because the attributes of the offending conduct justify such exclusion. For instance, offence-specific factors like actual loss caused to the principal and the extent of the offender’s abuse of position and breach of trust, which are prima facie relevant to an offence under ss 6(a) or (b), do not readily feature in an offence under s 5. In these circumstances, the absence of such factors in an instance of offending conduct under s 5 as compared to another instance of offending conduct under ss 6(a) or (b) where such factors were engaged says nothing about the relative severity of the two instances of offending conduct. Accommodating both s 5 and ss 6(a) and (b) offences within the same sentencing framework is therefore unworkable because the court has no intelligible means of classifying the severity of offending conduct under both provisions, using a common yardstick.

52 We similarly consider that the revised sentencing framework for private sector corruption offences under ss 6(a) and (b) of the PCA should not be extended to public sector corruption offences under the same provision.

53 We accept, as a matter of principle, that both public and private sector corruption offences under ss 6(a) and (b) of the PCA can be accommodated within the same sentencing framework. Since both offences involve the same offending provision, presumably, they would share a common pool of offence-specific factors for the purposes of step one of the revised sentencing framework.

54 However, accommodating both public and private sector corruption offences within the same sentencing framework can give rise to real difficulties. The overarching sentencing consideration in a case of public sector corruption is the distinct public interest in eradicating corruption in the ranks of public servants upon whom the smooth administration and functioning of the State are dependent (see Chua Tiong Tiong v Public Prosecutor [2001] 2 SLR(R) 515 (“Chua Tiong Tiong”) at [17]-[18]). On the other hand, in a case of private sector corruption, the public interest is in the private sector maintaining a reputation for being corruption free, with business being conducted in a fair and transparent manner so as to ensure that the public’s legitimate expectations of bona fides, commercial even-handedness and economic welfare are not prejudiced, and that the efficient operation of the market is not disrupted (see Public Prosecutor v Ang Seng Thor [2011] 4 SLR 217 (“Ang Seng Thor”) at [41]). Given these distinct overarching considerations, the context in which offence-specific factors come to be assessed in a case of public sector corruption differs quite significantly from that in a case of private sector corruption. Although our courts have eschewed a rigid public-private distinction in corruption cases (see, eg, Romel ([31] above) at [22]; Ang Seng Thor at [39]), that is quite different from saying that these offences must also be accommodated within the same sentencing framework, which requires that, at the very least, there be an intelligible means of classifying the severity of offending conduct in both situations.

55 Therefore, in our judgment, the revised sentencing framework should be limited only to private sector corruption offences under ss 6(a) and (b) of the PCA. It is to the components of the revised sentencing framework that we now turn.

56 We now turn to step one of the revised sentencing framework. We begin by considering the submissions which the Prosecution and Mr Tan have made in respect of the offence-specific factors to be included at step one of the revised sentencing framework. Both the Prosecution and Mr Tan take the position that the offence-specific factors set out by the High Court in Wong Chee Meng ([31] above) (at [62]) should similarly be adopted for the revised sentencing framework. However, they also urge us to include other offence-specific factors, in addition to those in Wong Chee Meng. There is largely common ground between them but some differences specifically in respect of the additional harm-related offence-specific factors. For ease of reference, we reproduce their respective positions in the table below:

57 The Prosecution also submits that some of these offence-specific factors, namely, (a) whether the public service rationale is engaged; (b) presence of public health or safety risks; and (c) presence of threats, pressure or coercion, should be regarded as “seriously aggravating”, in that where they are engaged, it would tend to attract a custodial sentence as a starting point for the offender.

58 Finally, both the Prosecution and Mr Tan also submit that the “amount of gratification given or received” (one of the offence-specific factors set out by the High Court in Wong Chee Meng) should be included as a culpability-related rather than as a harm-related offence-specific factor. This submission arises from a series of conflicting High Court decisions on this point. In Masui (HC) ([2] above) (at [245] and [250]), the court considered that the factor of “amount of gratification given or received” should be included as both harm-related and culpability-related offence-specific factors (subject to a sentencing court’s avoidance of double counting by stating exactly the relevance of the amount of the gratification to the two distinct independent variables of harm and culpability) because this factor was relevant to both harm and culpability. The court in Masui (HC) followed the High Court’s decision in Ang Seng Thor ([54] above) (at [46]), which held that the amount of gratification was “not only linked to the culpability of the offender … [but] also related to the harm caused by the offence” [emphasis in original]. In Wong Chee Meng (at [71]), however, the court considered that the amount of gratification was best regarded as a factor only going towards culpability.

59 We agree with the Prosecution and Mr Tan that the offence-specific factors set out by the High Court in Wong Chee Meng (at [62]) should be included at step one of the revised sentencing framework. For reasons that we will come to later, we are also of the view that the factor of “amount of gratification given or received” should be a culpability-related rather than a harm-related offence-specific factor (see [86]-[88] below). Returning to the offence-specific factors in Wong Chee Meng, they had not been adopted with regard to the specific context with which that case had been concerned (an offence under s 6 read with s 7 of the PCA), and we are of the view that they are equally applicable and relevant to offences of private sector corruption under ss 6(a) and (b) of the PCA. We set out these offence-specific factors:

60 The jurisprudence in connection with these offence-specific factors is well-developed and we need only to reiterate them briefly. In connection with the harm-related offence-specific factors:

(a) Corruption offences which occasion real harm to the agent’s principal are considerably more aggravated than those where the principal suffers little or no harm (see Heng Tze Yong v Public Prosecutor [2017] 5 SLR 976 (“Heng Tze Yong”) at [27]-[28]; Tan Kok Ming Michael ([47] above) at [99(b)(ii)]). The court should be alive to the fact that detriment can arise in a number of different ways. While the detriment to the principal will often be closely correlated to the profit obtained or benefit secured by the giver of gratification, this is not invariably the case (see Wong Chee Meng at [64]).

(b) Any benefit obtained by the giver of gratification should be viewed separately from any harm suffered by the principal. Although the benefit obtained by a giver of gratification will often take the form of profits which are realised as a result of the giver being able to enter into specific profitable transactions, the court should be alive to other less readily apparent but nonetheless very valuable benefits that the giver may derive from the corrupt transaction (see Wong Chee Meng at [65]).

(c) In considering the type and extent of loss to third parties or any public disquiet (if any) caused by the offending conduct, the court has regard to the wider impact of corruption offences on society (see Ang Seng Thor at [46]). The court considers if the offences in question are of a sort that have the effect of causing loss to third parties, or are capable of generating a sense of unease in the public (see Wong Chee Meng at [67]).

(d) The fact that an offender commits an offence as part of a criminal syndicate or group is in and of itself aggravating as it raises the spectre of organised crime, which is detrimental to society (see Wong Chee Meng at [68]; Ang Seng Thor at [33(d)]; Tan Kok Ming Michael at [99(d)]). For example, in Lim Poh Tee v Public Prosecutor [2001] 1 SLR(R) 241 (“Lim Poh Tee”) (at [31]), the court considered it an aggravating factor that the offending conduct of one police officer had the effect of drawing other junior officers into a “web of corruption” which rendered more officers beholden to corrupt gratification provided by the giver. Although that case was one of public sector corruption, it stands for the general proposition that the harm levels of offending conduct would be aggravated if the offence had been committed in a group setting, which is equally applicable to cases of private sector corruption.

(e) The presence of a transnational element serves to aggravate an offence (with no exception for corruption offences) given the greater difficulties involved in detecting and prosecuting such offences, as well as the need to take a firm and uncompromising stance against cross-border crime (see Public Prosecutor v Law Aik Meng [2007] 2 SLR(R) 814 at [42]).

