1. Introduction
On 5 June 2026, Whitby J delivered judgment in Wright v Lemon [No 5] [2026] WASC 218, ordering that five individuals and entities who funded, brokered or otherwise facilitated Julian Wright’s unsuccessful claim against the interests of his late brother and his sister be jointly and severally liable, together with Julian, for the defendants’ costs of the trial – costs estimated at $5.76 million (at [117], [277]).
The decision is the most comprehensive statement to date by the Supreme Court of Western Australia of the principles governing costs orders against non-parties who stand behind litigation. Its central holding is doctrinal: a causal connection between the non-party and the incurrence of costs is not a precondition to the exercise of the power under s 37(1) of the Supreme Court Act 1935 (WA). The touchstone is whether the non-party had a connection to the litigation sufficient to warrant the exercise of the power, judged against what is fair and just between the parties (at [102]–[110], [124]). In so holding, Whitby J declined to follow the reading of Kaur v Sikh Gurdwara Perth (Inc) [No 2] [2018] WASC 99 that had been pressed upon the court, and aligned Western Australian practice with the Full Court of the Federal Court’s decision in Court House Capital Pty Ltd v RP Data Pty Ltd [2023] FCAFC 192.
The decision warrants attention well beyond the parties for three reasons. First, it resolves a point of genuine uncertainty in Western Australian costs law, on which first-instance authority had been read as requiring a ‘real and direct’ causal link. Secondly, the non-parties exposed were not institutional litigation funders. They were a friend who sourced money from his own contacts and worked as a ‘legal assistant’ on a speculative basis, a broker who structured funding through unit trusts and corporate trustees, and an investor who introduced fellow investors. The judgment demonstrates that informal and quasi-professional funding networks attract the same exposure as commercial funders. Thirdly, the decision supplies practical guidance on almost every argument a non-party might deploy in resistance – onus, causation, apportionment, control, impecuniosity, security for costs, want of warning, and corporate or trust structuring – and rejects each on the facts.
The decision is of immediate relevance to civil litigation practitioners in Western Australia: those acting for successful parties contemplating recovery against funders; those advising funders, brokers and investors on exposure; and those acting for funded plaintiffs, whose disclosure obligations under O 9A of the Rules of the Supreme Court 1971 (WA) acquire renewed significance.
2. Relevant Legal Framework
Section 37(1) of the Supreme Court Act 1935 (WA) provides that the costs of and incidental to all proceedings in the Supreme Court ‘shall be in the discretion of the Court or judge’, and that the court ‘shall have full power to determine by whom or out of what estate, fund, or property, and to what extent such costs are to be paid’ (set out at [87]). The discretion must be exercised judicially but is otherwise ‘absolute, unconfined and unfettered’: Frigger v Lean [2012] WASCA 66 at [53], cited at [88]. The overarching guide is what is fair and just between the parties: Latoudis v Casey [1990] HCA 59; (1990) 170 CLR 534, 558, cited at [88].
It has long been established that statutory costs powers in this form extend to orders against persons who are not parties to the proceedings, including those who fund or financially assist litigation: see Knight v FP Special Assets Ltd [1992] HCA 28; (1992) 174 CLR 178, in which the High Court confirmed that a general costs jurisdiction permits orders against non-parties in appropriate cases. Whitby J treated the proposition as well established (at [89]).
Within that broad jurisdiction, the authorities have grappled with how to confine the power. In Bischof v Adams [1992] 2 VR 198, 205, Gobbo J examined both ‘the connection between the non party and the proceedings’ and ‘the causal connection between the non party and the costs’, requiring a connection that is ‘real and direct’ and observing that mere benefit from the litigation would not, without more, suffice (quoted at [100]). The Privy Council in Dymocks Franchise Systems (NSW) Pty Ltd v Todd [2004] 1 WLR 2807 expressed the view that if costs would have been incurred without the non-party’s involvement, the non-party should not ordinarily be liable for them (discussed at [98]–[99]).
