1 In 2004, three major workplace accidents occurred in Singapore – the collapse of Nicoll Highway, the fire on the vessel Almudaina at Keppel Shipyard and the worksite accident at the Fusionopolis building. Collectively, 13 lives were claimed and numerous others were injured in these three workplace accidents. The series of accidents which took place within the short span of less than a year added greater impetus and urgency to the existing efforts to fundamentally reform workplace safety and health practices.

2 The efforts culminated in the enactment of the Workplace Safety and Health Act (Cap 354A, 2009 Rev Ed) (“the WSHA”), which covers the safety, health and welfare of persons at work in workplaces, in 2006. The WSHA seeks to create a strong culture of safety at workplaces, and requires the various stakeholders to take reasonably practicable steps to ensure the safety and health of workers and others affected by work.

3 Numerous prosecutions have taken place under the WSHA since its enactment in 2006 but no prior case has come before the High Court on appeal. In this appeal, the Prosecution seeks to enhance the sentence of a fine of $150,000 that was imposed by the learned District Judge on the respondent, GS Engineering & Construction Corp (“the Respondent”), for an offence under s 12(1), read with s 20 and punishable under s 50(b), of the WSHA. The Respondent had breached its duty as an employer to take necessary measures to ensure the safety and health of its employees at work insofar as this was reasonably practicable. As a consequence of this breach, two of the Respondent’s workers fell to their deaths from the seventh floor of the worksite while they were loading an air compressor onto an unsecured loading platform.

4  The Prosecution takes the position that the sentence is manifestly inadequate as a fine of at least $300,000 ought to have been imposed. On a broader level, it submits that the sentences imposed by the district court in previous cases are too low, with the majority falling below 30% of the maximum sentence of $500,000 prescribed by the WSHA (in respect of companies as opposed to natural persons). It argues that the sentences thus do not adequately uphold the statutory intent and the public policy concerns behind the WSHA. The Prosecution therefore submits that this appeal presents a useful opportunity for the High Court to set out a sentencing framework and provide guidance on the correct approach towards sentencing for such offences.

5 Having considered the parties’ submissions, I allow the appeal and increase the quantum of the fine to $250,000. While I agree that the sentence imposed by the District Judge is inadequate, I do not accept that a two-fold increase to $300,000 as submitted by the Prosecution is warranted. Looking at the sentencing precedents, I agree that the sentences hitherto imposed for this offence are on the low side; they do not adequately utilise the sentencing range prescribed by Parliament and do not have sufficient deterrent effect. Thus, in the course of this judgment, I will set out some sentencing considerations that are, in my view, relevant in dealing with such offences. These will hopefully offer some assistance and guidance for the lower courts in the exercise of their sentencing discretion.

6 I begin by setting out the brief facts of the fatal accident in question and the events that led to it. Unless otherwise specified, the facts are undisputed and are mostly extracted from the Statement of Facts (“SOF”), which the Respondent had admitted to.

7 The Respondent is a South Korean company in the business of civil engineering and general construction. It was engaged by Jurong Town Corporation as the main contractor to construct two towers (Tower A and Tower B) at Fusionopolis Way, Ayer Rajah Avenue. The two towers were to be 11 and 18 storeys high respectively. The Respondent sub-contracted the structural works of Tower A to another company, Zhang Hui Construction Pte Ltd (“Zhang Hui”). Zhang Hui was to supply labour, materials, tools, equipment and provide supervision for all formwork installation works. The project commenced on 23 November 2011 and was scheduled to be completed by 23 March 2014.

8 On 22 January 2014, at or about 12.30pm, an accident occurred at the worksite. Two of the Respondent’s employees, Mr Ratan Roy Abinash Roy (“the first deceased”) and Mr Rajib Md Abdul Hannan, died as a result. The two deceased persons had been working as construction workers for the Respondent since 23 November 2012 and 5 June 2012 respectively.

9 At the material time, the two deceased persons and three other employees of the Respondent were loading an air compressor onto a loading platform at the seventh storey of Tower A under the instructions of their foreman, Mr Nurun Novi Saydur Rahman (“Mr Nurun”), who was also employed by the Respondent. Instead of being securely installed, the loading platform was left suspended at the edge of the seventh storey of Tower A by a tower crane via four lifting chain slings. The air compressor rolled away from the edge of the building when it was loaded onto the loading platform, causing the platform to tilt. Consequently, the two deceased persons, who were standing on the platform and in the way of the air compressor, fell out along with it. The air compressor landed on another loading platform that was installed two storeys down, while the two deceased persons fell to ground level. They were pronounced dead by paramedics who arrived at the scene shortly after.

10 An investigation into the cause of the accident was carried out. It revealed that the Respondent’s employees were originally only scheduled to shift the loading platform from the tenth storey of Tower B to the eighth storey of Tower A. ¹ The loading platform, which was retractable, was used to facilitate the lifting of bulky materials and items from one part of the worksite to another. ²

11 The plan changed when an employee from Zhang Hui approached the Respondent’s site supervisor, Mr Miah Rashed (“Mr Miah”), to request help from the Respondent’s employees to move the air compressor using the loading platform. Mr Miah agreed and asked Zhang Hui to provide five additional workers to help to load the air compressor onto the loading platform. ³ Thereafter, Mr Miah instructed the foreman, Mr Nurun, to deploy a group of workers to first shift the loading platform from Tower B to the seventh storey of Tower A to load the air compressor before installing the loading platform at the eighth level of Tower A to facilitate Zhang Hui’s works there. Mr Miah however instructed Mr Nurun not to install the loading platform at the seventh storey of Tower A, but to simply suspend it by a tower crane.

12  At or about 11.50am that morning, Mr Nurun and his team of workers (which included the deceased persons) commenced the task of shifting the loading platform. There was no lifting supervisor present to oversee the lifting operation. ⁴ After the loading platform was shifted from Tower B to the seventh storey of Tower A, Mr Nurun asked Zhang Hui for the additional manpower that was promised. ⁵ Zhang Hui was unable to supply any workers as it was lunch time. The Respondent’s workers, led by Mr Nurun, decided to carry out the lifting of the platform along with the air compressor notwithstanding that.