61 As for the culpability-related offence-specific factors:

(a) For reasons that we will come to later (see [85]-[88] below), the amount of gratification is a factor going towards culpability because a person who gives or receives a larger bribe is generally regarded as more blameworthy. It is important to reiterate that there is no material difference, as far as culpability is concerned, between gratification taking the form of an outright gift and that in the form of a loan (see Wong Chee Meng at [73]). The repayment of any of the gratification received by the receiver has no relevance as a mitigating factor in the context of culpability because the giver of a gratification (to whom such repayment is made), far from being a victim, is in effect a co-conspirator in a corruption offence (see Wong Chee Meng at [74]). There can be situations where an offender actually returns the gratification (or genuinely attempts to do so) as part of a sincere effort to resile from the corrupt transaction (see Wong Chee Meng at [74]). We acknowledge that the policy of the law should encourage such behaviour. However, we leave this point for consideration when it arises given that we have not heard arguments on the same in this appeal.

(b) All other things being equal, an offence committed with planning and premeditation will be more aggravated than one which is committed opportunistically or on impulse, as the presence of planning and premeditation evinces a considered commitment towards law-breaking and reflects greater criminality (see Logachev ([5] above) at [56]). Similarly, an offence committed by sophisticated means will be more aggravated than one which is committed simplistically (see Logachev at [57]). For example, concerted efforts by an offender to avoid detection, such as through the falsification of accounts, would be considered as aggravating (see Wong Chee Meng at [75]).

(c) All other things being equal, an offence perpetrated over a sustained period of time will be more aggravated than a one-off offence because it indicates how determined the offending conduct is (see Logachev at [59]). Thus, custodial sentences have been imposed where corruption offences were committed over a long period of time, even if the amount of gratification was relatively low (see Ang Seng Thor at [63]). However, the sentencing court should be careful not to regard this as a separate aggravating factor if there are several charges before the court such that it might choose instead to address the point by running sentences consecutively (see Ye Lin Myint v Public Prosecutor [2019] 5 SLR 1005 at [76]).

(d) An egregious abuse of position or breach of trust can be treated as aggravating the offender’s culpability. This arises in cases where a significant degree of trust had been reposed in the offender by his principal, because it goes without saying that a principal who appoints such an agent ought to be able to rely on his fidelity unquestioningly (see Wong Chee Meng at [77]). For example, in Wong Teck Long v Public Prosecutor [2005] 3 SLR(R) 488 (“Wong Teck Long”), the offender, who had been the assistant vice-president and manager of a private bank in Singapore, was convicted of an offence under s 6(a) for obtaining gratification from a customer in return for recommending the grant of RM14.5m in credit facilities (meant only for high net-worth individuals) to six individuals referred by the customer and whose accounts were to be operated by the customer as an authorised third party. In fact, these six individuals were not high net-worth individuals (see Wong Teck Long at [2] and [5]). The court considered that an aggravating factor which warranted a lengthier sentence was the fact that the offender’s breach of trust and abuse of position were significant by virtue of the senior office which he held (see Wong Teck Long at [35]).

(e) The motive of an offender in corruption offences will typically be greed but there are some situations in which an offender’s motive might mitigate his culpability, such as if he had been a giver of gratification who only did so to avoid harm being inflicted on himself by the receiver (see Wong Chee Meng at [78]). However, the offender’s culpability will be aggravated if his motive had been to facilitate and conceal criminal acts. That was the case in Public Prosecutor v Tay Sheo Tang Elvilin [2011] 4 SLR 206 (at [22]), where the High Court considered it an aggravating factor that the offender, a police officer who had bribed his fellow officers as an inducement for them forbearing to report him to his supervisor for misappropriating monies and contraband cigarettes seized during a police raid, had been motivated by the need to conceal his criminal acts. This case was one of public sector corruption but it nevertheless stands for a general proposition on how the offender’s motive might come to be an aggravating factor, which is equally applicable to cases of private sector corruption (see also [60(d)] above).

62 We next consider whether the further offence-specific factors which the Prosecution and Mr Tan have urged upon us should be included at step one of the revised sentencing framework, in addition to those already set out in Wong Chee Meng ([31] above). We begin by considering the harm-related offence-specific factors before turning to the culpability-related offence-specific factors.

63 Both the Prosecution and Mr Tan urge us to include this as a harm-related offence-specific factor. This factor was identified by the High Court in Masui (HC) (at [241]) and it was said to apply in cases where the corrupt acts caused or have the potential to cause significant detriment to the giver of gratification, such as when the legitimate business of the giver was in fact interfered with because of the receiver’s conduct.

64 For the moment, we are not inclined to include this as a harm-related offence-specific factor under the revised sentencing framework. This is so for two reasons. First, in broad terms, the injury to the public interest arising from a private sector corruption offence refers to the prejudice caused to the public’s legitimate expectation of commercial even-handedness and transparent decision-making in business (see [54] above; see also Wong Chee Meng at [67]). For that reason, the harm caused by a private sector corruption offence is generally located by looking beyond the immediate parties to the corrupt transaction, such as the principal and the third parties with whom the agent comes to deal with, or by a consideration of any other public interest ramifications associated with the offending conduct. Second, it appears to us that this factor might be more appropriately considered in connection with the offender’s culpability because it is indicative of the manner in which the offence had been committed. Logically, if a giver of gratification stands to suffer harm from his continued participation in a corrupt scheme, then all other things being equal, one would expect him to withdraw from it to extricate himself from any further harm. The fact that harm had been caused to the giver demonstrates the inability of the giver to withdraw himself from the corrupt scheme. This speaks of the means which the agent had employed in obtaining the gratification, and that such means had been so effective in causing the giver to counter-instinctively act against his self-interests.

65 However, we acknowledge that, in cases where the agent is paid a bribe at his behest so that he will forbear from inflicting harm on the giver, even though there may have been no lawful basis for the infliction of such harm (see Romel ([31] above) at [26(c)]; see, eg, Masui (HC) ([2] above) at [289] and Public Prosecutor v Tai Ai Poh HC/MA 9046/2014, discussed in Romel at [53]-[59]), the giver’s position is akin to those third parties dealing with the agent whose legitimate interests have been compromised as a result of the agent’s solicitation of a bribe. In other words, the harm caused to such a giver is capable of being seen as falling outside of the corrupt transaction itself. We therefore leave it to a future occasion to determine whether harm caused to a giver of gratification should be analysed as a harm-related or culpability-related offence-specific factor under the revised sentencing framework, or, if it is to be included as a harm-related offence-specific factor, whether it can be more specifically analysed in the terms which we have set out in this paragraph.

66 The “public service rationale” is a sentencing principle which refers to the public interest in preventing a loss of confidence in Singapore’s public administration through corruption within (see Chua Tiong Tiong at [17]-[19]). It is presumed to apply in cases where the offender is a government servant or an officer of a public body (see Ang Seng Thor ([54] above) at [33(c)]). In these cases, a custodial sentence is typically justified (see Public Prosecutor v Chew Suang Heng [2001] 1 SLR(R) 127 at [10]-[11]).

67 It is also established law that private sector corruption offences can engage the public service rationale if the subject matter of the offence involves a public contract or an essential service akin to those provided for by public bodies (see Lim Teck Chye v Public Prosecutor [2004] 2 SLR(R) 525 (“Lim Teck Chye”) at [67]; Ang Seng Thor at [33(c)]). For example, in Lim Teck Chye, the appellant was a director of a company, “CBS”, which provided bunkering services. He was found guilty of having abetted by conspiracy with his employees to corruptly pay gratification to marine surveyors to falsely certify that CBS had supplied the correct quantity and quality of marine oil to vessels serviced by CBS. The court held that, although the appellant’s corrupt actions had taken place in the private sector, they had the potential to adversely affect public confidence in the independence of marine surveyors and Singapore’s bunkering industry and therefore the public service rationale was engaged (at [68]). What was significant and which warranted the extension of the public service rationale in Lim Teck Chye was the fact that the subject matter of the corrupt transaction involved a regulatory or oversight function akin to that discharged by a public body, and the fact that it had been discharged by private and not public actors in that case made no difference (see Ang Seng Thor at [33(c)]; cf Romel at [37]).

