1 The emergence of fintech companies and digital payment platforms has brought about many changes to the financial industry. While these developments afford greater convenience and accessibility for consumers, they have also created new risks and challenges for regulators. In light of these developments, the Payment Services Act 2019 (No. 2 of 2019) (“PSA”) was enacted to provide a regulatory framework that is adaptive to the changing landscape of payment services.

2 The PSA enhances the overall efficiency and security of payment systems, promotes innovation, and ensures that payment service providers are held accountable for their actions. In particular, the PSA aims to protect consumers and merchants from potential risks associated with payment services. The regulatory framework under the PSA is designed to ensure that payment service providers meet certain minimum standards. Failure to comply with the framework not only exposes consumers and merchants to financial risks but also undermines the integrity of Singapore's financial system. Providing payment services without a licence is an offence under s 5(3) of the PSA. Unlicensed payment service providers, in particular, pose significant risks to consumers; these risks include fraud, money laundering, and terrorism financing. Hence, there is a need to strongly deter such offences to protect the integrity and stability of Singapore's financial system.

3 The above considerations come to the fore in the present appeal against the sentence of two weeks’ imprisonment imposed pursuant to the appellant’s plea of guilt to a charge under s 5(1) punishable under s 5(3)(a) of the PSA for the provision of cross‑border money transfer services (or money remittance services) without a licence. An important question of law is raised for this Court to consider the appropriateness of developing a sentencing framework under s 5(3)(a) of the PSA for offences that involve individuals providing payment services without a licence.

4 Specifically, this appeal presents a timely opportunity to consider the establishment of a sentencing framework for the specific offence of providing money transfer services without a licence under s 5(3)(a) PSA.

5 Having considered the parties’ submissions, I am of the view that it is desirable and appropriate to lay down such a sentencing framework. I adopt the “single starting point” framework with a starting point of three weeks’ imprisonment. Applying the framework to the present case, I affirm the sentence of two weeks’ imprisonment that was imposed on the appellant. I set out my reasons below for dismissing the appeal.

6 The appellant pleaded guilty to a single charge under s 5(1), punishable under s 5(3)(a) of the PSA, which involved carrying on a business of providing the payment service of cross-border money transfer services without a licence: ¹

7 The appellant admitted and consented to the following charge under s 6(2) of the Money-Changing and Remittance Businesses Act (Cap 187, 2008 Rev Ed) (“MCRBA”) (which has since been repealed) being taken into consideration for the purpose of sentencing: ²

8 The appellant admitted to the Statement of Facts without qualification. ³ The appellant owned and ran a company, East Village Pte Ltd, which dealt with the import of medicinal products from India and the sale of international calling cards. Through this business, he also offered remittance services to his customers. The appellant’s customers initially asked him for assistance in filling up paperwork for remittance at licensed remittance agents, and the appellant eventually decided to provide the remittance services himself.

9 The appellant provided the remittance services using the “hawala” method. This entailed enlisting the aid of his nephew in Myanmar to disburse the monies to beneficiaries in Myanmar after the appellant had collected monies from his customers in Singapore. The relevant details of each transaction were keyed into a spreadsheet on the Google Drive file-sharing service, which the appellant’s nephew accessed in Myanmar. The monies collected in Singapore were then used to buy goods in Singapore and shipped to the appellant’s nephew in Myanmar.

10 In providing the remittance services, the appellant charged between $2 to $10 per transaction and a further bank charge of between $1 to $3 depending on bank fees in Myanmar. The appellant also sold international calling cards to those customers who needed to call their families to obtain the details of the beneficiaries in Myanmar.

11 The appellant started providing the remittance service sometime in 2018. Between 3 February 2020 to 28 June 2020, he collected and remitted $10,123.20 and earned a service fee of $80 from those transactions.

12 The respondent sought a custodial term of at least three weeks’ imprisonment, citing general deterrence as the main sentencing consideration. The respondent referred to two unreported precedents as a comparison, as the only available reported case for offences under s 5 of the PSA, Public Prosecutor v Lange Vivian [2021] SGMC 116 (“Lange Vivian”), involved the unlicensed provision of a digital payment token service, a different type of payment service under the First Schedule of the PSA. The respondent also submitted that the sentencing precedents for equivalent offences under the repealed MCRBA should not be followed for offences under s 5 of the PSA, as the available sentencing data from the Sentencing Information and Research Repository (“SIRR”) suggested that sentences for offences under s 6 of the MCRBA tended to cluster at the lower end, with about 72.73% or 24 out of 33 cases resulting in fines being imposed. The sentencing courts did not appear to have fully utilised the sentencing range. Furthermore, the cases did not appear to have considered the custodial terms imposed by the High Court even for first offenders in prior cases, a point noted by District Judge Audrey Lim (as she then was) in Public Prosecutor v Shahabudeen s/o Asappa Abdul Hussain [2003] SGDC 122 (“Shahabudeen”) (at [14]).

13 The appellant submitted that a fine of $8,000 should be imposed. He claimed that it was common practice for Myanmar businesses to collect monies from workers in Singapore to pay for goods that were exported to Myanmar and for the Myanmar importers to disburse monies to the beneficiaries of those workers as payment for the goods. This was due to difficulties in remitting monies. Furthermore, his remittance service benefitted the workers as many of their beneficiaries did not have access to the formal banking system. The particulars of the workers were also recorded. The appellant carried on the remittance service to help his fellow countrymen, especially as the COVID-19 pandemic had exacerbated their difficulties in remitting monies to Myanmar.

14 In Public Prosecutor v Vijay Kumar [2022] SGMC 62, the learned Principal District Judge (the “PDJ”) convicted the appellant on his plea of guilt and sentenced him to two weeks’ imprisonment. The PDJ found that general deterrence was the main consideration for offences under s 5(3) of the PSA. The PSA’s objective was to enhance the regulatory framework for payment services in Singapore, which included cross-border money transfers. Key risks identified in this regard include loss of customer monies due to the insolvency of service providers, money laundering and terrorism financing.