13 The Respondent’s workers pushed the air compressor, which was mounted on a steel frame (fitted with two wheels at the front and a smaller wheel at the rear), onto the loading platform. But the small rear wheel could not be mounted onto the platform due to the height difference between it and the floor slab. In the process of trying to load the air compressor onto the loading platform, the loading platform started to tilt. At this point, the first deceased and the other co-workers informed Mr Nurun that it was unsafe to continue pushing the air compressor onto the loading platform but Mr Nurun told them to continue doing so. The workers then used a galvanised pipe to pivot the air compressor, and the two deceased persons positioned themselves in front of the air compressor in order to pull it onto the loading platform.

14 After several attempts, the workers finally succeeded in pushing the rear wheel onto the platform. Unfortunately, the air compressor started rolling towards the two deceased persons once it was mounted on the loading platform, causing the platform to tilt. Both the deceased persons could not move away in time and fell off the loading platform together with the air compressor. Neither of them was wearing a safety harness.

15 The Respondent was prosecuted for contravening s 12(1) of the WSHA in failing to discharge its duty to take measures, so far as it was reasonably practicable, to ensure the safety and health of its employees at work. It admitted that the two deaths could have been prevented had it done so. ⁶

16 Specifically, the Respondent admitted in the SOF to the following three breaches:

(a) failing to ensure that everyone involved in the lifting operation was trained to use the loading platform;

(b) failing to implement a safe system of work and ensuring, inter alia, that there was a permit-to-work or a lifting plan in place as required and that the risk assessment or safe work procedures were followed; and

(c) failing to provide fall protection equipment to its workers while they were working at height.

17 For completeness, I should point out that the SOF further sets out a fourth breach – that the Respondent had failed to ensure that there were no loose objects on the loading platform. While the Respondent had admitted to this breach as set out in the SOF, the District Judge held that the breach was not made out because the air compressor could not be regarded as “loose material” given that it rolled off immediately after it mounted the loading platform and further, there was no evidence that it was not going to be secured (at [31] of Public Prosecutor v GS Engineering & Construction Corp [2016] SGDC 89 (“the GD”)). As the Prosecution has not pursued this point on appeal, I will not address it any further save as to say that I share the District Judge’s view.

18 I will go on to briefly set out the details of the three breaches for two reasons. First, these details are material to the determination of the extent of the Respondent’s culpability, which is in turn crucial for sentencing. Second, there are some areas in dispute between the parties in respect of some of the breaches, notwithstanding that the Respondent had pleaded guilty and had admitted to a fairly comprehensive statement of facts. It may thus be helpful for me to set out the parties’ respective contentions, and thereafter my findings.

19 The Respondent admitted in the SOF that four of its workers who were involved in the lifting operation, including the first deceased, were not trained to install, dismantle or use the loading platform. ⁷ On appeal, it does not dispute that it had failed to ensure that all the workers were trained, but curiously, there is some suggestion in its written submissions (at para 41) that the first deceased was trained. In the light of its admission in the SOF that the first deceased was not trained and the absence of training records showing the contrary, for the purposes of the appeal, I disregard the Respondent’s present assertion that both deceased persons were trained. In any event, the fact remains that the Respondent admits that it had failed to ensure that all its workers were adequately trained before they were allowed to perform works involving the loading platform.

20  The SOF sets out four ways in which the Respondent had failed to implement and ensure a safe system of work.

21 First, the Respondent failed to ensure that there was a permit-to-work system in respect of the lifting of the loading platform. This was required under reg 11(1)(c) of the Workplace Safety and Health (Construction) Regulations 2007 (“the Construction Regulations”). The regulation in question specifies that the occupier of a worksite has to ensure, as far as reasonably practicable, that a permit-to-work system is implemented if any high-risk construction work is to be carried out. This is to ensure that (a) the high-risk construction work will be carried out with due regard to the safety and health of the persons involved; (b) the relevant persons are informed of the hazards that are associated with such work and the precautions that they have to take; and thereafter that (c) the necessary safety precautions are taken and enforced when the work is being carried out (per reg 11(2) of the Construction Regulations). An application for a permit-to-work system is usually made to a safety assessor, who would only issue the permit-to-work after conducting a site inspection and ensuring that all reasonably practicable safety measures are put in place (per reg 14 of the Construction Regulations).

22 The Respondent does not dispute that it did not apply for the requisite permit-to-work in respect of the lifting of the loading platform, which qualified as a high-risk task, and had therefore breached its duty under s 12(1) of the WSHA. In seeking to reduce its level of culpability, the Respondent argues in mitigation that Zhang Hui was “best placed” to do so. ⁸ On the material day, Zhang Hui had applied for a permit-to-work for lifting operations by crane and had done up a lifting plan for that purpose, but it did not include the lifting of the loading platform in either. ⁹ The second level of safety assessment of the permit-to-work was approved by the Respondent’s safety supervisor, and the permit itself was thereafter approved by the Respondent’s construction manager. ¹⁰ I address the Respondent’s argument on Zhang Hui’s role at [78] below in the section where I examine the Respondent’s culpability.

23 Second, the SOF also states that the Respondent failed to establish a lifting plan for the lifting of the loading platform, which involved the use of a crane. ¹¹ As with the permit-to-work system, the presence of a lifting plan is mandated by statute, specifically by reg 4 of the Workplace Safety and Health (Operation of Cranes) Regulations 2011 (“the Operation of Cranes Regulations”). Regulation 4 states that where any lifting operation involving the use of a crane is carried out in a work place by a crane operator, it is the duty of the “responsible person” to establish and implement a lifting plan that is in accordance with generally accepted principles of safe and sound practice. A proper lifting plan would have taken into account important information such as the details of the load that was to be carried, the lifting equipment and gears. Such a plan would have ensured that the operation zone was barricaded with warning signs and barriers, and that the sequence of lifting was planned. ¹² A “responsible person” is defined in reg 2 of the Operation of Cranes Regulations as either the employer of the person who operates the crane or the principal under whose direction the person operates the crane.