68 In his written submissions, Mr Tan argues that ss 6(a) and (b) offences under the PCA involving the public service rationale should be excluded from the revised sentencing framework given that the present focus is on purely private sector corruption offences. Mr Tan’s submission finds support in the position taken by the High Court in Masui (HC), which was of the view that the modified framework set out in that case only applied to “purely private sector corruption cases”, that is, private sector corruption offences which do not engage the public service rationale (see Masui (HC) ([2] above) at [79(c)] and [88]).

69 With respect, we disagree that the revised sentencing framework should be confined to purely private sector corruption offences which do not engage the public service rationale. In the first place, there is no reason in principle which justifies such a rigid distinction between private sector corruption offences engaging the public service rationale and those which do not. The only distinction between the two types of offences is that offending conduct engaging the public service rationale occasions damage to the public confidence in the public administration or provider of public services, in addition to the injury to society which ordinarily arises from a paradigm private sector corruption offence (see [70] below). The involvement of the public service rationale in a private sector corruption offence therefore does not render it so qualitatively different from one which does not as to render both offences incomparable for the purposes of stage one of the revised sentencing framework. Maintaining a rigid distinction between the two types of offences is also undesirable. Modern day governments have increased the outsourcing and privatisation of public services to private actors and so there is an increasing need to ensure these private actors, who are now also able to influence the public interest, are held accountable for the public services that they are responsible for delivering and the manner in which public money is spent (see Romel ([31] above) at [23]). In our view, confining the revised sentencing framework to private sector corruption offences that do not engage the public service rationale unduly and unnecessarily constrains its width, and undermines the objective which that framework seeks to achieve, namely, to provide a broadly consistent methodology for sentencing courts for all private sector corruption offences.

70 In our judgment, the public service rationale should be included as a harm-related offence-specific factor. Where the public service rationale is engaged in a case of private sector corruption, there is damage caused to public confidence in the provider of public or essential services, in the same way that damage is caused to public confidence in the public administration in a case of public sector corruption. This constitutes a distinct injury to society, apart from those which typically arise from offending conduct under ss 6(a) and (b) of the PCA, such as the immediate harm to the agent’s principal or third parties with whom the agent deals with. It should therefore separately count in the measure of harm caused by the offending conduct in question.

71 The Prosecution urges this court to include the following as harm-related offence-specific factors: (a) presence of public health or safety risks; (b) involvement of a strategic industry; (c) bribery of a foreign public official. We agree. Where any of these factors are engaged in a particular case, there is a distinct injury to society apart from those which typically arise from offending conduct under ss 6(a) and (b) of the PCA.

72 Where the offending conduct under ss 6(a) or (b) gives rise to public health or public safety risks, there is a distinct injury to members of the public who are reliant on the agent’s dutiful performance of his duties to mitigate or manage those risks, as they now stand to be impacted by such risks that they otherwise would not have if the offending conduct had not taken place (see, eg, Public Prosecutor v Marzuki bin Ahmad and another appeal [2014] 1 SLR 623 (“Marzuki bin Ahmad”) at [31]; Romel at [42]-[44]; Tan Kok Ming Michael ([47] above) at [99(b)(i)]). For example, in Romel (at [42]-[45]), the High Court considered that the offending agent, who overlooked high-risk defects in vessels that he was required to inspect on account of bribes which he had solicited in the performance of his duties, had posed both safety risks to the oil terminal and workers inside the terminal, a factor which it considered the court below ought to have appreciated in sentencing the offender.

73 Where the offending conduct involves a strategic industry or sector, there is a distinct injury to society because of the detriment caused to the development of that strategic industry or sector. For example, in Wong Teck Long ([61(d)] above) (at [36]), the court considered it an aggravating factor that the offender’s conduct stood to undermine the integrity of Singapore’s banking and financial industry, as well as Singapore’s reputation as a regional and financial hub. In Zhao Zhipeng v Public Prosecutor [2008] 4 SLR(R) 879, the offender, who was a football player in a football club in Singapore’s professional football league “S.League”, was convicted of offences under s 6(a) of the PCA for accepting gratification in return for not playing to the best of his ability so that his team would lose by a certain number of goals. The court considered it an aggravating factor that the offender’s conduct stood to destroy the reputation and commercial viability of the S.League and would hamper the development of international football in Singapore if left unchecked (at [28]-[31]).

74 Our courts have held that offending conduct involving a strategic industry will occasion a loss of public confidence (see, eg, Lim Teck Chye ([67] above) at [68]). Despite similar references to the loss of public confidence, it is important to emphasise that the involvement of a strategic industry and the public service rationale are distinct harm-related offence-specific factors (see Ang Seng Thor ([54] above) at [34]). But in cases involving a strategic industry, what renders offending conduct injurious is not the loss of public confidence per se, but the detriment caused to the development of the relevant strategic industry by such a loss of public confidence. We emphasise this distinction because one of the arguments put forth by the appellant is that the DJ had erred by regarding the public service rationale as being engaged as a result of her finding that KFELS was in a strategic industry. That argument is clearly misplaced because it is premised on the harm-related offence-specific factors of the involvement of a strategic industry and the public service rationale being interconnected, when they are not.

75 Turning now to the last of these harm-related offence-specific factors, the bribery of a foreign public official (whether by an offender within Singapore’s borders, or by a Singaporean offender outside of Singapore) gives rise to a distinct injury to the public interest because it can undermine Singapore’s international reputation for standing resolutely against corruption and also runs contrary to Singapore’s international obligations to combat transnational corruption (see Tan Kok Ming Michael at [75]-[92]). Thus, in Tan Kok Ming Michael, the court considered it an aggravating factor that the corrupt transactions in that case had implicated persons who were foreign public officials.

76 Mr Tan urges us to include this as a harm-related offence-specific factor. According to him, this concerns the dashing of the principal’s expectations as to whether and how the agent has carried out his duties to the principal. Mr Tan articulates this by reference to the three broad and non-exhaustive categories of private sector corruption set out by the High Court in Romel ([31] above) (at [26]):

77 Mr Tan submits, based on the Romel categories, an agent can fail to perform his duties faithfully in the following ways. This can involve the agent either improperly exercising his discretion (as in the first Romel category) or acting in dereliction of his duty by refraining from performing what he was duty-bound to do (as in the second Romel category) on account of gratification that he has received from the giver. It can also involve the agent doing either of the above on account of gratification that he has solicited from the giver for his own benefit (as in the third Romel category).

78 The Prosecution submits that the inclusion of this as a harm-related offence-specific factor is unnecessary because the manner in which an agent has failed to perform his duties will be captured by other harm-related offence-specific factors such as actual loss caused to the principal and benefit to the giver of gratification. It says that this factor only takes into account the nature and extent of the agent’s failure to perform his duties on a superficial analysis and does not entail a deeper consideration of the specific consequences resulting from such a failure or the resultant harm (as the other harm-related offence-specific factors do), and it is the latter which are more significant and important for the purposes of sentencing. Finally, the Prosecution also submits that including this factor would run the risk of double-counting to the prejudice of the offender.

79 We are unable to agree with Mr Tan and do not see it necessary to include this as a harm-related offence-specific factor. We accept that the nature and the degree of the agent’s failure to perform is the central target of s 6 of the PCA. However, every offence under ss 6(a) or (b) of the PCA would necessarily involve an agent who had acted in breach of duty or had failed to perform his duties faithfully. Under the Logachev ([5] above) framework, an offence-specific factor is meant to distinguish a case in terms of the severity of the offending conduct. Thus, if the factor which Mr Tan urges upon us is to provide any qualitative distinction between the cases, a sentencing court must inquire into the nature and extent of such breach of duty and/or failure to perform. Such an inquiry would require a sentencing court to embark on an analysis of the extent of the agent’s fulfilment of his equitable duties, which would be unduly onerous for the court. It would entail the court having to identify with precision the full scope of duties owed by the agent, without which the severity and significance of a particular breach of duty cannot be appreciated. We are not persuaded that such a complex exercise is one that a sentencing court, which is often limited to facts which are material for the purposes of the charges before it, is well-placed to undertake.