In Western Australia, Le Miere J drew on both authorities in Kaur v Sikh Gurdwara Perth (Inc) [No 2] [2018] WASC 99 at [13], stating that ‘the authorities establish that there must be a causal link’ and that the link ‘must be real and direct’. In Kaur, his Honour also indicated that the applicant bears the onus of establishing that the order is just (at [16]). Those statements were applied at first instance, including by Lundberg J in Sprintex Limited [No 3] [2025] WASC 59, where a causal connection was one of several factors considered (discussed at [101], [108]).
The federal authorities developed differently. In Gore v Justice Corporation Pty Ltd (2002) 119 FCR 429, the Full Court held that the correct approach is ‘to examine what did happen, putting to one side issues of speculation’ about whether the litigation could have continued unaided (at [61] of Gore, quoted at [104]). In Hardingham v RP Data Proprietary Limited [2023] FCA 480 at [19]–[23], Thawley J held that the power is exercised where the non-party’s connection to the litigation is sufficient to warrant it; that assistance founded in family or social ties, without commercial interest, is typically insufficient; that it is ‘not exceptional to order costs against the litigation funder who facilitates litigation for their own commercial gain’; and that it would be unfair to allow a non-party to fund litigation in the hope of benefit without facing the corresponding costs risk (quoted at [90]–[93]). The Full Court upheld that decision in Court House Capital Pty Ltd v RP Data Pty Ltd [2023] FCAFC 192, expressly rejecting a ‘but for’ requirement (at [34] of Court House Capital, quoted at [103]) and confirming that a funder with a commercial interest may be liable even absent any control over the proceedings (at [35], citing Gore at [64]; quoted at [239]).
Two further strands complete the framework. In Ballantyne Suites Pty Ltd v Ballantyne Chambers Pty Ltd (in liquidation) [2014] VSCA 223 at [45], the Victorian Court of Appeal held that ‘[t]here is no onus of proof in an application for the exercise of a discretion such as that in the present case’, a holding since cited with approval in Ipex ITG Pty Ltd (in liq) (receivers appointed) v Victoria [2014] VSCA 315 at [45] and Mistrina Pty Ltd v Australian Consulting Engineers Pty Ltd – Costs [2020] NSWSC 633 at [24] (discussed at [95]–[96]). And in Carter v Caason Investments Pty Ltd [2016] VSCA 236, the Victorian Court of Appeal identified factors relevant to the discretion – including whether the non-party’s commercial interest goes beyond mere recovery of funds advanced – and upheld an order against the sole director of a corporate funder, observing that it would not be in the interests of justice if corporate funders could be established with virtually no assets so that ‘those truly standing behind the litigation would not be exposed to any adverse costs order’ (at [55] of Carter, quoted at [229]).
Before Wright v Lemon [No 5], therefore, Western Australian first-instance authority sat in apparent tension with the federal and Victorian appellate authorities on two points: onus, and the necessity of a causal connection. The decision resolves both.
3. The Facts of the Case
The underlying litigation
Ernest (Peter) Wright carried on mineral exploration through Wright Prospecting Pty Ltd (WPPL). On his death in 1985 his shares passed to his three children, Michael, Angela and Julian (at [1]). In 1987 Julian sold his one-third shareholding to Michael and Angela (at [2]). Proceedings brought in 2001 by Julian’s children were resolved in 2008 by a Deed of Settlement to which Julian was party and which contained a covenant not to sue (at [3]).
On 24 February 2017, Julian commenced proceedings claiming that the assets of his father’s estate and matters material to WPPL’s mining interests had not been fully disclosed to him and that he did not receive full and fair value for his shares; he sought reinstatement as a one-third shareholder or, in the alternative, equitable compensation and damages (at [4]–[5]). On 2 July 2021, Le Miere J dismissed the claims, finding they were barred by the 2008 Settlement Deed and that commencing the proceedings breached the covenant not to sue: Wright v Lemon [No 2] [2021] WASC 159 (at [7]). Costs orders followed on 8 February 2022: Wright v Lemon [No 2] [2021] WASC 159 (S) (at [8]). The Court of Appeal dismissed Julian’s appeal on 1 March 2024 (Wright v Lemon [2024] WASCA 19) and special leave to appeal to the High Court was refused on 8 August 2024 (at [8]).