24 There appears to be some dispute as to who the “responsible person” was in this case. This is notwithstanding that the SOF clearly stated (at para 21) that “it [was] the duty of the [Respondent] to establish and implement a lifting plan for the safe lifting of the loading platform”. Yet, the Respondent seems to suggest (both in the proceedings below and on appeal) that Zhang Hui, and not it, bore this duty because Zhang Hui had full control of the crane that was used for this specific lifting operation. ¹³ Further, there is also a suggestion by the Respondent that this was not a case where no lifting plan was present, because a plan had been provided by Sante Machinery Pte Ltd, one of the suppliers of the loading platforms. ¹⁴

25 These two arguments that the Respondent has put forward seem to suggest that the Respondent is qualifying its admission in the SOF that it had breached its duty in not establishing a lifting plan. Alternatively, it could also be that the Respondent is simply saying in mitigation, as with its argument in respect of the permit-to-work, that Zhang Hui was in a better position than it to establish the requisite lifting plan; indeed, as with its mitigation in respect of the permit to work, the Respondent submitted that Zhang Hui was “best placed” to establish the lifting plan. ¹⁵ There is nonetheless some ambiguity.

26 Having reviewed the relevant portions of the SOF and the Respondent’s submissions, I conclude that the Respondent is not denying that it did not establish a lifting plan despite being obliged to do so. It is merely seeking to argue in mitigation that its culpability is lessened by the fact that Zhang Hui had a greater responsibility to ensure that the lifting plan was done. In any case, I am unable to see any reason why the Respondent should be permitted to qualify its previous admission to the SOF before the appellate court. I discuss Zhang Hui’s role and its consequent impact (if any) on the Respondent’s culpability at [78] below.

27 Third, the Respondent failed to implement the control measures that it had identified in the risk assessment and to carry out the safe work procedures that it had developed to address the risks involved in the lifting operation.

28 Prior to the accident and for the purposes of the project as a whole, the Respondent had conducted a risk assessment and had developed a set of safe work procedures for lifting operations involving different types of cranes. In its risk assessment, it identified the following control measures that had to be undertaken:

(a) a lifting supervisor had to be present to ensure and check that the correct rigging was performed and that the chain slings were secured;

(b) the lifting supervisor had to brief the workers involved in the crane operation on the risk assessment and safe work procedures before the start of the lifting operation; and

(c) the lifting supervisor and the site supervisor must periodically check that the hoisting operations were in order.

29 Similarly, the safe work procedures developed by the Respondent stipulated that all crane lifting operations had to be supervised by a qualified lifting supervisor. In addition, the site supervisor must apply for a permit-to-work for lifting operations and comply with and complete all the required checklists before the operation.

30 The Respondent did not carry out any of the control measures or the safe work procedures on the material day. ¹⁶ It neither applied for a permit-to-work nor arranged for a lifting supervisor to oversee the lifting of the loading platform. Mr Nurun, the foreman who was giving instructions to the workers, was not trained as a lifting supervisor. The workers involved in the lifting operation were also not briefed on the risk assessment and safe work procedures before they embarked on the task.

31 Fourth, the Respondent did not carry out any risk assessment or safe work procedure for the installation, use and dismantling of the loading platform (as opposed to lifting operations, which was discussed above). Having agreed to the SOF, the Respondent has admitted that it should have done so and should also have thereafter ensured that the relevant information was disseminated to all the workers who were involved. It also conceded that a proper assessment of the risks involved in the use of the loading platform would have highlighted that no loading should take place until the loading platform was installed and secured. ¹⁷

32 In respect of the final two points, the Respondent attempted to argue in the proceedings below that these were not entirely its fault for two reasons. It submitted first, that a lifting supervisor should have been provided by Zhang Hui and second, that it had a system in place to ensure proper supervision was present, but its lifting supervisor, Mr Miah, had detracted from it. This, it argued, was out of its control. On appeal, the Respondent does not explicitly make these arguments, but the general tenor of its submission is still that its culpability ought to be considered in the light that the accident was out of its control and was largely due to the actions or omissions of Zhang Hui and the Respondent’s workers on the ground. Again, as these issues go towards its level of culpability rather than the question of liability per se, I address them at [78] below, together with the other similar arguments that I have alluded to earlier.

33 The final breach was that the Respondent had failed to ensure that all the workers wore safety harnesses that were anchored securely before they started working at height. None of the workers, including the two deceased persons, was wearing any fall protection equipment. The Respondent did not issue safety harnesses to them on the day of the accident. The Respondent also admitted in the SOF that in any event, even if they had worn them, there would have been nowhere for the workers to anchor their harnesses.

34 While its admission by way of its agreement with the SOF appeared to be unequivocal, the Respondent again tried to downplay its level of culpability in respect of this breach in its submissions. It argues in mitigation that it could not really be faulted because it had issued security harnesses to all its workers at the beginning of the project (though not specifically on the material day), and further, it had periodically engaged external training providers to conduct on-site refresher training to remind the workers to use the safety harnesses. ¹⁸ The short point of its argument is that it cannot be blamed, or at least not fully so, for its workers’ election not to wear the safety harnesses. Further, there was also some suggestion that Zhang Hui should have been the one to provide the lifeline or anchorage point. ¹⁹

35 The District Judge imposed a fine of $150,000 for the offence, with an order of attachment to be issued in default. The Prosecution submitted for a fine of $300,000, while the Respondent submitted for a fine of $100,000.

36 The District Judge began by stating that he did not agree with the Prosecution’s submission that the courts should impose higher sentences just because the maximum punishment prescribed by law for such offences were increased by Parliament with the enactment of the WSHA. He was of the view that an increase in the prescribed punishment does not necessarily equate to an increase in the sentences that should be imposed as ultimately, the important consideration is still the offender’s culpability. In particular, he did not seem to think that the Parliamentary debates relied on by the Prosecution supported its submission. He was also not persuaded by the Prosecution’s submission that the existing sentences for WSHA offences were too low. He found the Prosecution’s argument in this regard to be “neither here nor there” as the Prosecution, in not appealing the earlier sentences, seemed to have accepted that the sentences imposed in those cases were adequate.

37 The District Judge was also not convinced by the Prosecution’s next argument that he should impose a high deterrent fine because of a rise in workplace fatalities. Two reasons led him to conclude that the statistics submitted by the Prosecution should be considered “with a pinch of salt”. First, while the Prosecution’s statistics pointed to a rise in the fatalities, the statistics produced by the defence showed a contrary picture. Second, no information on how many of the fatalities were due to breaches of the WSHA was provided.