80 Both the Prosecution and Mr Tan urge us to include this as a culpability-related offence-specific factor. This factor was identified by the High Court in Masui (HC) ([2] above). However, it is apparent from the reasoning in Masui (HC) (at [243]) that the court considered this factor as only being applicable in the case of offences committed by agents under s 6(a) of the PCA. This is because the court considered this factor as capturing “a narrow stratum of cases” falling under the third category in Romel ([31] above). To recall, the third Romel category involves a receiving party who seeks out payment from the paying party and couples it with a threat to inflict harm on the paying party even though there may be no lawful basis for doing so (see Romel at [29]; see also [76] above). By definition, only the offending conduct of agents who obtain gratification for their own benefit under s 6(a) can come within the third Romel category.

81 We agree that an agent who couples his demand for gratification with threats, pressure or coercion is more culpable than one who did not. Such conduct, which entails an interference with or deprivation of a person’s legitimate rights unless a bribe is paid, is antithetical to everything that Singapore stands for and destroys the notion that business in Singapore is clean and transparent (see Romel at [30]). Such an agent is obviously more blameworthy because not only did he engage in the wrongful conduct of demanding gratification, but he also did so with the means (oftentimes unlawful) to make his demand more effective.

82 However, for the purposes of the revised sentencing framework, we are of the view that this factor should not be confined only to offending conduct under s 6(a) of the PCA. This is because distinctions in the culpability of a giver of gratification can conceivably also be drawn, based on whether he had coupled his giving of gratification with the use of threats, pressure or coercion. Regardless of whether the offender had been an agent-receiver or a giver, the manner in which he initiated the corrupt transaction and whether threats, pressure or coercion had been used in that process, are relevant considerations for a sentencing court in the assessment of the offender’s culpability. We therefore agree with the Prosecution and Mr Tan that this be included as a culpability-related offence-specific factor, and that it be extended to offending conduct under both ss 6(a) and (b) of the PCA.

83 The Prosecution urges us to include “whether the offender initiated the corrupt scheme” as a culpability-related factor because the fact that a party had initiated a corrupt scheme is a factor which increases his blameworthiness. The Prosecution’s position is shared by Mr Tan, although he frames this somewhat differently as “the identity of the party who initiated the bribe”.

84 We agree that whether the offender had initiated the corrupt scheme or the bribe is a relevant factor going towards his culpability. All other things being equal, a giver of gratification who initiated the corrupt transaction is more culpable than a giver who succumbed to the solicitation and pressure of the recipient (see Heng Tze Yong ([60(a)] above) at [34]). Similarly, an agent who merely received gratification from the giver when offered is less culpable than one who had actively sought out gratification from the giver (see Romel at [29]). However, we prefer to frame this factor more generally as “the role played by the offender in the corrupt transaction” (see Heng Tze Yong at [30]). In our view, it is preferable to examine holistically whether the offender’s role in the corrupt transaction had been active or passive, rather than to focus restrictively on whether he had initiated or solicited the corrupt scheme or the bribe. After all, the focus of the inquiry on culpability should necessarily be on the role of the offender.

85 We next consider the offence-specific factor of “amount of gratification given or received”. Both the Prosecution and Mr Tan urge us to follow Wong Chee Meng ([31] above) and regard the amount of gratification as only a culpability-related offence-specific factor. However, they differ slightly in their reasons. The Prosecution’s reasons are two-fold: (a) first, there is a risk of double counting if the amount of gratification is also regarded as a factor going towards harm; and (b) second, it may not be necessarily true that the higher the bribe amount, the greater the corrupt influence exerted or the greater the subversion of the public interest. On the other hand, Mr Tan says that the amount of gratification should not be included as a harm-related offence-specific factor because any injury occasioned by the quantum of gratification is already catered for by the court including “harm caused to the giver of gratification” as a harm-related offence-specific factor.

86 As we have explained earlier (at [64]), we are presently not inclined to include “harm caused to the giver of gratification” as a harm-related offence-specific factor under step one of the revised sentencing framework and so we say no more about Mr Tan’s submission for the moment. In our view, the amount of gratification given or received is undoubtedly an important factor in assessing the severity of the offending conduct and it should go towards the assessment of culpability rather than harm. We say this for two reasons.

87 First, although the amount of gratification given or received may to some extent serve as a barometer of the degree of harm caused, it is unnecessary that it separately features as a harm-related offence-specific factor. This is because the relevant degree of harm that is associated with the amount of gratification is sufficiently taken into account by the other offence-specific factors going towards harm (see Wong Chee Meng at [71]). In Ang Seng Thor ([54] above) (at [46]), the High Court considered the amount of gratification as relating to the harm caused by an offence because the higher the amount of a bribe, the greater the corrupt influence exerted, which presumptively leads to a greater subversion of the public interest for transactions and decisions to be carried out fairly and transparently. However, exactly what dimension of this public interest is subverted is context-specific and depends on the situation in which the corrupt transaction takes place. It is therefore preferable that any such injury to the public interest associated with the amount of gratification be accommodated by the other offence-specific factors, which cater for the diverse types of harm to which corruption offences may give rise (see also Wong Chee Meng at [71]).

88 Second, as a matter of principle, the amount of gratification is more properly analysed as a factor going towards culpability. “Culpability” is a measure of the degree of relative blameworthiness disclosed by the offender’s actions and is assessed chiefly in relation to the extent and manner of the offender’s involvement in the criminal act (see Logachev ([5] above) at [35]). In the case of both ss 6(a) and (b) offences under the PCA, all other things being equal, the amount of gratification typically (but not necessarily) bears a relationship to the blameworthiness of the offender:

(a) In the case of an offence under s 6(b) committed by a giver of gratification, the amount of gratification speaks of the means by which the offence was committed. The higher the quantum of gratification, the greater the incentive the giver wanted to create for the agent to compromise the performance of his duties to his principal, and the greater the influence or advantage the giver likely sought to obtain through the gratification (see also Ang Seng Thor at [47]). All other things being equal, a s 6(b) offender who offered a bigger gratification would be more blameworthy than one who has offered a smaller gratification.

(b) An agent who committed an offence under s 6(a) by receiving or demanding gratification of any amount in exchange for doing or forbearing to do something is necessarily blameworthy because he has subordinated his loyalty to his principal in furtherance of his own interests, when he ought to have placed his principal’s interests before his own. However, the greater the quantum of gratification received or demanded, the more it shows that the agent had viewed his position as nothing but a mere conduit for personal gain, and the more blatantly he had disregarded his duty of loyalty to his principal out of greed for personal monetary gain (see also Ang Seng Thor at [47]). Thus, all other things being equal, an agent who received or demanded a bigger gratification must be more blameworthy than one who has received or demanded a smaller amount.

89 Finally, we consider the Prosecution’s submission that the following offence-specific factors be designated as “seriously aggravating”: (a) whether the public service rationale is engaged; (b) presence of public health or safety risks; and (c) presence of threats, pressure or coercion.

90 In its written submissions, the Prosecution initially took the position that a designation of these offence-specific factors as “seriously aggravating” means that, in a case where any of these factors are engaged, it would presumptively attract a custodial sentence as a starting point for the offender. At the hearing before us, however, the Prosecution clarified that it did not intend to pursue that position and it had simply taken that position because the established case law has considered a custodial sentence as being justified in cases where any of those factors were engaged.

91 In our view, the Prosecution’s concession was rightly taken. We accept that a reading of the case law does suggest that the custodial threshold is generally regarded as being crossed in cases where any of those factors that the Prosecution identified as “seriously aggravating” had been present, but in each of those cases, that conclusion followed from the court’s assessment of the severity of the offending conduct as a whole, and not simply because particular “seriously aggravating” factors were present.