By chamber summons dated 17 December 2025, the defendants sought orders that six non-parties be jointly and severally liable with Julian for the defendants’ costs of the trial – not the appeal (at [9]–[10]). The application was discontinued by consent against one respondent, leaving three groups: Mr Ian Thom and Barbirolli Investments Pty Ltd (the Thom Parties); Mr David Purcell and Litigation Funding Solutions (Australia) Pty Ltd (LFSA) (the Purcell Parties); and Mr John Trudgian, who did not appear (at [11]). The summons was heard over three days in April 2026 (at [11]).
The Thom Parties
Mr Thom, a friend of Julian since 2012, approached Julian about reviewing the files from the earlier proceedings with a barrister to assess whether Julian had a claim (at [24]–[31]). In return for introducing Julian to counsel, assisting to find funding for a legal opinion and helping to review documents, Julian entered into a deed dated 16 May 2013 with Barbirolli – a company of which Mr Thom was sole director and ‘controlling mind’, which did not trade, held no assets and kept no financial records (the Barbirolli Deed) (at [32]–[33]). The Barbirolli Deed entitled Barbirolli to 50% of any settlement monies up to and including $400 million, 30% of any excess, and 15% of any entitlement Julian had to future dividends or royalties (at [147]).
In 2013, Mr Thom procured $125,000 from four of his own contacts under agreements styled ‘loan agreements’, repayable only upon a successful outcome and carrying entitlements to a share of any settlement (at [34]–[38], [127]). Between 2013 and 2014, opinions were obtained from junior and senior counsel (at [36]–[40]). From 1 April 2016, Mr Thom was employed by Julian’s solicitors, Butcher Paull & Calder, as a ‘legal assistant’ working exclusively on Julian’s claim, remunerated only upon successful conclusion of the claim and payable directly by Julian (at [42]–[43], [133]). In August 2016, further advice supportive of a claim was obtained (at [46]). Mr Thom introduced Julian to Mr Purcell by no later than 2019 (at [49], [138]). At trial, Mr Thom attended on most days, sat in the position usually occupied by an instructing solicitor, engaged with Julian’s legal team and passed messages to counsel (at [71], [140]–[144]).
The Purcell Parties
Mr Purcell was sole director of LFSA, a self-described ‘broker or originator of litigation funding’. By a Mandate Agreement of 3 August 2019, Julian granted LFSA an exclusive period to secure funding offers, acknowledging that LFSA would be compensated by any funder it introduced (at [49]–[50]). Four funding tranches followed, structured through unit trusts with corporate trustees of which Mr Purcell was sole director: the LFS Funding Agreement of 17 July 2020 ($2 million, in exchange for 10% of any recovery capped at $1.2 billion, secured by first mortgage over the Marri Wood Park property owned by Julian’s company Nattim Pty Ltd) (at [56]–[57], [216]); the 2020 M&R Funding Agreement of 12 October 2020 ($550,000 for 5.5% of recovery, secured by second mortgage, entered while judgment was reserved) (at [58]–[59], [220]); the J4J arrangement of March 2021 ($1.1 million) (at [61]); and the 2021 M&R Funding Agreement of 18 May 2021 ($550,000 for 6.875% of recovery) (at [62], [224]). In each trust, A Class unitholders (the investors) were entitled to repayment and 80% of net income; B Class units carrying 20% of net income were held by or shared with LFSA (at [217]–[218], [221]–[222], [225]–[226]). Mr Purcell, through LFSA, stood to gain up to $47.7 million in total, and his entities received fees from each advance (at [219], [223], [227]–[228]).
Mr Trudgian
Mr Trudgian, introduced to Julian’s cause by Mr Purcell, advanced approximately $425,000 to $475,000 of his own and family trust funds across the funding agreements (at [51]–[52], [260]–[261]). Under a further agreement with LFSA, he was rewarded with shared B Class units for each investor he introduced – the schedule recording eight introductions (at [266]–[267]). His total potential gain was up to $37.2 million (at [273]).
Events after the litigation failed
After special leave was refused, the mortgages securing $2.55 million of funding were called and discharged, and Marri Wood Park was sold on 18 November 2024 for $5,450,000 to a company controlled by Julian’s son (at [64]–[66], [248]). On Mr Purcell’s applications, the two trustee funding companies were deregistered on 22 January 2025, Mr Purcell declaring to ASIC that each had assets worth less than $1,000 (at [251]–[252]). There was evidence, accepted by the court, of a real likelihood that Julian would be unable to pay the defendants’ costs or a substantial portion of them (at [187]–[192]).