38 Applying the framework set out in the English Court of Appeal case of R v F Howe & Sons (Engineering) Ltd [1999] 2 All ER 249 (“R v Howe”) which dealt with similar offences under the English Health and Safety at Work Act 1974 (Cap 37), the District Judge considered three broad categories of factors—mitigating factors, aggravating factors and other relevant factors—in determining the appropriate fine. In respect of mitigating factors, the District Judge took into account the Respondent’s early plea of guilt, the fact that it had conducted an investigation and taken measures after the accident to ensure that there would not be further accidents of a similar nature, as well as its good safety record as evidenced by the awards that had been given by agencies such as the Ministry of Manpower (“the MOM”). The District Judge regarded the fact that the Respondent’s lapses caused two deaths as aggravating, though he did not think that this was a dominant factor.

39 The District Judge went on to consider the Respondent’s culpability. He did not consider the Respondent’s breaches to be as egregious or its culpability as high as that submitted by the Prosecution. This was partly because the District Judge accepted that Zhang Hui played a major part and its actions and omissions had contributed to a “perfect storm”. He accepted that Zhang Hui was in the best position to apply for a permit-to-work for the tasks as well as to establish a safe work procedure or to do a risk assessment. Further, he did not think that the Respondent had been overly lax in safety, even though lapses were present. To his mind, a fine of $150,000 was appropriate as it would achieve a deterrent effect and yet, at the same time, be proportionate to the severity of the offence and the culpability of the Respondent.

40 The Prosecution submits that the sentence imposed by the District Judge is manifestly inadequate, and that a fine of at least $300,000 ought to have been imposed. It makes the following arguments in support of its position:

(a) First, the District Judge failed to give sufficient weight to the legislative intent behind the increase in the maximum punishment for such offences from $200,000 under the Factories Act (Cap 104, 1998 Rev Ed) (“the Factories Act”)—the predecessor of the WSHA—to $500,000 under the WSHA.

(b) Second, the District Judge failed to give due consideration to the full range of sentences available to him.

(c) Third, the District Judge failed to give sufficient weight to the principle of deterrence.

(d) Fourth, the District Judge downplayed the Respondent’s culpability.

(e) Fifth, the District Judge erred in law and fact in giving excessive weight to the role of Zhang Hui.

41 On a broader level, the Prosecution submits that this is an area of sentencing that would greatly benefit from a review and invites this court to set out a sentencing framework. It argues that the sentences imposed for such offences have not been consistent, and are, in general, too low. In essence, it submits that the current sentencing practice for such offences falls foul of the same three criticisms as summarised at (a) to (c) of the preceding paragraph. In its submissions, the Prosecution proposed a detailed sentencing framework which is premised on two principal factors: the culpability of the offender and the harm that would potentially result from the offender’s actions.

42 The Prosecution’s submissions fall into two categories: one, on a broader level, concerns the question whether there is a need to review the current sentencing practice, and the other, on a specific level, relates to the particular factual matrix of this case. The first three submissions set out at [40] above come within the former category while the remaining two submissions fall into the latter. I begin by addressing the first category of submissions.

43 The Prosecution submits that the present sentencing practice should be reviewed because the sentences (a) fail to give sufficient effect to the legislative intent of the WSHA; (b) do not utilise the full spectrum of sentences; and (c) do not have sufficient deterrent effect, having regard to the increase in workplace fatalities over the years. I discuss each in turn, though they are related and largely overlap.

44 There can be no dispute over the Prosecution’s submission that the legislative intent of a statutory provision is relevant, and indeed important, for sentencing. This is well-established, as can be seen from the observations of the High Court in cases such as Mehra Radhika v Public Prosecutor [2015] 1 SLR 96 (“Mehra”) (at [27]) and Public Prosecutor v Pang Shuo [2016] 3 SLR 903 (at [14]).

45 The Respondent takes issue with the Prosecution’s submission that the benchmark sentences imposed for such offences should have been increased when the maximum penalty for this offence was more than doubled with the enactment of the WSHA. It is common ground that the sentences for such offences have not increased since the enactment of the WSHA in 2006, but the Respondent argues that there is nothing wrong with this because the courts are not obliged to impose higher sentences simply because the maximum prescribed punishment has been increased. This was also the finding of the District Judge (as summarised at [36] above), who further took the view that the Prosecution must be taken to have “accepted” that the sentences imposed in the precedents were correct given that it did not appeal against the sentences.

46 In my view, there is in fact little difference between the parties’ respective positions (and the District Judge’s finding). Ultimately, whether the court should correspondingly increase the sentences it imposes when the maximum prescribed punishment for any particular offence is increased is dependent on the rationale and intention behind that legislative amendment.

47 The increase in the maximum prescribed punishment may, in some cases, be due to Parliament’s view that the sentences for that offence need to be higher, either because the prevailing benchmarks are too low or because there has been an upward trend or an increased prevalence in that offence. In such cases, the underlying rationale is that the existing maximum prescribed punishment appears to be inadequate and a wider sentencing range with a higher upper limit is warranted. The courts should then take cognisance of such considerations in sentencing. In this connection, Sundaresh Menon CJ made the following observations in Mehra (at [27]):

48 In other cases, however, the increase may be for an entirely different reason. Take for instance, the offence of culpable homicide not amounting to murder under s 304(a) of the Penal Code (Cap 224, 2008 Rev Ed). In 2008, the maximum determinate imprisonment sentence for this offence was increased from ten years to 20 years. The rationale of this increase was not because Parliament had viewed the offence with increased severity. The amendment was so as to bridge the gulf between the then-maximum determinate punishment of ten years and the term of life imprisonment. It was to accord the courts with a greater range of discretion in calibrating the sentences according to the facts of each case and to deal with the culpability of the offender in a more nuanced fashion (see the observations of the Court of Appeal in Public Prosecutor v Vitria Depsi Wahyuni (alias Fitriah) [2013] 1 SLR 699 at [22] – [23]). Another example of this is the amendment to increase the maximum sentence for an offence of giving false information to a public servant under s 182 of the Penal Code (see Koh Yong Chiah v Public Prosecutor [2016] SGHC 253 at [20]).

