Awarding costs in the Federal Court: Does the winner take all?

Awarding costs in the Federal Court: Does the winner take all?

Awarding costs in the Federal Court: Does the winner take all?

The judgment of Seiko Epson Co v Calidad Pty Ltd [2018] FCA 104 provides useful guidance on how the Federal Court apportions costs between parties who have each been partly successful in the litigation. The decision is particularly interesting because it is an example of a case where the “winner” was unsuccessful on a number of issues, but successful on a core issue.

This article considers how the Court will approach such a case with reference to a number of recent decisions, and highlights the complexity of awarding costs when the result of a case is mixed.

The parties’ arguments on costs

The substantive Australian Federal Court proceeding between Seiko and Calidad, involving claims of patent infringement, trade mark infringement, misleading and deceptive conduct, and breach of statutory duty, was resolved late last year. Calidad successfully defended all causes of action except patent infringement, where Seiko proved infringement in relation to certain categories of Calidad’s refurbished printer cartridges but not others.

Calidad submitted that it should be awarded all of its costs with respect to each of the causes of action for which it was wholly successful, and that the parties should bear their own costs of the patent infringement case (excluding two costs orders made during the interlocutory stages). In the alternative, Calidad argued that only 25% of the costs of the patent issues should be awarded to Seiko. Calidad agreed that it should pay Seiko’s costs in relation to the peripheral issue of joint tortfeasorship in the context of patent infringement.

Seiko, relying on Les Laboratories Servier v Apotex Pty Ltd [2016] FCAFC 27 (Servier), argued that Calidad should pay 60% of its costs of the whole proceeding, without dissecting costs by reference to the specific causes of action. The 40% reduction proposed by Seiko was based upon the centrality of the dispute concerning patent infringement. Seiko submitted that neither the time nor cost of the proceeding was significantly increased by the claims made in respect of the cartridges imported into and sold in Australia by Calidad which were found not to infringe Seiko’s patents.

Justice Burley’s reasons and costs orders

Justice Burley began by acknowledging that the Court has a wide discretion in awarding costs. His Honour rejected Seiko’s argument that the costs should not be dissected based on causes of action (and noted the “common practice” of costs being awarded on an issues basis where they can sensibly be separated), as the causes of action in question were discrete and clearly separable and Seiko had been wholly unsuccessful in relation to some of them.

Further, his Honour observed that Seiko’s apportionment was not supported by evidence to establish that the 40% reduction was appropriate and, “in the absence of further information, the question of whether or not it represents a fair allocation of costs is a matter of chance”. Therefore Calidad was awarded all of its costs with respect to each of the causes of action for which it was wholly successful, being trade mark infringement, misleading and deceptive conduct, and breach of statutory duty.

In relation to the patent infringement case, his Honour found that Seiko was entitled to its costs with a 25% reduction to take into account the fact that it had failed to establish infringement in respect of three of the seven categories of Calidad’s refurbished cartridges. His Honour rejected Calidad’s argument that each party bear its own costs in relation to the patent case on the basis that Seiko had succeeded in demonstrating patent infringement, albeit limited to only certain categories of goods.

Calidad has appealed against the trial decision (which includes these costs orders), and Seiko has both cross-appealed and issued a Notice of Contention, meaning that these costs issues will be revisited as part of the appeal.

The Federal Court’s approach to apportioning costs

There has been a clear trend towards costs not being apportioned on an issue-by-issue basis according to which party was successful on each issue or claim. This is due in part to a statement of principle to this effect from the Full Court in Servier (although, in the same case, there were also comments made recognising the apportionment of costs on an issue-by-issue basis).

In Servier, Apotex raised a number of grounds of invalidity of the patent in question, but was successful in relation to only one of those grounds, which resulted in the patent being revoked. The primary judge then made costs orders reflecting that outcome on an issue-by-issue basis, which resulted in Apotex paying some of Servier’s costs, despite the fact that Apotex had successfully invalidated the patent.

The Full Court overturned the primary judge’s decision to apportion costs based on issues, on the basis that the judge had erred in not starting from the position that the successful party is entitled to its costs, had failed to consider the extent to which there was a common substratum of fact in respect of issues on which the successful party succeeded and failed, and had not taken into consideration whether the raising of certain issues was justified.

Following Servier, in Sandvik Intellectual Property AB v Quarry Mining & Construction Equipment Pty Ltd (No 2) [2017] FCAFC 158, both parties accepted that the starting point was that Quarry, as the successful party, was entitled to its costs, and that some apportionment was appropriate given that Quarry had not succeeded on inventive step. The Full Court ordered that Sandvik pay 80% of Quarry’s costs, deciding that no higher reduction was justified because Sandvik’s success on this issue had not affected the outcome of the appeal.

More recently, in Idenix Pharmaceuticals LLC v Gilead Sciences Pty Ltd (No 2) [2018] FCAFC 7, Gilead succeeded in having Idenix’s appeal dismissed, however it failed on the issues raised in its Notice of Contention. The Full Court awarded Gilead all its costs with a small 10% reduction given that the length and complexity of the appeal was generated primarily by Idenix, not by Gilead’s Notice of Contention.

In Morroccanoil Israel Ltd v Aldi Foods Pty Ltd (No 2) [2017] FCA 1393 (Morroccanoil), Justice Katzmann opted not to apportion costs based on the success of the parties in relation to various issues due to a desire to avoid “a process of taxation that seems to be almost universally deplored”. Aldi was ordered to pay 35% of Morroccanoil’s costs.

In a case with a slightly different outcome, Shord v Commissioner of Taxation (No 2) [2018] FCAFC 27, the appellant was successful on one ground of appeal but the second ground of appeal was dismissed. The parties were ordered to bear their own costs of the appeal (rather than each party being awarded its costs in relation to the ground on which it had succeeded).

The reasoning in Seiko seems to be a departure from the trend demonstrated in the cases outlined above and, if Burley J’s costs order is upheld on appeal, the case may steer the way for similar costs orders in the future when multiple causes of action are raised and the results are mixed.

Lessons on the apportionment of costs

  1. The award of costs and the way in which costs are apportioned remains a matter of discretion for the Court.
  2. A party to litigation should bear the potential costs implications in mind when deciding whether to include ancillary causes of action, to bring a cross-appeal, or to issue a notice of contention. Raising ancillary causes of action which are ultimately unsuccessful or taking unsuccessful procedural steps in a proceeding may result in a favourable costs award being reduced, even if you are ultimately successful in the core dispute, particularly where these matters take up significant time and resources or are held by the Court to have had little prospects of success. This of course must be weighed up against the potential benefit to be gained from putting all possible legal arguments to the court.
  3. Based on authorities such as Servier, the usual starting position is that the successful party is entitled to recover its costs on a party-party basis (subject to certain limited exceptions generally linked to any disentitling conduct of the successful party), rather than the Court dividing up the costs entitlements of the parties on an issue-by-issue basis. However, as demonstrated by Seiko, the Court may depart from this general approach if multiple causes of action are raised, and award costs based on the outcome of each cause of action where circumstances make it reasonable to do so, such as where there are discrete and clearly separable causes of action (with limited overlap in the underlying facts).
  4. Ultimately, as seen in Morroccanoil, the Court may order that the costs recoverable by the successful party be reduced by an overall (and not formulaic) percentage so as to balance the successful versus the unsuccessful causes of action. This can avoid the costly and time-consuming process of dividing costs on an issue-by-issue basis. Relevant factors in determining the reduction will include the extent to which the unsuccessful claims contributed to the duration and complexity of the trial and whether it was reasonable in the circumstances to raise these claims.