The credibility findings
Whitby J accepted the evidence of the defendants’ witnesses (at [69], [72]) but found Mr Thom neither credible nor reliable, identifying five instances in which his evidence was inconsistent with the documents, his own affidavit or other witnesses, and finding that he sought substantially to downplay his role (at [73]–[86]). His evidence was not accepted except where it amounted to an admission, was inherently probable, or was corroborated (at [86]). Mr Purcell elected to adduce no evidence (at [19], [207]).
4. Analysis of the Court’s Reasoning
No onus of proof
The Thom Parties, relying on Kaur at [16], submitted that the applicant bears the legal onus of establishing that a non-party costs order is just (at [94]). Whitby J preferred the Victorian Court of Appeal’s holding in Ballantyne at [45] that there is no onus of proof in an application for the exercise of such a discretion, noting that Kaur was decided without reference to Ballantyne and Ipex and that the point was not live in Kaur (at [95]–[97]). The court added that the application did not, in any event, turn on onus (at [97]).
Causal connection is not a precondition
The central doctrinal question was whether the applicant must demonstrate a causal connection between the non-party and the incurrence of costs – in substance, a ‘but for’ precondition. Both the Thom Parties and the Purcell Parties relied on Kaur at [13] (at [98]). Whitby J held that Kaur, Sprintex and Bischof do not stand as authority for such a precondition, for five reasons (at [102]–[110]):
(1) the Full Court in Court House Capital unanimously rejected the proposition that a ‘but for’ finding is required, the court rejecting the contention as unsupported by authority and unsound (at [103], quoting Court House Capital at [34]);
(2) a ‘but for’ test would require the court to speculate about what might have happened absent the funding, an exercise the Full Court in Gore held should be put to one side in favour of examining ‘what did happen’ (at [104], quoting Gore at [61]);
(3) the landscape of litigation funding has changed since Bischof and Dymocks: funding as a commercial profit-making enterprise is increasingly common, and applications must be determined in that context (at [105], citing Hardingham at [21]);
(4) Kaur and Sprintex are not inconsistent with Court House Capital: neither involved a commercial litigation funder, and in neither was causal connection the sole determinant – it was one factor in a fact-specific enquiry (at [106]–[109]); and
(5) imposing such a precondition is inconsistent with the wide and unfettered nature of the s 37 discretion (at [110]).
The governing principle, repeatedly stated, is that ‘the proper enquiry is whether a non-party has a sufficient connection to the proceedings and whether it is fair and just to make a costs order against the non-party’ (at [124]; see also [112], [130], [210]).
The factors guiding the discretion
Whitby J emphasised that factors or guidelines ‘are not a substitute for, and are not a fetter upon’ the discretion, and are not closed (at [111]). Relevant considerations may include (at [113], citing Carter at [19], [49]–[50] and Court House Capital at [38]):
(1) whether the non-party has a commercial interest in the subject matter going beyond mere recovery of funds provided;
(2) whether the non-party had any right to be involved in decision-making and/or was entitled to protections with respect to settlement;
(3) the financial state of the unsuccessful party, particularly whether it can meet a costs order;
(4) any failure to seek security for costs and/or to warn non-parties that a costs order will be sought; and
(5) whether the proceedings involved matters of public interest extending beyond the parties.
This is ‘neither a checklist nor an exhaustive list’ (at [114]).
Apportionment is not required and selective pursuit is permissible
The Thom Parties contended the application must fail because the defendants had pursued only selected non-parties and made no attempt to apportion the costs sought between them (at [115]–[116]). Whitby J rejected both limbs. It is a matter for the applicant which non-parties it pursues; on the defendants’ estimate there were 52 individual funders, and requiring applications against each would be unreasonable and impractical – non-parties made liable may themselves seek contribution from others (at [119]). Nor must the applicant apportion: in Court House Capital at [39], a submission that the order must correspond with the degree of funding was rejected as offending ‘the broad and discretionary nature of the power’, and in CPC Patent Technologies Pty Ltd v Apple Pty Ltd (No 2) [2025] FCA 1671 four separate non-parties were held jointly and severally liable despite having funded in different circumstances (at [122]–[123]). The amounts connected to the respondent non-parties were, in any event, approximately $4.4 million of the $4.75 million advanced (at [120]).
