1 Following my decision in Mah Kiat Seng v Attorney-General and others [2023] SGHC 14, I heard parties on costs. One of the issues arising involves the principles on which costs of a litigant in person are assessed, on which there is a paucity of authority. For this reason, it is preferable to put my costs decision in writing. I adopt the abbreviations used in my earlier judgment.

2 The AG submits that costs should follow the event, and describes the event as: ¹

(a) success for Mah on two claims, namely against Rosli for unlawful apprehension and against the SPF in respect of the search conducted on him and his bag; and

(b) success for Tan on the claim of assault and for the SPF on the balance claims.

3 Mah contends that I should consider the “thesis” of his case, ² which he submits was wrongful arrest, and as he succeeded on that he should be considered the successful party even though he failed on other claims.

4 While Mah’s approach is appropriate where the court is considering the situation where some but not all of the claims against one defendant succeed, it is not appropriate to the situation where there are claims against two different defendants and the plaintiff succeeds against one defendant but not the other. In these circumstances, what the event is in relation to each defendant should be considered separately. That the defendants had the same legal representation does not affect the question of incidence of costs, although when the court turns to quantum it will have some relevance: see [20] below. In my view, Mah is entitled to the costs of his successful claim against Rosli but must pay costs for his unsuccessful claim against Tan.

5 As for the claims against the SPF, the AG is correct that there were some claims that Mah made that concerned the SPF generally and were not directed at specific actions of Rosli or Tan for which the SPF would be vicariously liable. In relation to those claims, Mah did succeed on the point of the search of his bag and person but failed on various other points. I consider that overall, the SPF was successful in its defence and accordingly is entitled to costs, although account must be taken of Mah’s limited and partial success against SPF when it comes to quantifying those costs.

6 I turn now to assessing the quantum of costs.

7  Mah is a litigant in person. By O 59 r 18A of the Rules of Court (Cap 332, R 5, 2014 Rev Ed) (“ROC 2014”) (which has been in substance carried forward into the Rules of Court 2021 as O 21 r 7), a litigant in person may be “allowed such costs as would reasonably compensate the litigant for the time expended by him, together with all expenses reasonably incurred”.

8  Mah referred to a decision of Chief Master Marsh in the English case of Campbell v Campbell [2016] EWHC 2237 (Ch). But that is a decision on certain provisions of the English Civil Procedure Rules (“CPR”) concerning costs management. Unlike in England, our rules do not set any hourly rate for litigants in person. That decision at [28] also refers to a cap contained in the CPR on recovery for a litigant in person set at two-thirds of the amount that would have been allowed were the litigant in person legally represented. The rationale for this is a presumed 50% mark up by lawyers on their underlying expense rate, which should not be allowed to a litigant in person as otherwise they might profit from the costs of the litigation. This mirrors the principle that a legally represented litigant may not recover as costs more than he has actually incurred (where this is below the figures indicated by applicable costs guidelines or costs precedents).

9 The first point taken by the AG is that Mah should have brought these proceedings in the Magistrate’s Court, given that the damages obtained by him are well below the jurisdictional limit of that court. For this reason and by virtue of O 59 r 27(5) of the ROC 2014, Mah’s costs should not exceed what he would have been entitled to had he proceeded in the appropriate court. By Part IV of Appendix 2 to the ROC 2014, where a plaintiff is awarded up to $20,000 (as was the case here), the costs to be allowed are $3,000 to $6,000 (excluding disbursements). Nonetheless, the court has a discretion to order otherwise under O 59 r 31(2). In my view, given that this case, despite its low monetary value, is not a straightforward one, I would exercise my discretion to order otherwise. Thus, I am prepared to consider that Mah would have spent much more time on these proceedings than might be the case for a typical Magistrate’s Court case. Nonetheless, as I explain at [16] below, when it comes to fixing the appropriate hourly rate the principle of proportionality suggests that the Magistrate’s Court scale is a helpful reference point.

10 This brings me to the question of how to assess the amount that would reasonably compensate Mah for the time expended by him. In my view, there are two parts to this. The first is to determine the time expended by him and the second is to fix an amount (for example an hourly rate) that compensates him for that time.

11 In the oral hearing, Mah suggested that he had expended about 1,000 hours in total on these proceedings. There were six days of trial, and he had to prepare for cross-examination as well as for his own evidence. He had to file closing submissions after the trial. He had to view video recordings. He referred to the amount of legal research he had done, and the results of that effort were certainly visible in the proceedings.

12 The AG did not offer any estimate of the time that Mah might have expended. They argued that he had failed to substantiate the number suggested by him.

13 In my view, there is no need for a litigant in person to produce time sheets. The court is able to use its own experience of litigation to estimate the reasonable time spent. A simple calculation, excluding matters for which there have been separate interlocutory costs orders, would be 460 hours, comprising 60 hours for the trial days, 300 hours for pre-trial preparation and 100 hours for post-trial work. I am satisfied that this figure is justified. It is important to note that this is an estimate of what would be the reasonable amount of time spent. It may well be that Mr Mah in fact spent 1,000 hours on the matter, because of its personal importance to him, but the court’s inquiry involves an objective yardstick of reasonableness.