92 While we were initially minded to agree with the Prosecution that these identified offence-specific factors be designated as “seriously aggravating”, we ultimately decline to do so, for reasons of practicality and principle. First, we do not see what utility such a designation yields, given that no presumptive sentencing position is to follow by virtue of any of the “seriously aggravating” factors being engaged by the facts of the case. Second, the weight to be attributed to each of the offence-specific factors is sensitive to the facts and circumstances of the case (see Wong Chee Meng ([31] above) at [85]). Exactly what weight should be attributed to each of those factors is a matter for the sentencing court’s exercise of discretion, having taken into account all the relevant facts. Thus, the fact that any of these “seriously aggravating” factors are engaged per se does not give rise to the conclusion that the offending conduct in question is more serious. That is a conclusion which follows from the assessment of the offending conduct and the facts of the case as a whole.

93 Let us illustrate this with an example. Consider an agent who used threats in soliciting gratification, but which the giver knew the agent was not in a position to carry out and so the threats had no effect whatsoever on the giver, as compared to a case where no threats had been used, but only because the agent knew, from the position of influence which he occupied, that a demand for gratification alone would be just as effective and so it was unnecessary to resort to any such threats. In the example just posited, it is not a forgone conclusion, by virtue of the presence of threats alone, that the former case is more serious than the latter. While the presence/absence of threats, pressure or coercion (or indeed, for that matter, any other “seriously aggravating” factor) will generally provide a barometer by which the severity of the offending conduct may be appreciated, it is not conclusive of that issue. It would be quite inconsistent with the fact-sensitive exercise of sentencing for a court to begin with a presumptive view about the severity of the offence by virtue of such “seriously aggravating” factors being engaged without first considering the actual severity of the offending conduct in the context of the specific facts of each case, which is what preoccupies the first stage of the revised sentencing framework.

94 The point we make here is that a sentencing court should not ordinarily be required to make a predetermination of the severity of the offending conduct simply because some factors had been, as a matter of form, engaged by the facts of the case. In our view, that is precisely the effect of designating certain offence-specific factors as being “seriously aggravating”. That would have the untoward effect of fettering the discretion of a sentencing court, which fundamentally undermines the objectives which the adoption of a sentencing framework like the present seeks to achieve.

95 To summarise the above discussion, we are of the view that the following offence-specific factors should be included under step one of the revised sentencing framework. It bears repeating here that the various offence-specific factors under step one of the revised sentencing framework are not of equal weight, and they can assume different weight and significance depending on the facts of each case (see [92] above). The factors that are included in addition to those already set out in Wong Chee Meng are emphasised below in italics:

96 We now turn to consider step two of the revised sentencing framework. Both the Prosecution and Mr Tan take the position that the sentencing matrix in the revised sentencing framework should not include a default reference to fines, unlike that adopted in Masui (HC) ([2] above). On the other hand, the differences between the Prosecution and Mr Tan can be classified as falling under two main areas:

(a) In a case of slight harm/low culpability, the Prosecution’s indicative sentencing range is a fine or up to six months’ imprisonment, while Mr Tan’s is only a fine. The Prosecution argues that excluding the possibility of a custodial sentence in slight harm/low culpability cases would unnecessarily fetter the discretion of a sentencing court.

(b) The Prosecution’s sentencing matrix has provided for a broader sentencing range in cases of severe harm/medium culpability, moderate harm/high culpability and severe harm/high culpability, which it said is necessary to provide a sentencing court with the necessary scope to calibrate the sentence in these cases, which are likely to be aggravated and present a greater number of offence-specific factors. On the other hand, Mr Tan has reserved a broader sentencing range only in egregious cases of severe harm/high culpability.

97 In the light of the parties’ submissions, there are three main issues arising for this court’s determination:

(a) first, whether the indicative sentencing range for a case of slight harm/low culpability should be restricted to only a fine;

(b) second, in what harm/culpability categories should the revised sentencing framework provide for a broader sentencing range; and

(c) third, whether the sentencing matrix in the revised sentencing framework should include or omit a default reference to fines.

We consider each in turn.

98 We agree with the Prosecution that the indicative sentencing range for a case of slight harm/low culpability should include a custodial term of up to six months. In our judgment, offending conduct coming within the slight harm/low culpability category can nevertheless attract a custodial sentence, if particular offence-specific factors are engaged on the facts of the case. We illustrate this with two examples.

(a) A custodial sentence was imposed in Public Prosecutor v Lam Kim Heng [2018] SGDC 98, where the accused, a coffee shop staff, had corruptly offered a packet of cigarettes to a building inspection officer who worked for CPG Facilities Management Pte Ltd, which was authorised by the Housing and Development Board (“HDB”) to ensure that owners of food and beverage outlets do not infringe HDB regulations. This was after the officer informed the accused that there was illegal placement of tables and chairs in violation of HDB rules. The accused offered the cigarettes for the officer to show leniency in his inspection and also for the officer to inform him in advance of future inspections. The accused pleaded guilty to one charge under s 6(b) of the PCA and was sentenced to three weeks’ imprisonment (the accused’s sentence was subsequently varied on appeal in HC/MA 9080/2018/01 to a fine of $250 but specifically on account of judicial mercy as the accused suffered from a terminal illness).

(b) In Public Prosecutor v Su Fengxian [2018] SGDC 40, the accused had offered the technical director of Singapore Table Tennis Association gratification of €2000 as an inducement to show leniency in her son’s disciplinary case with the association. The accused pleaded guilty to one charge under s 6(b) of the PCA and was sentenced to six weeks’ imprisonment. The court observed that the public service rationale applied given that, amongst other things, the association performed a public service in recruiting and training players to represent Singapore to play against foreign players and teams in competitions overseas and received a substantial amount of funding from public funds for their work. The disciplinary proceedings had an “important oversight function” in ensuring that national players conducted themselves properly (at [85]-[86]).

We should add that, notwithstanding our reference to these cases involving guilty pleas, the sentencing matrix is concerned with benchmark sentences in relation to accused persons who claim trial (see Ng Kean Meng Terence v Public Prosecutor [2017] 2 SLR 449 (“Ng Kean Meng Terence”) at [40]). We have nevertheless referred to these cases because they are still helpful comparators in ascertaining the appropriate indicative sentencing range.

99 We now turn to the second issue. In our judgment, a broader sentencing range with an indicative starting sentence of three years’ imprisonment should only be provided for in cases of severe harm/high culpability. As a matter of principle, providing for a broad sentencing range in too many categories within the sentencing matrix will undermine the consistency in methodology which a sentencing framework is meant to provide. A broad sentencing range should only prescribed for categories of cases where it is required, such as where the situations coming within that category fall along a broader than usual spectrum and so the court requires greater width to calibrate a sentence accordingly. The category of severe harm/high culpability presents such a necessity given the great variation in offending conduct that might come within that category - such as borderline cases and those where the offending conduct is extremely severe and falls at the uppermost end of that category. Providing for a broader sentencing range in cases of severe harm/high culpability would allow them to be punished proportionately, ie, as deserved for the offence “having regard to the seriousness of the harm caused or risked by the offender and the degree of the offender’s culpability” (see Guay Seng Tiong Nickson v Public Prosecutor [2016] 3 SLR 1079 at [42]).We also note that, based on current sentencing practice, the imprisonment term for an offence under ss 6(a) or (b) of the PCA rarely exceeds two years - according to statistics from the LawNet’s Sentencing Information and Research Repository, between 22 October 2001 and 12 August 2022, there were only three cases involving offences under s 6(a) of the PCA in which sentences of over two years’ imprisonment were imposed, and only seven such cases in respect of s 6(b) of the PCA.

