• Lawyers in Sydney
• Lawyers in Melbourne
• Lawyers in Brisbane
• Lawyers in Perth
• Family lawyers
• Employment lawyers
• Criminal lawyers
• Wills and estates lawyers
• Find a lawyer

DMG had faced a judgment exceeding $2.5 million in Mr Bartlett's favour. On appeal, it challenged negligence, contributory negligence and damages. Its challenge to the finding of negligence failed. It succeeded on the absence of contributory negligence and on aspects of the damages assessment. Those successes reduced its liability by slightly more than $1 million, to just over $1.5 million ([1]).

After the appeal, the Court directed the parties to provide submissions about the form of orders and costs. The parties agreed on the judgment amounts and cross-claim contribution orders, and on a revised order for trial costs. The trial costs position differed from the order made by the primary judge because the reduced judgment meant a Calderbank offer was not bettered ([2]). The only outstanding dispute was the parties' appeal costs ([2]–[5]).

DMG said it had succeeded on two of its three broadly framed issues: contributory negligence and damages. It highlighted the reduction in its liability and the reversal of a special costs order. But it accepted that the usual rule that costs follow the event should not simply apply, because negligence had occupied most of the hearing. DMG proposed that the respondents pay 50% of its appeal costs ([3]).

Mr Bartlett argued that hearing time was a useful proxy for preparation time and costs. He relied on Elite Protective Personnel Pty Ltd v Salmon (No 2) [2007] NSWCA 373, Bostik Australia Pty Ltd v Liddiard (No 2) [2009] NSWCA 304, and Manhattan Homes Pty Ltd v Burnett [2025] NSWCA 32. His solicitor's notes suggested that liability and vicarious liability submissions took about three hours of a three-and-a-half-hour hearing. He therefore submitted that he should recover the bulk of his costs, estimating 70–75% ([4]). The Workers Compensation Nominal Insurer did not seek its costs and made no submission about Mr Bartlett's costs ([5]).

The Court's orders recorded the agreed judgment sums: $1,503,624.20 against DMG and $1,038,241.80 against the Nominal Insurer. The cross-claim orders allocated contribution between them at 10% and 90%, respectively, in the terms stated in the orders. DMG was also ordered to pay Mr Bartlett's ordinary trial costs up to 11 October 2024. The contested appeal costs issue was resolved by making no order, with each party bearing its own costs ([12]).

Why the court decided this way

The Court began from principles that were not disputed, while noting that the parties disagreed about how those principles applied ([6]). It reproduced six principles from Bostik at [38]. In summary, courts generally do not divide costs issue by issue merely because a successful party lost some arguments. A different approach may be appropriate if a particular issue or group of issues was clearly dominant or separable, or if an unsuccessful issue materially increased the hearing time. A separable issue can be a disputed question of fact or law, even if the party succeeds on the overall outcome. The ultimate apportionment remains discretionary.

The Court identified two prominent considerations: the effect of issues on the amount of liability, and the time they occupied, which is ordinarily a reasonable proxy for costs incurred ([7]). It did not treat either as decisive by itself. This mattered because the financial and time-based measures pointed in different directions.

On the financial measure, DMG had achieved substantial success. Its successful grounds reduced its liability by slightly more than $1 million, and the reduction also affected the trial costs position ([1]–[2]). On a simplified, outcome-based assessment, that was a significant appellate result.

On the time measure, however, Mr Bartlett's point had force. The Court examined the transcript and noted that the bulk of the parties' submissions concerned DMG's negligence grounds, on which DMG had failed. It also noted that only the last four of the 21 pages of DMG's opening submissions addressed economic loss, with a small part of the earlier material addressing contributory negligence ([7]). The unsuccessful negligence challenge thus consumed most of the hearing and submissions.

The Court characterised DMG's appeal as substantial, but distinguished between relatively straightforward challenges to aspects of the primary judge's reasons and the more ambitious challenge to liability, which dominated the hearing ([8]). It also observed that if DMG had limited its appeal to contributory negligence and damages, it would have enjoyed a costs order in its favour. Conversely, if Mr Bartlett had conceded all or part of the challenged damages assessment, he might have been entitled, or well placed, to obtain a costs order in his favour ([8]). The Court expressly said this was not a criticism of either party's forensic decisions. Its point was that decisions about the issues pursued or conceded could affect the eventual costs position.

The Court then cautioned against treating apportionment as arithmetic: “mathematical precision in this area is illusory” ([9]). It weighed DMG's substantial success on relatively discrete issues against Mr Bartlett's retention of a judgment of about $1.5 million, whose challenge had taken the majority of the time and costs ([10]). Those considerations “approximately balance[d] out”. The resulting order was no order as to appeal costs, with each active party bearing its own ([10]).

Before and after state of the law

The judgment presents its costs principles as established and undisputed, rather than announcing a new legal test. It expressly draws on the principles identified in Bostik at [38], including the discretion to depart from the usual rule, the relevance of dominant or separable issues, and the potential significance of time spent on unsuccessful arguments ([6]).

Accordingly, the “before” position stated in this judgment is that costs ordinarily follow the event, but that general approach may not resolve a case with mixed success. Courts may consider the parties' overall results, whether issues are separable, and how much hearing time unsuccessful issues consumed. The judgment also repeats that costs apportionment is an evaluative exercise, not one capable of exact mathematical division ([6]).