15 Furthermore, the PDJ found that the custodial threshold had been crossed. The PDJ distinguished the cases under s 6 of the MCRBA where fines were imposed, noting that he was unable to give much weight to the unreported cases as it was not possible to discern what weighed on the mind of the sentencing judge in those cases. The PDJ also noted that the High Court had, in fact, imposed significant custodial terms for offences under s 6 of the MCRBA, even for first offenders and where there was no loss caused. Taking reference from the sentencing factors adopted by the court in Lange Vivian, the PDJ concluded that in the present case, the following factors were relevant. First, there was a transnational element involved due to the provision of a cross-border money transfer service. This was considered an aggravating factor as the PSA regulated domestic transfers separately. Second, the quantum involved in the present case ($10,123.20) was also higher than the quantum in Lange Vivian ($3,350). Third, the PDJ considered that the appellant ran the remittance on a commercial basis, charging a fee and bank charge, and even sold international calling cards to customers who needed them to verify details of the remittance transaction. Fourth, the offence had been committed for close to five months. Finally, there was also a further charge under the MCRBA taken into consideration.

16 The PDJ rejected the appellant’s submission that the remittance services were provided to help Myanmar workers who were unable to remit monies through licensed avenues. Whatever shortcomings there may be in the Myanmar financial system, this cannot justify the operation of an unregulated and illegal shadow financial system, spanning two countries, that would have an impact on the reputation and integrity of Singapore’s financial system. As the appellant’s own submission suggested that such illegal services were widespread among the Myanmar community in Singapore, this justified a greater need for deterrence.

17 The appellant appeals against his sentence of two weeks’ imprisonment, repeating his submission below for a fine of $8,000 to be imposed instead. The appellant submits that the PDJ erred in concluding that the custodial threshold had been crossed. This is because the offence fell within the lowest end of the harm and culpability spectrum, given that there was no loss caused. Furthermore, the PDJ placed undue weight on the aggravating factor of the period of offending. The PDJ also erred in concluding that the appellant’s main business benefitted when he only received $80 in commission and in finding that only a custodial sentence would be an effective deterrent sentence when a fine could be just as effective.

18 The appellant argues that the following mitigating factors were not given sufficient weight. The appellant’s intention was to help his fellow countrymen who had to resort to unlicensed remittance businesses as there was a lack of such services servicing the rural areas in Myanmar. The monies remitted were not tainted with illegality. The remittance system which he employed was not complex and sophisticated.

19 On the issue of the applicable sentencing framework under s 5(3) of the PSA, the appellant makes no specific submission but merely relies on the approach adopted in Lange Vivian, ⁴ where the court derived the final sentence by considering the culpability and the harm caused by having regard to the offence-specific factors and offender-specific factors.

20 In response, the respondent submits that the PDJ arrived at the correct sentence of two weeks’ imprisonment. This is because the PDJ rightly found that general deterrence was the primary sentencing consideration, and the custodial threshold was crossed. Furthermore, the PDJ correctly weighed the mitigating factors raised by the appellant.

21 In relation to the appropriate sentencing framework for individuals who commit offences punishable under s 5(3)(a) of the PSA, the respondent submits that the sentencing precedents for offences under s 6(2) of the MCRBA should not be referred to in determining the appropriate sentence. This Court should establish a sentencing framework by adopting a “single starting point” framework and provide for a custodial term as the starting point. A starting point of four weeks’ imprisonment is appropriate for offenders who claim trial. A downward adjustment is warranted to the starting point, given the appellant’s plea of guilt to arrive at the final sentence of two weeks’ imprisonment.

22 As this was the first appeal to the High Court concerning a s 5(3) PSA offence, Ms Tai Ai Lin was appointed as a Young Independent Counsel (“YIC”) to assist this Court in considering the appropriateness of developing a sentencing framework for offences under s 5(3)(a) of the PSA. The following questions were put to her. First, are the sentencing precedents for offences under s 6(2) of the MCRBA relevant in determining the appropriate sentence for offences under s 5(3) of the PSA? Second, should the Court establish a sentencing framework for such offences, and if so, what would be an appropriate sentencing framework? Third, when is the custodial threshold crossed for such offences?

23 In relation to the first question, the YIC submits that while there are similarities in the offences between s 5(3) of the PSA and s 6(2) of the MCRBA, the sentencing precedents for the latter offence have limited relevance and precedential value. Nonetheless, those sentencing precedents can assist this Court in formulating a sentencing framework as they help in identifying the relevant factors to be taken into account in the sentencing framework for s 5(3) PSA offences. ⁵

24 In relation to the second question, the YIC suggests that this Court should establish a sentencing framework for s 5(3) PSA offences and that the appropriate framework is the two-stage, five-step framework set out in Logachev Vladislav v Public Prosecutor [2018] 4 SLR 609 (“Logachev”). ⁶

25 In relation to the third question, the YIC submits that the question of whether the custodial threshold is crossed should be answered after deciding on the sentencing framework. In this case, as the appropriate sentencing framework is the Logachev framework, the custodial threshold is crossed when the level of harm caused by the offender is at least moderate, and the level of culpability of the offender is at least medium, after all mitigating factors have been accounted for. ⁷

26 In evaluating the correctness of the PDJ’s sentencing decision, the key issues for determination in this appeal are:

(a) Are sentencing precedents under s 6(2) of the MCRBA relevant in relation to offences under s 5(3) of the PSA?

(b) Is there a need to establish a sentencing framework for s 5(3) PSA offences?