100 We also think it unnecessary for the sentencing framework to include a default reference to fines in each of the harm/culpability categories, as was done in the modified framework in Masui (HC) (at [213]-[216], [263]-[264] and [274]). This is because the imposition of fines is highly fact-sensitive, and fines are typically imposed in addition to an imprisonment term where the offender has substantially benefitted from his wrongdoing, and so it is necessary to disgorge the offender of his unlawful gains (see, eg, Public Prosecutor v Tee Fook Boon Andrew [2011] SGDC 211 at [78] (the fines imposed being subsequently affirmed on appeal in Public Prosecutor v Tee Fook Boon Andrew [2011] SGHC 192 at [38]) and Ang Seng Thor ([54] above) at [53] and [71]).

101 In these cases, fines generally served a purpose similar to that of s 13 of the PCA, which provides that an offender who is convicted of an offence under the PCA of accepting gratification which value can be assessed may be ordered to pay that amount as a penalty. Like a penalty order imposed under s 13 of the PCA, the imposition of fines served to “ensure that offenders are not able to retain their ill-gotten gains” (see Marzuki bin Ahmad ([72] above) at [61]; Leong Wai Kay v Carrefour Singapore Pte Ltd [2007] 3 SLR(R) 78 at [21]).

102 Since the question of whether a fine should be additionally imposed engages specific considerations of its own, it is quite distinct from the question of the appropriate custodial term. Therefore, in our view, the sentencing matrix of the revised sentencing framework should not refer to the sentencing option of a fine by default. As this court observed in Mao Xuezhong v Public Prosecutor and another appeal [2020] 5 SLR 580 (at [55]), fines and imprisonment are not “interchangeable”, as the reasoning of the High Court in Masui (HC) (at [303]) otherwise suggests.

103 In the light of the above analysis, and having regard to the stipulated sentencing range for offences under ss 6(a) and (b) of the PCA, the following sentencing matrix sets out the indicative starting sentences for an accused person who is convicted after trial:

104 For completeness, the parties do not dispute the subsequent steps to be applied under the sentencing framework, which have been canvassed above (at [45(c)] and [46]). To briefly recapitulate, after identifying the indicative sentencing range, the third step is to identify the appropriate starting point within that range, in an exercise of granulating the case with regard had to the same offence-specific factors as those considered at the first step. The fourth step involves making adjustments to the indicative starting point to take into account offender-specific factors, ie, established aggravating and mitigating factors personal to the offender. At the fifth step, where an offender has been convicted of multiple charges, the court will consider the need to make further adjustments to take into account the totality principle.

105 We now consider the issue of the sentence imposed on the appellant. We will discuss the Conspiracy Charges (see [10] above) and Non-Conspiracy Charges (see [12] above) separately in the following analysis.

106 In relation to the Conspiracy Charges, the following harm-related offence-specific factors are relevant:

(a) Actual loss caused to principal: $566,289.15, representing the amount that KFELS paid in wrongful marks-ups in respect of the invoices which are the subject of the Conspiracy Charges. Related to this, we consider that the appellant is mistaken in contending his offences were “victimless” (see [27] above). In the context of this case, KFELS was clearly the victim of his offences, as it suffered the harm of being made to pay more than it ought to have, in order to fund the illicit gains of the conspirators, a point which we will return to below (at [108]). KFELS also suffered the loss of ensuring that the best service providers were engaged, since the subject contracts were awarded based not on the quality of work of the vendors, but on account of the bribes offered by Titan, Spectrama and Growa.

(b) Benefit to the giver of gratification: The obvious benefit to Keh Choon, Goh and Fatkullah was that they were able to successfully secure contracts with KFELS, which they otherwise would not have been awarded.

(c) Type and extent of loss to third parties: Here, third party contractors were deprived of an opportunity to quote for jobs with KFELS, which they might otherwise have had without the intervention of Lim, who effectively decided which contractors would be invited to quote and thereafter be awarded the jobs. These jobs were awarded to Titan, Spectrama and Growa on account of the commission Lim received from Keh Choon, Goh and Fatkullah respectively, regardless of the terms which those vendors offered.

(d) Offences were committed as part of a group: The offences were part of a conspiracy between the appellant, Raj and Lim, wherein the former two decided they would seek vendors for jobs in KFELS and subsequently ask for the invoices to be marked up by more than 15%. Group offences have been observed to generally result in greater harm (see [60(d)] above), just as in the present case, where Raj was the one who knew Lim, and the appellant was the one who had contacts for the vendors, and the acts of the appellant, Raj and Lim in combination resulted in the offences.

(e) Involvement of a strategic industry: In our view, the DJ correctly observed that KFELS was in a strategic industry (see GD at [48]), which was said in Ang Seng Thor ([54] above) to include the bunkering and maritime industry (at [34], citing Lim Teck Chye ([67] above) at [68]). Here, the offences concerned jobs connected with the KFELS shipyard. As articulated in Romel ([31] above) (at [51]), the economic ramifications would be considerable should corruption take root in the maritime industry, which was observed in 2015 to account for up to 7% of Singapore’s gross domestic product and 170,000 jobs.

107 We note that the DJ was of the view that the only harm-related offence-specific factor was that of potential detriment, in the form of a possibility that there could be an unsuitable vendor which could then lead to “disastrous consequences” (see GD at [45]–[48]). The DJ did not however make any finding of actual harm. Before us, the appellant takes the same view that there was no actual harm. He argues that the finding by the DJ that an unsuitable vendor could lead to disastrous consequences is “hypothetical” and that “[i]n fact, this is a victimless crime”. The appellant submits that he was told by Raj to refer “capable vendors” to perform the jobs. The appellant further argues that KFELS did not pay more than it should have, because he had been told that the tenders submitted by the vendors as part of the conspiracy “had to be more competitive than other tenders”. Thus, KFELS had not been invoiced at a level above market rate.

108 These submissions by the appellant are mistaken. In our judgment, the DJ was led into error by the position taken before her by the Prosecution and the Defence that there was only potential harm in this case, which led to her mistakenly concluding that the harm was slight (a point of significance to which we will return later (see [112] below). It seems to us that all those in the court below failed to appreciate that the whole modus operandi of the offences was to cause the amounts paid by KFELS to be inflated in order to fund the bribes ultimately paid to the appellant, Raj and Lim. KFELS was therefore the victim in that it paid more than what it would otherwise have for those services rendered by Titan, Spectrama and Growa. To put things in context, KFELS paid about 45% more than it otherwise would have paid, as the aggregate mark-up amounted to about 45% of the total invoiced amount paid by KFELS to the vendors (before GST). There was therefore real and actual economic detriment suffered by KFELS.

109 There was additionally, as noted by the DJ, the potential harm which did not materialise, in terms of the downstream consequences of a sound procurement process within KFELS that had been bypassed as a result of the offences. We do not disagree that there was such potential harm. However, in our judgment, it is incorrect to consider this as the only harm arising from the Conspiracy Charges to the exclusion of the actual economic detriment suffered by the KFELS, as the DJ and the parties below have done. There was also, as just discussed (see [106] above), a number of other harm-related offence-specific factors which do not appear to have been considered in the court below, whether by the parties or the DJ. A proper assessment of these harm-related offence specific factors would thus necessarily produce a sentencing result markedly different from the one the DJ had reached.

110 We now turn to the culpability-related offence-specific factors for the Conspiracy Charges. The following factors are relevant:

(a) Amount of gratification given or received: $566,289.15, as noted above.

(b) Degree of planning and premeditation: There was a degree of planning and premeditation in the manner in which the bribes were sought, as described above (at [106(d)]). It is also significant that the appellant arranged for Ong and 3W to prepare fictitious invoices and delivery orders for Titan and Growa as part of the corrupt scheme (see [12(b)] above). We note that while the DJ had taken the position that the use of falsified invoices amounted to “efforts to cover up the corrupt transactions in the [Conspiracy Charges]” (see GD at [55]–[56]), that does not seem to be so - the issuance of false invoices was not so much concerned with avoiding detection, but was rather the means by which the corrupt scheme was carried out. On those occasions when 3W’s invoices were used, 3W was simply an alternative to Megamarine for invoicing Titan and Growa for the commission payable under the corrupt scheme.