49 The District Judge is thus correct in saying that that an increase in the punishment prescribed by law does not inexorably or “necessarily” equate to a need for an increase in sentences imposed. In the latter category of cases, no such corresponding increase in sentences is required. The real question is thus whether Parliament, in increasing the maximum penalty for this offence from $200,000 to $500,000 (in relation to companies), intended for higher sentences to be imposed for greater deterrent effect.

50 In this regard, I would respectfully disagree with the District Judge’s view (at [33] of the GD) that this was not the intention of Parliament. The relevant Parliamentary debates, coupled with the background to the enactment of the WSHA, reflect such an intention. As I stated at the introductory paragraph, the WSHA, which replaced the Factories Act, was enacted following the government’s review of workplace safety regulations after three high-profile accidents in 2004. The legislative intent of the WSHA was clearly expressed by the then-Minister for Manpower, Dr Ng Eng Hen, at the second reading of the Workplace Safety and Health Bill (Bill 36 of 2005) (“the Bill”), where he said as follows:

In a later portion of the same speech, Dr Ng elaborated on the rationale behind the introduction of higher penalties for poor safety management and performance:

(Singapore Parliamentary Debates, Official Report (17 January 2006) vol 80 at cols 2206 and 2215).

51 It is discernible from the above extracts that the introduction of more severe penalties for such offences in the WSHA was part of a concerted effort to deter poor safety management and effect a cultural change for employers and other stakeholders to take proactive measures to prevent accidents at the workplace. The necessary implication is that Parliament’s intention was for the courts to impose higher penalties, where appropriate, in order to achieve such a deterrent effect and ensure that the true economic and social costs of such risks and accidents are borne by the responsible parties. Insofar as the District Judge and the Respondent are saying that the culpability of the offender is important and relevant, this is uncontroversial and is undoubtedly correct. But this does not detract from the fact that it was Parliament’s intent, as evidenced from the extracts from the Parliamentary debates during the second reading of the Bill, that more severe penalties are warranted for such offences and that the courts ought to thus have taken heed of this and to have acted in tandem in sentencing.

52  On a related note, the Prosecution submits that the sentences that have thus far been imposed for such offences do not utilise the range of penalties available. The maximum prescribed punishment for such offences is a fine of $500,000 for a first-time offender and a fine of $1m for a repeat offender, where the offender is a company. In support of this submission, the Prosecution points to the fact that the large majority of sentences imposed falls below 30% of the maximum penalty of $500,000, and that at the time of sentencing in this particular case, no penalty of above $200,000—this being the maximum penalty under the old Factories Act regime—had even been imposed. The district court had since imposed a fine of $220,000 on a first- time offender in Public Prosecutor v Dawn Plastic Industries (DSC 900133 of 2016) (“Dawn Plastic”) on 21 June 2016.

53 As observed by Menon CJ in Poh Boon Kiat v Public Prosecutor [2014] 4 SLR 892 at [60] and more recently in Janardana Jayasankarr v Public Prosecutor [2016] 4 SLR 1288 (“Janardana”) at [21], the maximum sentence that is stipulated for an offence signals the gravity with which Parliament views that offence. A sentencing judge ought therefore to take this into account when determining precisely where the offender’s conduct falls within the entire range of punishment devised by Parliament. As in the case of Janardana which involved the offence of violence against a domestic helper, this assumes particular importance because Parliament had specifically acted to enhance the sentencing powers of the courts in the case of such offences.

54 In Ong Chee Eng v Public Prosecutor [2012] 3 SLR 776, Chao Hick Tin JA sounded the following caution (at [24]):

The Prosecution submits that this is exactly what has happened in this case. ²⁰ It argues that this is borne out by the fact that notwithstanding that a decade has passed since the enactment of the WSHA, a large majority of the sentences falls below 30% of the maximum penalty. The second point that the Prosecution raises in support of this submission is even stronger. This is that in Public Prosecutor v L & M Foundation Specialist (MOM Summons No 1258 of 2010) (“L & M”), a repeat offender had only been sentenced to a fine of $160,000 for a breach of its duty under s 12 of the WSHA, which had resulted in the death of an employee. The offender had previously—just slightly more than a year ago—been fined $80,000 for a similar breach which had also resulted in a death. As submitted by the Prosecution, it is apparent that the sentence of a fine of $160,000 is not only vastly below the maximum punishment for a repeat offender (which is $1m), but is also vastly below the maximum punishment for a first-time offender. There is also merit to the Prosecution’s submission that s 51 of the WSHA, which prescribes enhanced penalties for repeat offenders causing death, would be rendered otiose if the courts consistently sentence first-time offenders to the lower end of the sentencing range as in that case, the enhanced range of between $500,000 and $1m would not be utilised. ²¹

55 Having examined the sentencing precedents, I agree with the Prosecution that the present sentencing benchmarks for such offences do not sufficiently utilise the available sentencing range. To be clear, I accept the Respondent’s point that the maximum sentence for a case is reserved for the worst type of cases falling within the prohibition and that in the case of the present offence, this would be something close to the scale of the Nicoll Highway collapse. ²² Fortunately, the cases that have come before the courts are not anywhere near that end of the spectrum. But this does not assist the Respondent or justify why the sentences imposed thus far are all concentrated within the lower end of the spectrum; it merely justifies why sentences at the higher end (say, between $400,000 and $500,000) have not been imposed.

56 The third argument raised by the Prosecution in support of its submission that the sentencing regime for such offences is in need of review is that the sentences imposed thus far fail to give sufficient weight to the principle of deterrence. In the context of such offences, this argument is closely linked to, and is a consequence of, the two points that have been addressed.

57  Insofar as the Prosecution’s argument is that the sentences that have been imposed thus far do not have sufficient deterrent effect, I agree with it. But if it is seeking to argue further, that there is, in recent years, an increased need for deterrence following the rise of workplace fatalities, I am not wholly convinced from the material placed before me that there is a basis to come to such a conclusion. As pointed out by the District Judge, the statistics submitted by the Prosecution provide limited assistance. First, some parts of it are directly countered by another set of statistics that was tendered by the Respondent. Second, it is unclear if the recorded fatalities were the result of conduct that was in breach of the regulations under the WSHA. Further, I also do not think it is entirely appropriate for me, in deciding if the District Judge’s sentence should be altered, to take into account statistics or statements made by the Prime Minister and the present Minister for Manpower after the date the sentence was meted out. Some of the statistics and documents that the Prosecution relies on at the appeal fall within this category. In any event, as I had intimated during the appeal hearing, I do not think we should be unduly fixated on the interpretation of those statistics. What is important for our present purposes is that fatalities and injuries arising from any workplace accident should be avoided as much as possible, and that the sentences imposed for such offences should be capable of achieving the intended deterrent effect.