(c) What is the appropriate sentencing framework for s 5(3) PSA offences?

27 I note that the parties have been asked to address the issue of when the custodial threshold is crossed for s 5(3) PSA offences. I will address this in connection with the determination of the appropriate sentencing framework to be adopted.

28 The appellant parts ways with both the respondent and the YIC on the relevance of the precedents under s 6(2) of the MCRBA. The appellant contends that given the common objective between the MCRBA and the PSA in preventing money-laundering activities and terrorism financing, the sentencing precedents involving the unlicensed remittance of monies under the MCRBA ought to be relevant when considering the present offence of unlicensed remitting of monies under the PSA. ⁸ According to the appellant, this is borne out by the statement of then-Minister for Education Mr Tharman Shanmugaratnam at the Second Reading of the Money-Changing and Remittance Businesses (Amendment) Bill (Singapore Parliamentary Debates, Official Report (15 August 2005) vol 80) on the objective of the MCRBA:

29 The appellant points further to the speech of then-Minister for Education Mr Ong Ye Kung at the Second Reading of the Payment Services Bill 2019 (Singapore Parliamentary Debates, Official Report (14 January 2019) vol 94), where a similar objective was raised concerning the four key risks arising from payment services, including money-laundering activities and terrorism financing:

30 Notwithstanding the above, I agree with the YIC that it would be hasty to conclude from this overlap of objectives that the precedents under s 6(2) of the MCRBA ought to be fully relevant for s 5(3) of the PSA. ⁹ There are five good reasons, canvassed by both the respondent and YIC, why the sentencing precedents under s 6(2) of the MCRBA have limited precedential value. I agree that these reasons are sound and adopt them accordingly.

31 First, sentencing precedents under s 6(2) of the MCRBA have a restricted scope of application. ¹⁰ The requirement for a valid licence under s 6(1) of the MCRBA extends only to persons carrying on a “remittance business”, which is in turn defined as out-bound remittances (see s 2(1) of the MCRBA). In contrast, 5(3) of the PSA covers a broader scope of seven categories of payment services as specified in Part 1 of the First Schedule of the PSA: ¹¹ (a) account issuance services; (b) domestic money transfer services; (c) cross-border money transfer services; (d) merchant acquisition services; (e) e-money issuance services; (f) a digital payment token services; and (g) money-changing services.

32 Second, s 6(2) of the MCRBA and s 5(3) of the PSA have different prescribed punishments. ¹² The broadening of the sentencing range under s 5(3) PSA translates into lower precedential value of the s 6(2) MCRBA precedents (see Kow Keng Siong, Sentencing Principles in Singapore (Academy Publishing, 2nd Ed, 2019) at [13.131]–[13.132]). The broader sentencing range under s 5(3) of the PSA can be seen in the following table helpfully prepared by the YIC: ¹³

33 Third, the s 6(2) MCRBA precedents lack clear reasoning to provide reliable guidance. As noted by the court in Public Prosecutor v Chen Jiantao (MAC 907914 of 2021) (unreported) (“Chen Jiantao”), which was cited in Public Prosecutor v Tan Khoon Yong [2022] SGMC 43 (“Tan Khoon Yong”) at [45] and[46], there is no reported High Court guidance on the appropriate sentencing framework for a s 6 MCRBA offence. As the YIC observes, of the 39 cases she has located from the SIRR which concern a sentence under s 6(2) of the MCRBA, there appears to be only one reported High Court decision (see Chinpo Shipping Co (Pte) Ltd v Public Prosecutor [2017] 4 SLR 983) and three reported District Court decisions (Shahabudeen; Public Prosecutor v Abdul Bashar Khan [2016] SGDC 203; Public Prosecutor v Ng Ah Ghoon [2020] SGDC 184) which set out the factual background and the court’s analysis on the sentence imposed on the offender. The remaining decisions are unreported and thus have little or no precedential value.

34 Fourth, the s 6(2) MCRBA precedents lack consistency. As the court in Chen Jiantao observed in its oral judgment at [2(c)], precedents after 2003 do not appear to have taken into account the two High Court cases of Public Prosecutor v Mohideen Kunji Mohamed Rafi (CR 16 of 2002) (“Mohideen”) and Public Prosecutor v Mohd Ziard Mohd Zarook (CR 17 of 2002) (“Zarook”) and the decision of DJ Audrey Lim (as she then was) in Shahabudeen. In those cases, custodial sentences were imposed on first-time offenders even though the monies remitted were not tainted, and no loss was caused to the customers.

35 Fifth, the actual sentences that have been imposed in s 6(2) MCRBA cases fail to utilise the full range of punishment prescribed even under s 6(2) of the MCRBA. ¹⁴ Nearly three-quarters of the sentences meted out in the past two decades under the MCRBA have congregated in the lower range of the sentencing spectrum which extends to two years’ imprisonment (Chen Jiantao at [2(b)]). No explanation is apparent from the cases for this trend.

36 Considering the limitations in the s 6(2) MCRBA precedents, they should accordingly be treated with caution. However, that is not to say that the precedents are devoid of relevance. The s 6(2) MCRBA precedents remain helpful insofar as they provide the relevant sentencing factors, both aggravating and mitigating, for consideration within the applicable sentencing framework. This is considered below at [74]–[78].

37 The appellant did not put forward any specific submissions as to whether a sentencing framework was necessary or desirable. Both the YIC and the respondent agree that this Court should establish a sentencing framework. ¹⁵ This follows from their position that the s 6(2) MCRBA precedents lack precedential value, which would leave sentencing courts with few reasoned decisions under s 5(3) of the PSA, which is of relatively recent vintage, to draw guidance from. Accordingly, this Court should take the opportunity to provide such guidance in the form of a sentencing framework. The YIC similarly argues that this is an opportune moment for this Court to lay down a sentencing framework for s 5(3) PSA offences as it will provide much-needed guidance for future sentencing courts. This will have the added benefit of not only ensuring that the full range of prescribed sentences can be utilised where appropriate but also lending consistency to future sentencing outcomes.