(c) Duration of offending: The offending took place on numerous occasions over three years.

(d) Offender’s motive in committing the offence: The Statement of Facts to which the appellant admitted states that he had committed the offences because he “needed the money”. Further, he stated in mitigation that the period of 2015 to 2017 were “not good years for the machine industry” and that he was motivated by a “business opportunity”. Although the appellant argues that there should be a distinction between “profit motivation in business and greed” for sentencing purposes and emphasises that he did not personally pocket the gratification, which was used for the running of the company, we are of the view that such a distinction is untenable in the circumstances - according to the Statement of Facts, excluding what was paid by Megamarine as corporate tax, the appellant still received $191,115.89 in respect of the Conspiracy Charges. The DJ was correct to observe that the appellant “was not acting out of altruistic reasons when he committed the offences” (see GD at [58]).

111 In relation to the proceeded Non-Conspiracy Charges, the following harm-related offence-specific factors are relevant:

(a)  For the charge involving Raj, the appellant, as a giver of gratification, enjoyed the benefit of Megamarine successfully securing the contract from Rotating for the supply of hardware.

(b) For the charges involving Ong and 3W, the issuance of the fictitious invoices was the means by which the corrupt scheme involving the Conspiracy Charges were carried out, from which the appellant, as a giver of gratification in this transaction with Ong, benefitted financially.

As for the culpability-related offence-specific factors, it is relevant, in relation to the charge involving Ong and 3W, that the appellant had initiated the corrupt scheme, having asked Ong to help prepare the fictitious invoices in return for 8% of the invoiced amount.

112 As regards the Conspiracy Charges, the parties had in the proceedings below agreed that the assessment of harm in respect of those offences was slight, with medium culpability. But, as we have discussed above, the DJ and the parties had erred in considering that there was only potential harm in this case (see [108]-[109] above). While the appellant maintains in the appeal that his offences were of slight harm/medium culpability, the Prosecution now takes the view that the Conspiracy Charges were of moderate harm/medium culpability, in the light of a number of other harm-related offence-specific factors which do not appear to have been considered by the DJ (see [106] above).

113 In our judgment, the DJ erred in her decision that the Conspiracy Charges were of slight harm/medium culpability. This follows given our view that the Conspiracy Charges involved actual harm, and also since several other harm-related offence-specific factors that had not been canvassed in the proceedings below were, in our view, relevant in this case. We agree with the Prosecution that the correct assessment of the Conspiracy Charges is one of moderate harm/medium culpability, with the result that the indicative starting sentence for each of the Conspiracy Charges would be one to two years’ imprisonment. Given the erroneous basis on which the lower court proceeded in the assessment of harm, it is also not surprising that the sentence which we consider to be appropriate will be far higher than that imposed by the DJ (see [125] and [129]–[130] below).

114 As for the Non-Conspiracy Charges, we agree with the Prosecution that these may be assessed as being of slight harm/low culpability. The charge involving the appellant corruptly giving $3,000 to Raj was not connected with the Conspiracy Charges, having occurred prior to the appellant participating in the conspiracy (see GD at [11]), and was of a small amount. The appellant’s culpability was also low as the corrupt act had been initiated by Raj and not the appellant, who was informed by Raj that he (the appellant) would have to pay Raj some gratification if the appellant wanted business from Rotating. Next, although the charges involving the appellant corruptly giving sums of between $2,308.32 and $2,977.12 to Ong of 3W for preparing fictitious invoices furthered the purposes of the scheme that is the subject of the Conspiracy Charges, those offences did not have other overtly deleterious effects. As for culpability, since the appellant had played an active role by initiating the bribes in the charges involving Ong and 3W, this placed him at the upper end of low culpability (see [84] above), as opposed to falling under medium culpability.

115 Having regard to what has been said above (at [106]–[111]), in particular the significant amount of actual loss to KFELS occasioned by the Conspiracy Charges (see [106(a)] above), the appellant’s sentence for the amended first charge involving Keh Choon/Titan and a gratification of $107,000 (see [10(a)] above) should fall within the middle to high level of the indicative sentencing range. We agree with the starting sentence of 21 months’ imprisonment proposed by the Prosecution for that charge. The indicative starting sentences for the 14 remaining charges relating to Goh/Spectrama (see [10(b)] above) and Fatkullah/Growa (see [10(c)] above), each of which involve gratifications of relatively lower amounts between $21,835.41 to $46,170.50, but which otherwise also engage the same harm and culpability-related offence specific factors discussed above (at [106] and [110]), are adjusted downwards to 14 to 16 months’ imprisonment per charge, since the appellant’s culpability in respect of those charges would be correspondingly lesser (see [88] above).

116 In relation to the Non-Conspiracy Charges, the Prosecution recommends an indicative starting sentence at the lower end of the custodial threshold, of one to one-and-a-half months’ imprisonment. In our view, this would be appropriate for the offences involving Ong and 3W but not the offence involving Raj. As discussed above, the offence involving Raj was of slight harm/low culpability, with lower culpability as compared with the offences involving Ong and 3W, since the latter were related to the offences that were the subject of the Conspiracy Charges, and specifically were part of the means by which the corrupt scheme in the Conspiracy Charges had been carried out. Given the absence of any other offence-specific factors for the offence involving Raj, we do not think that the custodial threshold was crossed. We note as well that fines have generally been imposed in cases with similar facts, ie, where gratification has been requested from the paying party in order to advance that party’s business interests (see, eg, HC/MA 9094/2019/01 Loo Nee Soon v Public Prosecutor and Heng Tze Yong ([60(a)] above). However, an indicative starting sentence of one month’s imprisonment is appropriate for the charges involving Ong, bearing in mind the appellant’s higher culpability in initiating the corrupt scheme.

117 The DJ noted that the relevant aggravating factor was the 40 charges that had been taken into consideration for the purpose of sentencing (see GD at [81]). Further, the significant mitigating factors were the appellant’s full co-operation with the CPIB and his early plea of guilt (see GD at [84]). In our view, the DJ correctly considered that these factors were relevant.

118 The Prosecution has submitted that a reduction of around 25% from the indicative starting sentences of each of the charges is justified on account of the appellant’s plea of guilt. Although we agree that such a reduction is justified, consistent with the observations made earlier this judgment (see [43] above), we are of the view that a sentencing court should approach this issue qualitatively rather than quantitatively. In determining the extent of reduction to be applied on account of an offender’s plea of guilt, the task of the court is not a mathematical one, but one of calibrating the sentence to fit the facts of the case and arriving at a sentence that fits both the crime and the offender (see Ng Kean Meng Terence ([98] above) at [71]). With these considerations in mind, we are of the view that the sentences for the Conspiracy Charges may be reduced to about 15 months’ imprisonment for the charge involving Keh Choon/Titan and about ten to 12 months’ imprisonment for each of the 14 charges relating to Goh/Spectrama and Fatkullah/Growa.

119 Applying these same offender-specific factors to the Non-Conspiracy Charges, the starting sentence for the charges involving Ong/3W may be reduced to three weeks’ imprisonment per charge.

120 The appellant has argued that the fact that many charges were taken into consideration for sentencing was a result of the appellant’s decision to plead guilty. As such, these charges should not be assessed as an offender-specific aggravating factor as it would “negate his early plea of guilt”. He also argues that the fact of many offences should instead be an offence-specific factor going towards harm, as opposed to an offender-specific aggravating factor.