58 For the above three reasons, I agree with the Prosecution that the sentences that have been imposed for such offences in previous prosecutions are too low and that the sentencing regime ought to be reviewed. For future cases, the sentencing court should bear in mind that the legislative intent for the introduction of more severe penalties was to create a more palpable deterrent effect and encourage stakeholders to take more proactive measures to minimise the occurrence of accidents at workplace. The court must also take into consideration the range of available penalties, and calibrate the sentences in accordance with factors such as the offender’s culpability and the severity of the offence.

59 Before leaving this issue, I make a further observation. This relates to the point made by the Prosecution that the fact that it did not appeal against the sentences previously imposed does not mean that the correct sentences had been imposed. ²³ This point was made in response to the District Judge’s comment in the GD (at [45]) that the Prosecution’s submission that the present sentencing benchmarks do not satisfy the legislative intent of increasing penalties is “neither here nor there” because the Prosecution, in not appealing, seemed to have accepted that the sentences previously imposed were adequate.

60 It is ultimately the court’s role to assess and determine the appropriate sentence in the light of all the circumstances of the case (see the observations of Menon CJ in Janardana at [12]) and the mere fact that no appeal has been filed by either party does not mean that a particular sentence that was imposed is correct. But a natural and reasonable inference would be that the outcome was deemed acceptable to both parties such that they saw no necessity to seek the appellate court’s intervention. With respect, it appears logically incongruous and perhaps not entirely fair for the Prosecution to distance itself from its role in assisting the court in the sentencing process; higher sentences in all the precedents cited might otherwise have been imposed had the appropriate sentencing submissions been made (or appeals filed). The fact remains that the sentencing trend persisted over the years partly because the Prosecution had neither appealed nor, it would seem, addressed the court to seek higher sentences prior to the most recent cases (ie, the present case and Dawn Plastics).

61 I note that the Prosecution has highlighted L & M (see [54] above) as an instance where the full sentence range was not utilised. It now seeks to argue with the benefit of hindsight that the sentence imposed on a repeat offender was manifestly inadequate in that case. But the Prosecution had not deemed it necessary to lodge an appeal against sentence at the material time. It would also appear that there was no submission seeking a higher sentence. I accept of course that prosecutorial perspectives and policies can and do change over time, and perhaps a different view has now prevailed. Ultimately, I am grateful that the Prosecution has chosen to grasp the nettle and raise these matters for an appellate court to consider.

62 With that, I turn to consider the appropriate sentencing guidelines for such offences.

63 The Prosecution submits that as a starting point, the court should determine the severity of the offence based on the principal factual elements of the offence, which, it submits, would in this case be:

(a) the culpability of the offender; and

(b) the harm that may potentially result from the offender’s actions.

It submits that this is consistent with the legislative intent as seen from the Parliamentary debates as well as with the conceptual approaches of other jurisdictions, in particular the United Kingdom (“the UK”) as seen from its sentencing guidelines, ²⁴ towards similar offences.

64 The Prosecution submits that an inquiry into the offender’s culpability will require a holistic assessment of all the circumstances of the case, in particular the nature and circumstances surrounding the breach. It suggests that the following non-exhaustive list of factors would be useful for assessing an offender’s culpability: ²⁵

(a) the number of breaches or failures in the case;

(b) the nature of the breaches;

(c) the seriousness of the breaches – whether they were a minor departure from the established procedure or whether they were a complete disregard of the procedures;

(d) whether the breaches were systemic or whether they were part of an isolated incident; and

(e) whether the breaches were intentional, rash or negligent.

65 As for the second suggested factor for the starting point, the Prosecution submits that potential harm rather than actual harm should be considered because it is clear from the Parliamentary debates that the WSHA criminalises the creation of the risk of harm rather than actual harm. This is most obviously seen from Dr Ng’s statement during the second reading of the Bill that companies or persons who show poor safety management would and should also be penalised even if no accident has occurred (as set out at [50] above). Most likely taking reference from the UK sentencing guidelines, the Prosecution submits that the following non-exhaustive list of factors would be useful for assessing the potential harm resulting from an offender’s breach:

(a) the seriousness of the harm risked; and

(b) the likelihood of that harm arising.

66 The Prosecution goes further than simply setting out the relevant factors that ought to be considered in deriving a starting point. In its submissions, it proposed specific starting ranges depending on the level of culpability and potential harm for cases where the offender claimed trial:

The Prosecution suggests that the figures in the brackets for the respective ranges should be the starting points where deaths were resulted.

67 The Prosecution submits that after determining the starting point, the court should next consider the aggravating and mitigating circumstances present. Again taking reference from the UK sentencing guidelines, the Prosecution suggests that the following could be regarded as aggravating factors:

(a) serious actual harm resulted;

(b) the breach was a significant cause of the harm that resulted – in this regard, it submits that a significant cause need not be the sole or principal cause of the harm, and need only be a cause that has more than minimally, negligibly or trivially contributed to the outcome;

(c) the offender had cut cost at the expense of the safety of the workers;

(d) there was deliberate concealment of the illegal nature of the activity;

(e) there was a breach of a court order;

(f) there was an obstruction of justice;

(g) the offender has a poor record in respect of workplace health and safety;

(h) there was falsification of documentation or licences; and

(i) there was a deliberate failure to obtain or comply with relevant licences in order to avoid scrutiny by authorities.

68 It suggests the following as a non-exhaustive list of mitigating factors:

(a) the offender has voluntarily taken steps to remedy the problem;

(b) the offender provided a high level of cooperation with the authorities for the investigations, beyond that which is normally expected;

(c) there is self-reporting, cooperation and acceptance of responsibility;

(d) there is a timely plea of guilt;

(e) the offender has a good health and safety record; and

(f) the offender has effective health and safety procedures in place.