38 Given the limited guidance available from the s 6(2) MCRBA precedents as highlighted above (see above at [30]–[36]) and the paucity of sentencing precedents under s 5(3) of the PSA, I agree that it is opportune for this Court to establish a sentencing framework. A sentencing framework would provide structure and guidance for future sentencing courts and assist in the “quest for broad parity and consistency in sentencing” (Abdul Mutalib bin Aziman v Public Prosecutor and other appeals [2021] 4 SLR 1220 at [40]).

39 Before I turn to examine which sentencing framework is appropriate, I shall provide a brief outline of the possible sentencing frameworks based on current sentencing practice.

40 First, the “single starting point” framework. Here, the court begins with a single presumptive starting point for all cases before even considering the facts. From this starting point, the court makes appropriate upward or downward adjustments having regard to the aggravating or mitigating factors. The “single starting point” framework is suitable where the offence “almost invariably manifests itself in a particular way and the range of sentencing considerations is circumscribed”: Ng Kean Meng Terence v Public Prosecutor [2017] 2 SLR 449 (“Terence Ng”) at [28]).

41 Second, the “multiple starting points” framework. Here, the court establishes different indicative starting points, each corresponding to a different class of the offence. Once the court has established an indicative starting point by reference to the classification of the offence, the court then adjusts the sentence by reference to the aggravating and mitigating factors as the case may be: see Terence Ng at [29]. This was the approach in, for example, Vasentha d/o Joseph v Public Prosecutor [2015] 5 SLR 122, where the High Court held in the context of a drug trafficking offence that different indicative starting points would apply depending on the weight of the drugs trafficked. The “multiple starting points” framework is suitable where the offence is “clearly targeted at a particular mischief which is measurable according to a single (usually quantitative) metric that assumes primacy in the sentencing analysis” (Terence Ng at [30]).

42 Third, the “benchmark” framework. Under this framework, the court identifies or defines an archetypal case (or series of archetypal cases) and the sentences which should be imposed in respect of those case(s). This provides a focal point from which the sentences in the present case should take reference: see Terence Ng at [31] citing with approval Abu Syeed Chowdhury v Public Prosecutor [2002] 1 SLR(R) 182 at [15]. The “benchmark” framework is particularly suited for offences which “overwhelmingly manifest in a particular way or where a particular variant or manner of offending is extremely common and is therefore singled out for special attention” (Terence Ng at [32]).

43 Fourth, the “sentencing matrix” framework. Here, the court first considers the seriousness of an offence by reference to the “principal factual elements” of the case in order to give the case a preliminary classification. These are tabulated into a “matrix” wherein each cell features different indicative starting points and sentencing ranges: see for example Poh Boon Kiat v Public Prosecutor [2014] 4 SLR 892 (“Poh Boon Kiat”). Based on this assessment, the starting point will be identified. Then, at the second stage of the analysis, the precise sentence to be imposed will be determined having regard to other aggravating and mitigating factors that do not relate to the principal factual elements of the offence: see Terence Ng at [33] citing Poh Boon Kiat at [79]. The “sentencing matrix” framework is suitable for offences “crucially dependent on the availability of a set of principal facts which can significantly affect the seriousness of [the] offence in all cases” (Terence Ng at [34]).

44 Fifth, the “two-step sentencing band” framework introduced in Terence Ng. In the first step, the court first identifies under which band the offence in question falls within, having regard to the factors which relate to the manner and mode by which the offence was committed as well as the harm caused to the victim (ie, the offence-specific factors). Once the sentencing band has been identified, the court then determines precisely where within that range the present offence falls to derive an “indicative starting point”, which reflects the intrinsic seriousness of the offending act. In the second step, the court considers the aggravating and mitigating factors (ie., offender-specific factors) which are personal to the offender to calibrate the appropriate sentence for that offender. In exceptional circumstances, the court is entitled to move outside of the prescribed range for that band if, in its view, the case warrants such a departure. Although the courts have yet to articulate when precisely this framework is appropriate, Terence Ng at [35] suggests that the “two-step sentencing band” framework may be suitable where none of the other frameworks found above is suitable.

45 Sixth, the “five-step sentencing bands” framework introduced in Logachev for offences punishable under s 172A(1) of the Casino Control Act (Cap 33A, 2007 Rev Ed). Like the “two-step sentencing band” framework above, the courts have yet to articulate when precisely this framework is appropriate. However, the approach in Logachev suggests that this framework may be suitable where none of the above frameworks – including the “two-step sentencing band” framework – is suitable. The “five-step sentencing bands” framework involves the following steps:

(a) Step 1: Consider the offence-specific factors, and classify whether the level of harm was slight, moderate or severe; and whether the level of the offender’s culpability was low, medium, or high.

(b) Step 2: Identify the applicable indicative sentencing range according to the sentencing ranges set out in a matrix developed by the Court.

(c) Step 3: Identify the appropriate starting point within that indicative sentencing range.

(d) Step 4: From the starting point, make the appropriate adjustments to account for offender-specific aggravating and mitigating factors.

(e) Step 5: Consider whether there is need to make further adjustments in light of the totality principle, in cases where an offender has been convicted of multiple charges.