121 As to the first argument, in our judgment, it is apparent from the discount provided on account of the appellant’s guilty plea (see [117]-[118] above) that there has been no “negation” of the effect of the said plea. As to the second argument, it is incorrect as a matter of principle for multiple offences which have been taken into consideration for sentencing purposes to be considered an offence-specific factor. Offence-specific factors are those concerned with the “level of gravity of the crime in specific relation to the offence upon which the accused was charged” (see Ng Kean Meng Terence at [42], citing Public Prosecutor v Huang Hong Si [2003] 3 SLR 57 at [8]). Therefore, offences taken into consideration cannot have a bearing on the severity of the offence for which the offender is being sentenced, since they concern distinct offences which the offender has committed, and which are not the subject of the sentencing exercise itself. Instead, offences taken into consideration are only relevant as an aggravating offender-specific factor (see Logachev ([5] above) at [64]), since they pertain to the offender’s personal circumstances beyond the offences for which he is being sentenced.

122 As mentioned above, the fifth step of the revised sentencing framework requires the court to consider whether the sentences ought to be adjusted on account of the totality principle.

123 In view of the appellant’s multiple offences, we consider it would be appropriate to order, as the DJ had, that one sentence from the charges relating to each vendor (Titan, Spectrama and Growa), one sentence from the charges relating to Ong and one sentence from the charge relating to Raj run consecutively, having regard to the principles articulated in ADF v Public Prosecutor and another appeal [2010] 1 SLR 874 (at [146]) (see GD at [105]). The totality principle, in turn, requires the sentencing judge to take one “last look” at all the facts and circumstances and be satisfied that the aggregate sentence is sufficient and proportionate to the offender’s overall criminality (see Raveen Balakrishnan ([39] above) at [73]; Public Prosecutor v Su Jiqing Joel [2021] 3 SLR 1232 at [121]). It should be noted, however, that the totality principle not only possesses a limiting function in guarding against an overall sentence, but also a “boosting effect on individual sentences where they would otherwise result in a manifestly inadequate overall sentence” [emphasis in Azlin] (see Azlin ([46(b)] above) at [199] and [205]–[206]; Anne Gan ([46(b)] above) at [20]).

124 The first limb of the totality principle entails examining whether the aggregate sentence is substantially above the normal level of sentences for the most serious of the individual offences committed, while the second limb entails consideration of whether the effect of the sentence on the offender would be crushing and not in accordance with his past record and future prospects (see Mohamed Shouffee bin Adam v Public Prosecutor [2014] 2 SLR 998 at [54] and [57]; Wong Tian Jun De Beers v Public Prosecutor [2021] SGHC 273 (“Wong Tian Jun”) at [65]). We note, however, that there is limited utility in applying the first limb of the totality principle in the present case given the need to consolidate the approach to sentencing for ss 6(a) and (b) offences under the PCA going forward, and the fact that there appears to have been thus far underutilisation of the full sentencing spectrum for these offences (see [31] above). The second limb is, on the other hand, relevant in allowing this court to determine an appropriate overall sentence, bearing in mind the need that such sentence must be matching and proportionate to the offending behaviour (Wong Tian Jun at [68]).

125 We set out the sentences imposed by the DJ as well as the sentences which we considered to be appropriate under the revised sentencing framework after properly taking into account the relevant factors that we have identified in this judgment. The word “consecutive” is also noted in parentheses where a particular sentence was ordered to run consecutively.

126 Based on the above analysis, the aggregate sentence of 17 months and three weeks’ imprisonment imposed by the DJ, compared against the 37 months and three weeks’ imprisonment as well as a fine that we have arrived at by applying the revised sentencing framework and with the relevant factors properly considered, was lenient and cannot be said to be manifestly excessive. Given the substantial divergence between the sentences imposed by the DJ and that which we have arrived at on an application of the revised sentencing framework with a proper consideration of the relevant factors, a question arises as to whether the sentences ought to be enhanced. This is notwithstanding that the Prosecution did not file a cross-appeal against the sentences.

127 While the court hearing an appeal for reduction of sentence by an accused person will not normally enhance the sentence in the absence of a cross-appeal by the Prosecution (see Shafruddin bin Selengka v PP and other appeals [1994] 3 MLJ 750), such enhancement may nevertheless be ordered in exceptional cases where the sentence is manifestly inadequate (see, eg, Rosli bin Supardi v Public Prosecutor [2002] 3 MLJ 256 at 263). This was the case, for example, in Wong Tian Jun ([124] above). The High Court was of the view that the sentences imposed by the District Court on an offender for various cheating charges had not been properly calibrated, given that the offender had scammed his victims for sex and sexually explicit material and so the offences that the offender had committed were at the very highest end of the harm which might arise for offences under s 417 of the Penal Code (Cap 224, 2008 Rev Ed) (the “Penal Code”). As such, after specifically informing the parties that an enhancement of the aggregate sentence was possible and considering further submissions from them (see Wong Tian Jun at [3]), the court held that, on an application of a sentencing framework for charges under s 417 of the Penal Code, the individual sentences for those charges would be increased from between eight and ten months’ imprisonment to between 33 and 36 months’ imprisonment (see Wong Tian Jun at [51]).

128 Before us, Mr Singh stated that if the issue of there being more relevant offence-specific factors had been previously raised by the Prosecution in the proceedings below, he would have run the appellant’s defence differently, such as by emphasising the point raised in mitigation that the tenders that were the subject of the Conspiracy Charges had to be competitive. But, as we explained at the hearing, that argument as to a competitive price having been paid by KFELS for the subject jobs is misconceived. It is irrelevant in the context of an offence under s 6 of the PCA, which is to prevent the relationship between an agent and his principal from being undermined. Further, even if the appellant could have hypothetically proven that KFELS would have paid exactly the same amount in an open tender for those jobs, this does not change the fact that KFELS had in the present case paid substantially more for them in order to fund the activities of the conspirators, as discussed above. We therefore fail to see how any such adjustment in the running of the appellant’s defence would have assisted him. The Prosecution stated that while it was not appealing against the sentences imposed by the DJ, there were co-accused persons whose cases were pending before the State Courts, and concerns of parity meant that the present sentences should not be unduly low.

129 We do not think, however, that it lies in the Prosecution’s mouth to raise such an objection at this late stage, given that it had chosen not to appeal against the sentence imposed by the DJ in the first place. Nevertheless, the present case was one where the DJ’s decision on sentence had been premised on the misapprehension by the parties before her that the only relevant harm factor was one of “potential detriment to KFELS” (see GD at [45]). The appellant aligns himself with that position goes so far as to take the position before us that there was no real victim and that the contracts awarded by KFELS were in fact competitive. For the reasons that we have set out above (at [106] and [108]–[109]), and as we have explained to the parties at the hearing before us, that view is entirely misplaced, and it is clear that the DJ had proceeded on an erroneous basis in respect of the sentences she imposed for each of the Conspiracy Charges. Even without drawing any comparison with the sentences arrived at using the revised sentencing framework, we are independently satisfied that the sentences imposed by the DJ were manifestly inadequate and that an enhancement of the sentences for each of the Conspiracy Charges is necessary to fit the severity of the subject offences. We therefore set aside the sentences imposed by the DJ in respect of the Conspiracy Charges and replace them with the sentences arrived at by applying the revised sentencing framework. We do not however disturb the sentences imposed by the DJ in respect of the Non-Conspiracy Charges, as the DJ had not so erred in respect of the sentences for those charges. For avoidance of doubt, we also agree with the DJ’s decision as to which of the charges are to be run consecutively.

130 The resulting global sentence, taken into account those sentences which are to be run consecutively, is as follows:

131 We also take this opportunity to remind accused persons who contemplate filing appeals against their sentences to bear in mind that the court will consider enhancing sentence(s) in cases of plainly unmeritorious appeals, even in the absence of a cross-appeal by the Prosecution.

132 Accordingly, we dismiss the appeal, and the aggregate sentence imposed is enhanced to 37 months and three weeks’ imprisonment. It remains for us to thank Mr Tan for his thoughtful and comprehensive submissions, from which we have derived considerable assistance.