The “after” position is best described as an application of those principles to this appeal, not a change to them. The Court found that the financial effect of DMG's successful grounds and the time spent on its unsuccessful negligence challenge pulled in different directions. It resolved that case-specific balance by ordering no appeal costs ([7]–[10]). The judgment does not state that every appeal with partial success should result in each party bearing its own costs, nor does it set a percentage formula for future cases.

The judgment also records the agreed recalculation of trial costs. The reduced judgment meant that a Calderbank offer was not bettered, leading to a different trial costs result from the primary judge's order ([2]). The source does not provide the terms of that offer or the full reasoning behind the primary judge's original costs order. The point available from this judgment is limited to the parties' agreement and the Court's resulting orders.

Key passages with plain-English translation

At [6], the Court sets out the Bostik principle that where there are multiple issues, courts generally do not separate costs by each argument on which a party succeeded or failed. In plain English, losing some points does not automatically mean a successful party loses the costs of those points. A more specific order may be justified if an issue was clearly dominant or separable, or took up a significant part of the hearing.

At [7], the Court explains that the “bottom line” and time spent are both relevant. The bottom line means the extent to which successful arguments changed the amount of liability. Time spent in court may indicate the resources the parties used on particular issues. The Court treated hearing time as a proxy, not as a precise measurement of legal costs.

At [8], the Court says that both active parties could have improved their costs position through different forensic choices. This observation is expressly not a criticism. In practical terms, it illustrates that costs discretion can be influenced by the issues parties choose to pursue and the issues they choose to concede, while leaving the merits of those choices unjudged.

At [9], the Court states that mathematical precision is illusory. The phrase signals that the Court was not attempting to calculate a precise percentage of costs by counting hours or grounds. Instead, it assessed the whole appeal and compared the parties' relative success and the resources spent on the issues.

At [10], the Court gives its decisive assessment: DMG succeeded substantially on relatively discrete matters, but Mr Bartlett maintained a substantial judgment and defeated the challenge that consumed most of the hearing. The considerations approximately balanced. The plain-English result is that neither party was awarded the other's appeal costs.

What fact patterns trigger this precedent

The judgment is most directly relevant to costs disputes after an appeal has produced mixed results: one party has achieved a meaningful change to the judgment, but has failed on another issue that occupied substantial hearing time. Here, DMG reduced its liability by slightly more than $1 million but lost its challenge to negligence ([1], [7]).

It also illustrates the relevance of an appeal where the successful grounds are comparatively discrete while an unsuccessful ground dominates the submissions and hearing. The Court did not treat the number of broad issues won as conclusive. It considered how the issues related to the result and the resources spent on them ([3], [7], [10]).

A further relevant feature is that the respondent retained a substantial judgment despite the appeal reduction. The Court weighed that continuing result against the appellant's partial success ([10]). The judgment does not establish a general threshold for what counts as a substantial reduction, nor does it prescribe how much hearing time must be spent on an issue before it becomes dominant.

The source also shows that costs issues may remain after the substantive appeal has been decided and other orders have been agreed. Here, the amounts and cross-claim contribution orders were agreed, leaving only appeal costs for judicial determination ([2]–[5]). The decision's reach should be understood within those facts and the discretionary framework it applies.

How later courts have treated it

This judgment is dated 8 May 2025. The supplied text contains no subsequent judgments and therefore does not establish how later courts have treated this costs decision. It would be unsupported to describe it as followed, distinguished, or applied by later courts on the basis of the material provided.

The authorities identified within this judgment are earlier authorities. Bostik is the source of the principles reproduced at [6] and is applied in the Court's assessment. Elite Protective Personnel and Manhattan Homes appear as authorities relied on by Mr Bartlett's submissions at [4]. The judgment does not separately analyse their facts or adopt a distinct proposition from either beyond the general costs approach discussed in [6]–[10].

The earlier substantive appeal, De Martin & Gasparini Pty Ltd v Bartlett [2025] NSWCA 56, is identified at [1] as the judgment from which the appeal arose. It supplies the procedural and financial context for the costs decision. The present judgment should therefore be read as a short, fact-sensitive application of the costs discretion following that appeal, rather than as a comprehensive restatement of all costs principles.

Still-open questions

The judgment leaves the application of the discretionary principles dependent on the facts of each case. It does not specify how courts should quantify hearing time where the transcript, written submissions and preparation effort point in different directions. It describes hearing time as ordinarily a reasonable proxy for costs, but does not provide a formula or say that hearing time must be used ([7]).

Nor does it set a numerical threshold for when an issue is “clearly dominant or separable”, or for when time spent on an unsuccessful issue is sufficiently significant to affect the costs result. The principles quoted from Bostik guide the inquiry, but the Court stresses impression and evaluation rather than mathematical precision ([6], [9]).

The judgment also does not explain how a court should weigh the value of a successful ground against the resources consumed by an unsuccessful one in cases with different kinds of relief. Here, the contrast was between a substantial reduction in damages liability and a failed negligence challenge that occupied most of the appeal ([7], [10]). The Court does not turn that comparison into a general weighting rule.

Finally, the judgment provides limited detail about the Calderbank offer and the original trial costs order. It records that the reduced judgment meant the offer was not bettered and that the parties agreed to a revised trial costs order ([2]). The terms of the offer and the primary judge's full reasoning are not set out in this costs judgment, so no broader conclusion about those matters can be drawn from it.

This is general legal information, not legal advice. Zoe can make mistakes. Before acting on or relying on this analysis, have a qualified Australian lawyer check it.