46 The YIC and the respondent diverge on the appropriate sentencing framework. On the one hand, the YIC suggests that the “five-step sentencing bands” framework adopted in Logachev is appropriate. This is because offences under s 6(3) of the PSA can arise from seven different categories of payment services, namely: account issuance, domestic money transfer, cross-border money transfer, merchant acquisition, e-money issuance, digital payment token, and money-changing services. Thus, there is no “one particular way” or an “archetypical case” in which these offences are manifested. According to the YIC, the “five-step sentencing bands” framework in Logachev lends itself well to the sentencing of offences under s 5(3) of the PSA as it would best accommodate the wide variety of factual circumstances in which the offences can be committed, respect the PSA’s objective of mitigating the key risks of payment services, and facilitate the analysis of the considerations of harm and culpability.

47 On the other hand, the respondent invites this Court to adopt the “single starting point” framework. This is because the offence of carrying on payment services business (particularly, money transfers or money remittances) without a licence generally manifests itself in a particular way. Furthermore, the key sentencing factors determining the gravity of the offence is limited.

48 As the YIC pointed out during the hearing, the respondent has, at various junctures, adopted the terminology of a “single starting point” and the “benchmark” framework interchangeably. In my assessment, there is considerable overlap and substantial similarity between these two approaches. It is not necessary for present purposes to determine whether the two frameworks are indeed identical or conceptually distinguishable. I accept, as the respondent proposed in its oral submissions, that both the “single starting point” and “benchmark” framework involve the identification of an archetypal case. The references are thus used interchangeably solely for convenience, and all subsequent references I make will refer only to the “single starting point”.

49 Having considered the parties’ submissions, I agree with the respondent that the “single starting point” framework is the most appropriate framework for s 5(3)(a) PSA offences relating to the unlicensed provision of money transfer services (ie., money remittance) by individuals. The present case only engages this type of payment service. As for the appropriate sentencing framework(s) for the other types of payment services under the First Schedule of the PSA, I agree with the respondent that this is best left for future consideration in a more appropriate case where those payment services are engaged on the facts.

50 Turning to elaborate on my analysis of the appropriateness of the “single starting point” framework, the essential question to be asked is whether two criteria are met: (a) the offence invariably manifests itself in a particular way and (b) the range of sentencing considerations is circumscribed (Terence Ng at [28]). I am satisfied that the archetypal offences under s 5(3)(a) of the PSA, as characterised by the respondent, fulfil both criteria.

51 On the first criterion of the offence invariably manifesting itself in a particular way, the respondent submits that the offence under s 5(3)(a) of the PSA of carrying on a payment services business (in particular, money remittance) without a licence generally manifests in situations where the offender knew or ought to have known that he needed a licence for his business when he carried on the business of providing payment services, specifically money remittance-type services, without a licence. ¹⁶ In contrast, the YIC takes the position that s 5(3) PSA offences do not arise in a particular way as they can arise in various factual circumstances. Specifically, such offences can relate to seven different categories of payment services, namely: account issuance, domestic money transfer, cross-border money transfer, merchant acquisition, e-money issuance, digital payment token, and money-changing services. The YIC further noted that various other considerations might need to be taken into account, including, among other things, whether tainted funds are involved, the total value involved, the period of offending and the amount of benefit to the offender.

52 The rationalisations offered by the respondent and the YIC merit careful consideration. The YIC justifiably points out the fact that s 5(3) PSA offences can take place in varied factual circumstances. Indeed, different forms of payment services are explicitly contemplated within the First Schedule of the PSA. This would appear to weigh in favour of the YIC’s submission that such offences do not “invariably manifest [themselves] in a particular way”, adopting the language in Terence Ng at [28]. However, the respondent puts forward an equally if not more cogent argument that focuses on the “knowledge” element of such offences instead, viz. that such offences will almost invariably be committed in factual circumstances where the offender knew or ought to have known of the requirement for licensing.

53 Crucially, it should be emphasised that the respondent suggests in addition that the proposed sentencing framework should be confined to the narrower scenario of unlicensed provision of money transfer services. I see no principled objection to the establishment of a “narrower” sentencing framework based on a “single starting point” as a consequence. This is not unprecedented. In Terence Ng at [32], the decision of the High Court in Wong Hoi Len v Public Prosecutor [2009] 1 SLR(R) 115 (“Wong Hoi Len”) was cited by the Court of Appeal to illustrate a case where the court had laid down the benchmark sentence for the specific offence of causing hurt to public transport workers, involving prosecution under s 323 of the Penal Code (Cap 224, 2008 Rev Ed) (“Penal Code”). Section 323 is of course capable of far broader application to a variety of factual circumstances. The Court of Appeal also cited Public Prosecutor v Fernando Payagala Waduge Malitha Kumar [2007] 2 SLR(R) 334 (“Fernando Payagala”) in which the benchmark sentence was laid down for the specific offence of credit card fraud, prosecuted under the general cheating provision contained in s 420 of the Penal Code. In both Wong Hoi Len and Fernando Payagala, the High Court did not attempt to lay down sentencing guidance pitched at any higher level of generality beyond the specific fact scenario at hand.

54 In my assessment, the unlicensed provision of money transfer services is likely to represent an extremely common particular variant of the manner in which s 5(3)(a) PSA offences are committed. More importantly, I agree with the respondent that in providing such services, the offender would typically know, or at least ought to know, that he does not possess a valid licence when committing the offence. Prior to the enactment of the PSA, there were already long-standing requirements for licences in relation to certain payment services (especially money remittance business) or where such services were provided by regulated entities. ¹⁷

55 In relation to the various other considerations the YIC raises (see above at [51]), they can be properly assessed at the stage where the court considers the aggravating and mitigating factors in adjusting the notional starting point identified for the archetypal case.

56 The YIC further argues that given the paucity of precedents in relation to s 5(3) PSA offences, it would be impossible at this juncture to ascertain with certainty whether the archetypal case arises where an offender either knows or ought to have known of the need for a licence. In my view, this is not fatal to the respondent’s argument. The “archetypal case” of unlicensed provision of money transfer services is framed as being one contingent on the mental state of the accused in that, at the very least, an accused person ought, in the ordinary event, to have known of the need for a licence. It is difficult to contemplate many situations where an accused can dispute that he ought to have known of the licensing requirements given the long-standing requirements for licences in relation to certain payment services.