69 I agree substantially with the general sentencing framework and the list of potential aggravating and mitigating factors that the Prosecution has proposed. These provide helpful guidance in determining the appropriate sentence.

70 However, I do not agree with two of the proposed sentencing ranges, and will adjust the table (which sets out the starting sentencing ranges depending on the culpability of the offender and the potential for harm in a case involving an offender who has claimed trial) in the following manner:

While I have set out the various sentencing ranges from which the starting point for the sentence can be derived, I should caveat that these are merely guides and should not be taken to be rigid and inflexible. The upper and lower limits in the ranges, while meant as guides, should not operate as constraints if there is reason to depart from them. Further, the sentencing ranges may be further revised if necessary as we build up a corpus of sentencing precedents henceforth.

71 In this table, I have omitted the figures in brackets that appear in the Prosecution’s table at [66] above, which had been proposed as the specific starting points that should be adopted where death has resulted. I do not think that it is appropriate or necessary to stipulate starting points to such a degree of specificity for the following reasons.

72 In the Prosecution’s own words, s 12 of the WSHA “allows for a vast range of actions (or omissions) to be caught” and “the number and nature of safety breaches could vary across different offenders”. There are myriad ways in which a workplace accident could occur, and even more ways in which breaches of duty could be committed. The harm that could potentially result can also take many varied forms. The wide variety of factual scenarios of misconduct and the range of possible consequences make it difficult, though not impossible, to set out sentencing starting points with precision. The attendant risk is that these starting points may become too readily applied as the first thought and anchors over time such that they become rarely (if ever) departed from.

73 Further, this difficulty is exacerbated by the fact that the sentencing precedents accumulated over the past decade are, as I have found above, of little, if any, assistance to us. This distinguishes the present case from other cases where the court is able to rely on a body of case precedents which reveals a sentencing trend to derive the specific starting points for sentencing or to provide an illustration and guidance of what constitutes a certain level of culpability or severity. There may well be a right time—after a corpus of precedents have been built up following the present case which would hopefully help to re-focus the courts—for the courts to revisit if there is a need to set out a more detailed framework with specific starting points for sentences. In my view, this is not the right case to do so.

74 Moreover, I do not think there is a critical need to set out specific starting points to this level of detail. While I agree that the sentencing benchmarks for such offences ought to be reviewed, this does not necessarily entail the imposition of prescriptive starting points. In my view, the broad guidance and principles as well as the sentencing ranges set out in this case would be sufficient to guide and re-focus the sentencing inquiry for such offences. I am confident that the lower courts, with the guidance offered in the present case to utilise the sentencing range and to give effect to the legislative intent, would be able to fairly and judiciously exercise their discretion and impose appropriate sentences in future cases. In my view, it is important to leave room for sufficient flexibility in the sentencing guidelines that are laid down.

75 Most importantly, it would also be inconsistent with the Prosecution’s own framework to take into account the issue of whether death had been caused at the first stage of the inquiry. In determining the starting point, we are concerned only with the culpability of the offender and the harm that may potentially result. Whether death had been caused is a key consideration when we look at the harm that actually resulted. This should feature only at the second stage of the inquiry when we turn to examine the aggravating factors of the case.

76 Therefore, while I agree with the Prosecution’s proposed approach and the sentencing ranges as set out at [63]-[68] above, I decline to adopt its proposal in respect of fixed starting points for sentencing in cases where death has resulted.

77 To summarise, in deciding on the appropriate sentence to impose in such cases, the court should be guided by the following:

(a) The court should first determine the appropriate starting point by considering two principal factors: (i) the culpability of the offender; and (ii) the harm that could potentially have resulted. The court should generally take reference from the sentencing ranges set out in the table at [70] above in deriving the specific starting point.

(b)  The culpability of the offender may be dependent on the following non-exhaustive list of factors: (i) the number of breaches or failures in the case; (ii) the nature of the breaches; (iii) the seriousness of the breaches – whether they were a minor departure from the established procedure or whether they were a complete disregard of the procedures; (iv) whether the breaches were systemic or whether they were part of an isolated incident; and (v) whether the breaches were intentional, rash or negligent.

(c) The potential harm may be assessed by considering, among other things, (i) the seriousness of the harm risked; and (ii) the likelihood of that harm arising.

(d) Next, after deriving the starting point for sentencing, the court should calibrate the sentence by taking into account the aggravating factors and mitigating factors of the case.

(e) Aggravating factors include the following: (i) serious actual harm (including death) resulted; (ii) the breach was a significant cause of the harm that resulted – in this regard, a significant cause need not be the sole or principal cause of the harm, and need only be a cause that has more than minimally, negligibly or trivially contributed to the outcome; (iii) the offender had cut cost at the expense of the safety of the workers; (iv) there was deliberate concealment of the illegal nature of the activity; (v) there was a breach of a court order; (vi) there was an obstruction of justice; (vii) the offender has a poor record in respect of workplace health and safety; (viii) there was falsification of documentation or licences; and (ix) there was a deliberate failure to obtain or comply with relevant licences in order to avoid scrutiny by the authorities.

(f) Mitigating factors may include the following: (i) the offender has voluntarily taken steps to remedy the problem; (ii) the offender provided a high level of cooperation with the authorities for the investigations, beyond that which is normally expected; (iii) there is self-reporting, cooperation and acceptance of responsibility; (iv) there is a timely plea of guilt; (v) the offender has a good health and safety record; and (vi) the offender has effective health and safety procedures in place.

78 The Prosecution submits that the Respondent falls within the “high” culpability category because it had failed to put in place several measures that were recognised by the industry and mandated by law. The Respondent, on the other hand, submits that the District Judge had given sufficient weight to its level of culpability and had correctly found that its breaches were not so egregious and that Zhang Hui was in several ways the more responsible party instead. It is also suggested in some parts of the Respondent’s submission that the workers on the ground had to also bear some responsibility for not following the protocol.

79 In my view, the District Judge had downplayed the Respondent’s culpability by according too much responsibility to Zhang Hui and by not recognising that several of the breaches were serious in nature. I have some sympathy for the Respondent in that the fatal accident would not have occurred had its workers, in particular Mr Miah, not spontaneously agreed to do Zhang Hui a favour and move the air compressor.