14 However, from this total of 460 hours it must be determined how much is referable to the claims on which Mah succeeded. Taking a broad brush approach, I am satisfied that the issue of unlawful apprehension was by far the one involving most work, especially in terms of legal research but also in terms of factual analysis. Accordingly, I would take a proportion of 70%, resulting in a net figure of 322 hours.

15 I now turn to the compensatory hourly rate. In principle, I do not think it is necessary for the litigant in person to prove the opportunity cost to him of the hours expended. This is not a claim for damages but an exercise of costs jurisdiction, which is grounded first and foremost in policy considerations of access to justice. The court in an individual case does not have the benefit of the empirical research which would ground an hourly rate that could, for example, be fixed in the rules of court. Nonetheless, the court should use the resources available to it to reach an estimation.

16 Costs incurred must be proportional to the amount at stake. While there was a point of principle in this case, that principle could have been litigated in the Magistrate’s Court. Accordingly, it is important to bring back into reference the scale of costs applicable in the Magistrate’s Court. It is a reasonable assumption that typical cases resulting in damages of $20,000 might take one or at most two days to try, and three to four days in pre- and post-trial work. This suggests about 50 hours allowed for the costs of the lawyer representing the litigant. Taking the figure of $5,000 within the scale results in a figure of $100 per hour. Discounting this by a third to eliminate the profit margin that a lawyer is entitled to but not a litigant in person results in a net hourly rate of $66.66.

17 I would round this down to $60. Accordingly, multiplying $60 by 322, I fix costs payable to Mah at $19,320.

18  The AG has claimed $35,000 in costs in respect of the successful defence of Tan, and a further $50,000 for the partially successful defence of the SPF.

19  I agree that costs should be assessed on the High Court scale given that it was Mah who chose to bring these proceedings in the High Court. The AG refers to the Guidelines for Party-and-Party Costs in the Supreme Court of Singapore at Appendix G to the Supreme Court Practice Directions 2013 (the “Costs Guidelines”). Taking the category of Simple Torts in Section III, they arrive at a range between $66,000 and $132,000 in costs for pre-trial work, six days of trial and post-trial work. ³ Without giving any working, they then seek the amounts of $35,000 and $50,000 respectively totalling $85,000. In doing so they appear to treat the costs claims for Tan and the SPF as independent and cumulative.

20 At this point, the joint representation of the defendants is material. Ultimately, there is one total set of costs incurred by the defendants which should be apportioned among them. Otherwise, there is a risk of double counting and of litigants being awarded more costs than actually incurred. By way of example, the same lawyers were in court for the trial for all defendants. If the case had been entirely dismissed, the defendants could not each ask for the time of the same lawyers being in court.

21 Indeed, when it came to their disbursements, the AG only sought 50% of them. Yet for the costs, they seek, assuming the higher end of the Simple Torts category, more than 60% (ie, $85,000 against $132,000).

22 In my view, it is more helpful to look at the complexity of the case as a whole and then consider the proportion of the case taken up by the claims on which Tan and the SPF succeeded.

23 Once one considers the matter as a whole, it is apparent it should not be categorised as Simple Torts but as Torts. Applying the Torts category in the Costs Guidelines would give a range from $91,000 to $196,000 with a midpoint of $143,500. This compares with a midpoint for Simple Torts of $99,000. Thus, my starting point is actually higher than that of the AG.

24 However, I would then award only a proportion by reference to the proportion of time spent on the claims on which Tan and the SPF succeeded. In my view, that is only 30%, taken together. Accordingly, taking a reasonable and proportionate amount for the matter as a whole to be $160,000, I would fix costs in their favour at $48,000. I note that the AG has sought two separate amounts. I do not think it is necessary to do so but if I were to allocate costs between them then it would be $22,000 in respect of the claims against Tan and $26,000 in respect of the balance claims.

25 Mah is to have 70% of his reasonable disbursements incurred while the AG is to have 30%. If these disbursements cannot be agreed within 14 days hereof, then any disputed items can be resolved by the Registrar.

26  At the conclusion of the oral hearing, the AG requested that I stop time from running on the unpaid damages, as they had offered to pay those damages but Mah had suggested setting off those damages against any amount he might have to pay to the defendants as costs. I declined to do so, as it seemed straightforward to set off any interest that has accrued against the costs. As is apparent from this costs judgment, there is indeed a net balance of costs owing to the defendants against which the damages (and any accrued interest) may be duly set off. Indeed, even after that set off, there will still be a balance payable by and due from Mah.

27 Lastly, I would observe that it may appear incongruent that Mah is awarded $19, 320 in costs for the issues comprising 70% of the matter while the AG is awarded $48,000 for the issues comprising the balance 30%. However, this apparent incongruity is explained and justified by two points. First, compensation for the time of litigants in person will ordinarily be less than the amount that the court awards to litigants for the cost of legal representation, because of the principle that litigants must not profit from costs of legal proceedings. As an aside, litigants who engage lawyers will typically spend more on that representation than they are awarded by the court if successful against the other party. Secondly, in this particular case I have referred to the Magistrate’s Court scale as a reference point for Mah’s costs because this case could have been brought there, even though I have not limited the amount awarded to that scale. I have done this because of the principle of proportionality. By contrast, I have assessed the AG’s costs against the Cost Guidelines (which apply to High Court matters) because the AG did not choose the court, and a successful defendant to a matter brought in a higher court than the one appropriate to the claim is entitled to costs assessed on the scale or guidelines of that higher court.