57 There are few cases which might exceptionally not arise in this typical manner. An example would be where the offender commits an “unknowing” breach. ¹⁸ This can occur where, for instance, a PSA licence has lapsed because the licensed payment services operator inadvertently overlooks renewing the licence or paying the annual licence fee but nevertheless continues operating his business without falling foul of all the other requirements of the PSA. ¹⁹ In response to this example, the YIC pointed out at the hearing that the offender in this example ought to nonetheless have known of the licensing requirement such that it could not be considered an “unknowing” breach. While I can see the force of this argument, I am ultimately persuaded by the respondent’s submission that this can still be considered an example of an “unknowing” breach. I accept that had it not been for the offender’s genuine mistake, he would have been expected to be fully compliant with the applicable licensing requirement. In such circumstances involving a genuine mistake, the imposition of a custodial term would not be warranted.

58 On the second criterion of the range of sentencing considerations being circumscribed for offences under s 5(3)(a) PSA, I agree with the respondent that the range of sentencing considerations under s 5(3)(a) PSA tends to be circumscribed. In this regard, the key factors which determine the gravity of the offence are the knowledge of the need for a licence and the scale of the business operations. The latter factor would include the amount of profits generated and other related economic indicators.

59 At this juncture, I pause to deal with the YIC’s submission that the Logachev framework better facilitates the qualitative and contextual analysis of both harm and culpability that is required to assess the gravity of the offences. Arguably, in many instances, the “five-step sentencing bands” framework devised in Logachev may allow for more comprehensive consideration of the offence-specific and offender-specific sentencing factors through, for example, the assessment of harm and culpability at step one, sentencing ranges at step two, and adjustments of the starting point sentence based on the offender-specific factors at step four. However, the fact that various other sentencing frameworks have continued to be adopted by the courts suggests that they still have practical utility and relevance. The applicability of the different frameworks should thus continue to be determined with reference to the broad criteria set forth by the Court of Appeal in Terence Ng.

60 Furthermore, I am satisfied that the suitability of the “single starting point” framework to the offence here is supported by the Court of Appeal’s statement in Terence Ng at [28] that such a framework might be appropriate “for instance, where one is concerned with a regulatory offence”. In determining whether an offence is regulatory in nature, I concur broadly with the views of the District Judge in Public Prosecutor v Hamida Binti Sultan Abdul Kader [2021] SGDC 38 at [15]:

61 I find that an offence under s 5(3)(a) of the PSA would be properly characterised as a “regulatory” offence as it is concerned primarily with the prevention of money laundering or terrorism financing risks through such enforcement of minimum standards of conduct, being the procurement of licences in this context. Furthermore, as the respondent observed at the hearing, an offence under s 5(3)(a) is a strict liability offence. That being said, I should make it clear that this does not connote that such offences are somehow less serious or less deserving of disapprobation or that the sentencing norm should not deviate beyond the lowest end of the available sentencing spectrum.

62 I agree with the PDJ that general deterrence is the dominant sentencing principle for offences under s 5(3) of the PSA. This has been the consistent conclusion in previous cases dealing with offences under s 5(3) of the PSA (Tan Khoon Yong at [30]; Public Prosecutor v Zhu Yu [2022] SGDC 172 at [9]; Lange Vivian at [21]). Unlicensed providers of payment services that operate in the shadow financial system are unregulated and increase the risks of money-laundering and terrorism financing (“ML/TF”) brought about by new financial technologies. This is evident from then-Minister for Education Mr Ong Ye Kung’s Second Reading speech for the Payment Services Bill 2019 (Singapore Parliamentary Debates, Official Report (14 January 2019) vol 94):

63 As observed by the court in Tan Khoon Yong at [32]–[37], offences under s 5(3) of the PSA threaten the integrity and reputation of Singapore’s financial banking system, as payment services may be easily used for money laundering and terrorism financing due to the anonymous and borderless nature of the transactions. Payment services may also be easily used to facilitate money mule offences which are an essential component in the ecosystem of financial crime and have the potential to cause great harm. As the High Court held in Public Prosecutor v Law Aik Meng [2007] 2 SLR(R) 814 at [24(e)], general deterrence is warranted where offences affect the delivery of financial services and/or the integrity of the economic infrastructure and “[t]he courts will take an uncompromising stance in meting out severe sentences to offences in this category”.

64 Apart from the two criteria considered above, an additional consideration in favour of the “single starting point” framework is that it would better give effect to the intended legislative purpose of mitigating the key risks arising from unlicensed payment services. In particular, I agree with the respondent that a custodial term ought to be the general starting point for archetypal cases of individuals providing unlicensed money transfer services under s 5(3)(a) of the PSA. Unless there are compelling reasons to consider a fine, only a custodial term can mitigate the four key risks that Parliament has identified under the PSA by ensuring sufficient deterrence against such offences. These risks are: first, loss of customer monies; second, money-laundering and terrorism financing risks; third, fragmentation and lack of interoperability across payment solutions; and fourth, technology risks, including cyber security risks (see above at [29]). As I have noted above (at [62]), the provision of payment services without a licence must be strongly deterred to safeguard the integrity of Singapore’s financial institutions.