80 The fact remains, however, that the Respondent was the overall occupier (and thus overall in-charge) of the worksite, and was the employer of the deceased persons and other workers performing the lifting operation. Even more significantly, it must be noted that most, if not all, of the breaches involved the failure to perform safeguards and actions that would have to be undertaken for the Respondent’s own lifting operation and even if it did not have to move the air compressor for Zhang Hui. Despite the fact that it was scheduled to carry out the lifting operation that day, the Respondent did not acquire the necessary permit-to-work, did not implement the control measures that it had identified in the risk assessment or carry out the safe work procedures that it had developed for the lifting operation. It also did not carry out any risk assessment or safe work procedure for the installation, use and dismantling of the loading platform. Further, it would have remained its responsibility to ensure that the workers carrying out the lifting operation—even if we leave out the task of moving the air compressor—were all trained and were all wearing safety harnesses.

81 Moreover, even if we take the Respondent’s case at its highest and accept that it was Zhang Hui’s responsibility to ensure that the lifting operation and the moving of the air compressor were included in the permit-to-work system, the Respondent would still have failed to discharge its duty by approving the second level of safety assessment and the permit-to-work without having ensured that the permit reflected its own moving of the lifting platform and Zhang Hui’s scheduled move of the air compressor.

82 Further, s 10(c) of the WSHA makes it clear that a duty or liability imposed by the WSHA on any person is not diminished or affected by the fact that it is imposed on one or more other persons, whether in the same capacity or in different capacities. Simply put, the Respondent cannot absolve itself from liability or responsibility even if Zhang Hui was equally, or even more, responsible.

83 I am even less persuaded by the Respondent’s attempt to shift the responsibility for the offence or the breaches to its workers. Dr Ng’s observations at the second reading of the bill (Singapore Parliamentary Debates, Official Report (17 January 2006) vol 80 at col 2205) are pertinent:

It is thus the responsibility of the employer (or occupier or other “responsible persons”) to ensure that its workers are trained and are mindful of their safety at the workplace, and that proper systems are in place to ensure that steps are taken to minimise risks.

84 To be clear, I am not saying that an employer should be made liable for a breach even if it had instituted proper systems, issued countless reminders and had taken all reasonably practicable steps to minimise risks but its workers had wilfully refused to comply. Under the WSHA, workers also have a duty to cooperate with their employers and take steps to ensure their safety as well as the safety of others at the work place or they may be found guilty of an offence (see s 15 of the WSHA). It is possible to envisage a case where an employer should not be made liable or would have a very low level of culpability even if the offence is made out, if the accident was solely or mainly caused by the workers’ acts or negligence in spite of the proper systems that had been instituted by the employer. This, however, is not such a case. While some of the ground workers such as Mr Narun, who had given unwise instructions, may have some part to play in the accident and may be in breach of their duty under the WSHA, this does not absolve the Respondent from its responsibility.

85 Looking at the circumstances surrounding the accident, I am of the view that the Respondent’s culpability in the present case falls within the “medium” to “high” category. There were a number of breaches, though they cannot be said to be highly egregious breaches in that they are more akin to cumulative deviations from established procedures rather than a complete disregard of those procedures or of the workers’ safety. While I agree that there were some aspects beyond the Respondent’s control (as in the case of many accidents), I do not agree with the District Judge that it was a “perfect storm”. As admitted in the SOF, the accident could have been prevented had the Respondent taken reasonably practicable steps to ensure the safety of its workers.

86 I agree with the Prosecution that there was a high potential for harm in this case. The workers undertook high risk work without a safe system of work or the provision of any fall arrest equipment, and some of them were not properly trained. All of them, and not just the deceased persons, were at risk of falling. Further, it was not inconceivable and it was in fact likely that the moving of the air compressor onto the unsecured floating platform would cause a shift in weight and the platform to tilt, thus causing an accident – as it did. There was also the real possibility that the loading platform would have landed on other workers, resulting in further casualties or fatalities.

87 I should, however, state that I do not derive assistance from the Prosecution’s rather sweeping submission that the “potential for harm cannot be any higher than this”. This does nothing to advance its argument, which I accept, that the potential for harm is high. One need only think of workplace accidents such as the collapse of Nicoll Highway for this to be seen in perspective. While it cannot be gainsaid that every life is precious and every death is tragic, it also cannot be denied that the potential for harm—not only to the workers on the ground but to the public as a whole—can be far greater in other cases.

88 Weighing the two principal factors, I am of the view that the appropriate starting sentence for the Respondent is a fine of $300,000.

89 I turn next to consider how the sentence should be calibrated in the light of the aggravating and mitigating factors that are present.

90 The Prosecution submits that the relevant aggravating factors are that (a) serious actual harm was caused in that two lives were lost; and (b) the breaches were a significant cause of the harm. The fact that death has resulted has always been regarded as an aggravating factor. As observed by the court in R v Howe, “[g]enerally, where death is the consequence of a criminal act it is regarded as an aggravating feature [and] … [t]he penalty should reflect public disquiet at the unnecessary loss of life”. I agree with this observation. I also agree with the Prosecution that the Respondent’s breaches were cumulatively a significant cause of the harm that resulted.

91 There are several mitigating factors in the present case. The Respondent had pleaded guilty and had cooperated with the authorities from the outset. The Respondent has a good safety record, and had received awards from the Land Transport Authority and the MOM for its safe work practices. ²⁶ In particular, it had been awarded two such awards in 2013 in the same project where the accident in question occurred. This was its first fatal accident in more than two million accident-free man hours since 2011.

92 Further, the Respondent was also very proactive in ascertaining the cause of the accident and taking remedial steps to ensure as far as possible that no similar accidents recur. A week after the accident occurred, it commissioned a safety consultancy firm to carry out the investigations into the accident.

93 Weighing all the relevant aggravating and mitigating factors, I am of the view that the appropriate sentence in this case should be $250,000. Even after factoring in the aggravating factors, I am persuaded by the significant mitigating factors present to reduce the quantum of the fine from the starting point of $300,000 (see [88] above).

94 For the reasons above, I allow the Prosecution’s appeal and enhance the fine to a sum of $250,000.