65 As the respondent submits, imposing fines on individuals may not be enough to achieve the necessary deterrent effect. ²⁰ Imposing fines as the standard punishment for s 5(3)(a) PSA offenders could create a moral hazard. Fines may become just another risk or cost of doing business, especially since running payment services without a licence could be seen as part of profit-driven activities. In such cases, potential financial gains could outweigh the cost of the fine, and an offender who stands to gain more than the amount of the fine may be less deterred. Furthermore, it would be difficult to determine an appropriate fine for unlicensed operators, given that they may not keep accurate records. As a result, the actual profits of an offender may be challenging to detect and/or calculate, making it difficult to impose a proportionately deterrent fine. The difficulty in detection is a consideration warranting stricter treatment by our courts. This can be seen, for example, in Public Prosecutor v Mihaly Magashazi [2006] SGDC 135 at [25] in relation to the fraudulent use of credit cards which involved the deception of financial institutions. Such crimes are often easy to commit but difficult to detect.

66 This is exacerbated by the nature of money transfer services (or money remittance services), where the benefits to an offender can fluctuate within a short span of time. For instance, in the context of cross-border money transfers, a significant part of the benefits would depend very much on the exchange rate, which fluctuates from day to day. The mere imposition of a fine would incentivise offenders to not keep proper records of their transactions, compounding the risks of money laundering and terrorism financing that Parliament sought to address with the PSA.

67 The present case illustrates the difficulties posed by the lack of complete and reliable records. While the appellant had, to his credit, admitted to offering unlicensed remittance services since 2018, he could only be prosecuted for offences spanning a period of about seven months, from 17 November 2019 to 28 June 2020, as that was the extent of the available records.

68 The sentencing trends under both the PSA and the MCRBA support the imposition of a custodial term as the starting point. A survey of the s 5(3)(a) PSA precedents show that a custodial sentence has generally been imposed. According to the SIRR, the majority of sentences (ie, 23 out of 24) imposed since 22 October 2001 have been imprisonment terms. Imprisonment was not imposed in only one unreported case where the court imposed probation for a term of 12 months.

69 Notwithstanding my earlier observation that sentencing precedents under s 6(2) of the MCRBA are of limited relevance and should be approached with caution, the imposition of a custodial term as the starting point is not inconsistent with precedents under s 6(2) of the MCRBA. As seen from the two High Court cases of Mohideen and Zarook, and the decision of DJ Audrey Lim (as she then was) in Shahabudeen, custodial sentences were imposed on first-time offenders even though the monies remitted were not tainted and no loss was caused to the customers.

70 Given that a custodial term ought to be the default sentence, I am satisfied that the “single starting point” framework is appropriate. In my view, this sends a clear and certain signal to would-be offenders that a custodial term would be imposed in the ordinary course. This, in turn, translates into a stronger deterrent effect. This echoes the reasoning of V K Rajah J (as he then was) in Fernando Payagala at [74] and [75], where he found that the need to deter credit card fraud would be more unequivocally conveyed through the imposition of imprisonment as a starting point for credit card cheating offences under s 420 of the Penal Code. Moreover, adopting the approach of imposing a notional starting point ensures greater consistency in the sentencing of such offences, while allowing for a proper consideration and assessment of the individual facts of the case (Fernando Payagala at [74]).

71 Finally, I would add that the imposition of a default custodial term would not be inconsistent with the need for a sentencing court to explore the full spectrum of sentences contemplated by Parliament: see Poh Boon Kiat at [60] citing Ong Chee Eng v Public Prosecutor [2012] 3 SLR 776 at [24]. Fines remain an appropriate sentencing option in exceptional cases, eg. where the offender commits an “unknowing” breach or where there are exceptional mitigating factors which justify a departure from the custodial benchmark. ²¹ Fines may also accompany the custodial term in cases where there is evidence of profit, in order to disgorge the criminal benefits of the offender (see Koh Jaw Hung v Public Prosecutor [2019] 3 SLR 516 at [44]). ²²

72 As I accept that the “single starting point” is the appropriate framework, two questions then arise as to the components of this framework. The first is the length of the custodial term which will serve as the starting point for an offender claiming trial. The second is the appropriate aggravating and mitigating factors that will feature in adjusting the starting point.

73 On the length of the custodial term, the starting point for an offender claiming trial should be three weeks’ imprisonment. This would not only be a sentence at the very lowest end of the sentencing range of custodial terms imposed in the precedents (which is up to three years’ imprisonment) but it would also fall within the middle of the lowest sentencing range (two to four weeks’ imprisonment) imposed in previous cases for offences under s 5(3)(a) of the PSA. The sentencing ranges for 23 precedent s 5(3)(a) cases drawn from the SIRR is summarised in the table below.

74 The next question which arises is in relation to the relevant aggravating and mitigating factors under the “single starting point” framework. The “single starting point” framework presumes a starting point for all cases before even taking into account the facts of the case at hand. From this starting point, the court makes appropriate upward or downward adjustments according to the aggravating or mitigating factors. The question that arises here is, therefore, the appropriate aggravating and mitigating factors that should feature in this framework.

75 In this respect, I am satisfied that the sentencing factors that have been helpfully identified by the YIC can serve as the aggravating and mitigating factors for the chosen framework. These factors are largely derived from Lange Vivian which has been endorsed by sentencing courts on s 5(3) PSA without contention. ²³ Slight modifications are made through the addition of the two factors of “benefit to the offender” ²⁴ and “offender’s role”. ²⁵ To elaborate on these factors, the “benefit to the offender” comprises both the commission or profit earned by the offender and any other kind of tangible or intangible benefit accruing to the offender. This would encompass a situation where there was an increase in customers to the offender’s other business units (related or unrelated) by reason of the unlicensed payment service provided to customers. Next, the “offender’s role” refers to the criticality of that role in the success of the enterprise. This is to be distinguished from the “level of sophistication” which looks at the specific methods used by the offender to carry out his role in the offence.

76 In dealing with sentencing for an offence of providing money transfer services without a licence, a sentencing court should take into account the following offence-specific factors. I note that these factors going towards harm and culpability would be considered aggravating factors.

77 As for the offender-specific factors, these have been derived from a review of the relevant precedents under s 5(3) of the PSA and s 6(2) of the MCRBA. ²⁶ As these factors are of general application, I accept that they are relevant sentencing factors.

78 As for the circumstances which would justify a fine rather than a custodial sentence, I accept the respondent’s submission that a fine would be warranted only in exceptional circumstances. Taking reference from the archetypal case of an individual providing money transfer services without a licence under s 5(3)(a) of the PSA, the critical factor in determining whether the custodial threshold is crossed is whether the person knew or, at the very least, ought to have known that he needed a licence (see above at [54]). Therefore, an example where a fine would be warranted may be where the offender commits an “unknowing” breach. As noted above at [57], an “unknowing” breach may occur where there is no basis to infer that the offender knew or ought to have known of the licensing requirements. One conceivable example is where a PSA licence has lapsed because the licensed operator, by virtue of a genuine mistake, inadvertently overlooks renewing the licence or paying the annual licence fee, but nevertheless continues operating his business without falling foul of all the other requirements of the PSA or exposing his client(s) to risk. In this example, a custodial term would not be warranted as it does not engage the PSA’s concern of “preventing an unregulated shadow financial system” (Lange Vivian at [14]).

79 I would emphasise that even if a relatively low sum was involved, this would not necessarily bring a case below the custodial threshold. The critical indicator remains whether the offender knew or ought to have known of the licensing requirements. Furthermore, the value of monies cannot be used as the primary factor in deciding whether the custodial threshold is crossed because it may not be possible to delineate a clear point where one offence involving a specified monetary value is more serious or aggravating than another.

80 As the respondent points out, assigning a threshold based on a monetary value can be arbitrary. Moreover, the amount of money involved in such an offence often cannot be accurately quantified due to incomplete or insufficient records. This means that when the matter is brought before the court, the monetary value involved is likely to be inaccurate. Indeed, this was the scenario in the present case since the appellant had not retained full records of his unlicensed remittance activities prior to the period specified in the charge. As the scale and scope of these activities could not be accurately ascertained, taking reference from the monetary value to determine whether the custodial threshold is crossed would be inappropriate. After all, it is by design that unlicensed money remittance businesses operate below the radar.

81 Having regard to the sentencing considerations identified above, I took into account the total value involved being $10,123.20. I note that this is not a particularly insignificant sum even though it is not exceedingly large.

82 Next, I considered the duration of offending which lasted 4.5 months for the proceeded charge. Further taking into account the TIC charge, this extends to about seven months. The appellant also admitted to providing unlicensed remittance services since 2018. The offence was not one-off or ad hoc, and thus this is a relevant aggravating consideration.

83 While there was a transnational element in the appellant’s employment of the hawala method in providing the cross-border money transfer services, such an element can be said to be inherent in the nature of the offence itself involving cross-border money transfers. The offence also required the cooperation of a foreign actor, but once again, this was perhaps an inevitable consequence of the offence itself.

84 On the amount of commission or profit, the facts indicate that the appellant only made a modest profit of $80 from the remittance service but also enjoyed further benefits from the sale of international calling cards to those who needed them to verify details of the remittance transactions. The quantum of such profits was not specified but it was unlikely to have been substantial. While I would not accord substantial weight to this as an aggravating factor, I should add that this was not a case where there was no benefit at all to the appellant. To be clear, even where there is no evidence of any profit, this would not carry mitigating weight but would only be a neutral factor in sentencing.

85 With the appellant’s employment of the hawala method, this resulted in the absence of any actual money flows between the appellant and his counterpart in Myanmar. The money trail was effectively severed once the appellant collected the monies. On balance, however, I am unable to accept that the offence was committed through particularly complex or sophisticated means. I would therefore not accord significant aggravating weight to this consideration.

86 In the round, I am of the view that the aggravating factors do not compel the imposition of a sentence at or above the proposed starting point sentence of three weeks’ imprisonment, which is calibrated based on an offender having been convicted after trial. There are no exceptional mitigating circumstances to warrant a non-custodial sentence. Taking into consideration the appellant’s plea of guilt and full cooperation with the authorities, I am satisfied that a sentencing discount should be awarded from the proposed starting point sentence of three weeks’ imprisonment. This results in a final sentence of two weeks’ imprisonment, corresponding to the sentence imposed by the PDJ below.

87  For the reasons above, I dismiss the appeal. To sum up, I conclude that:

(a) The s 6(2) MCRBA precedents are relevant insofar as they provide the relevant offence-specific and offender-specific factors for the sentencing framework chosen by this Court (see above at [28]–[36]).

(b) This is an opportune moment for this Court to establish a sentencing framework given the limitations in the s 6(2) MCRBA precedents and the paucity of sentencing precedents under s 5(3) PSA (see above at [37]–[38]).

(c) The appropriate sentencing framework is the “single starting point” framework. This is because s 5(3)(a) PSA offences involving the unlicensed provision of money transfer services invariably manifest in an archetypal way, the range of sentencing considerations under s 5(3)(a) PSA is circumscribed and there is a need for a custodial term as the starting point to achieve the legislative intent of the PSA in mitigating the risks inherent in payment services (see above at [49]–[71]).

(d) The starting point sentence should be three weeks’ imprisonment for cases involving offenders who are convicted after trial (see above at [73]).

(e) Finally, in applying the “single starting point” framework to the present case, a sentence of two weeks’ imprisonment is adequate and appropriate (see above at [81]–[86]).

88 In closing, I would like to commend the YIC, Ms Tai Ai Lin, for her diligent, thoughtful and well-researched submissions on the novel issues raised in this appeal. She offered illuminating and insightful perspectives which greatly aided the court's understanding of the case. It leaves me to convey my deep appreciation for her contributions.