The Thom Parties: facilitation without advancing funds
The Thom Parties advanced none of their own money. That did not preclude the order: costs orders may be made against a non-party whose role is to source funds for the litigation from others – Maylord Equity Management Pty Ltd v Nauer (No 2) [2017] NSWSC 1467 at [52], applied at [130] and [210]. The court found Mr Thom procured the initial $125,000; the ‘loan’ label did not matter because the advances were repayable only upon success and carried entitlements to a share of the proceeds, bearing ‘all the characteristics of funding agreements’ (at [127]). His facilitation was ‘substantial and sustained’: introductions to counsel and to Mr Purcell, document collation, employment as a legal assistant remunerated only out of any proceeds, and daily attendance and engagement at trial (at [128]–[146]).
The commercial interest analysis turned on the construction of the Barbirolli Deed. Applying orthodox principles – recitals may aid construction of operative clauses (Franklins Pty Ltd v Metcash Trading Ltd (2009) 76 NSWLR 603 at [380]); a deed is construed as a whole, preferring harmonious constructions (Australian Broadcasting Commission v Australasian Performing Right Association Ltd [1973] HCA 36; (1973) 129 CLR 99, 109; AIG Insurance Australia Ltd v McMurray [2023] WASCA 148 at [136]); and a general instrument is read by reference to what the parties contemplated, as disclosed by the recitals (Grant v John Grant & Sons Pty Ltd (1954) 91 CLR 112, 131) – the court held the Deed extended to any action by Julian concerning ownership of shares or assets of WPPL, whenever commenced, and that its condition was satisfied by the 2016 advice (at [158]–[169]). The submission that the Deed had ‘run its course’ after unfavourable advices in 2013–2014 was rejected as without merit (at [170]).
The alternative contention – that Julian repudiated the Deed in 2020 by saying ‘the deal was over’ and that Mr Thom accepted the repudiation – failed on the evidence. Renunciation requires conduct conveying to a reasonable person repudiation of the contract as a whole or of a fundamental obligation, is a serious matter not lightly inferred, and must be clearly proved: Armada Balnaves Pte Ltd v Woodside Energy Julimar Proprietary Limited [2022] WASCA 69 at [509], [515], applied at [174]–[175]. The only evidence was Mr Thom’s own, first asserted in 2026; it was uncorroborated, inconsistent with Julian’s redaction of the Deed in 2022, with Mr Thom’s continued close involvement at trial, and with his retention of the Deed when he retained almost nothing else; and it was implausible that entitlements of that magnitude would be surrendered on an oral remark (at [176]–[177]).
The result was a finding that Mr Thom was ‘an integral part of the litigation’, with Barbirolli – and through it Mr Thom – having the most to gain from success: ‘It ultimately stood to gain more than even Julian himself’ (at [205]–[206]).
The Purcell Parties: brokerage, trust structures and risk-free funding
Mr Purcell’s primary answer was that he and LFSA were mere brokers engaged by the funders, that the corporate trustees acted only as trustees, and that neither he nor LFSA provided funding or stood to benefit (at [210]–[211], [214]). The court looked to substance. LFSA’s fee was paid out of the funding advanced to Julian, and LFSA held B Class units entitling it to 20% of the net income of each trust – a total potential gain of up to $47.7 million (at [212], [219], [223], [227]–[228]). Mr Purcell ‘was in substance the person who sourced the funding’; it would be unfair and unjust to allow him ‘to escape a non-party costs order by hiding behind the corporate structures and trust arrangements he set up’ (at [230]), the court applying the observation in Carter at [55] concerning funders established with virtually no assets (at [229]).
Three further arguments failed. As to timing, funding provided late – including while judgment was reserved – still attracts liability where it ensures the proceedings continue and confers a share of the outcome: CPC Patent Technologies at [108]–[109], applied at [234]; and the only reasonable inference was that Julian required the funding to continue (at [235]). As to control, each agreement disclaimed direct or indirect control while requiring consultation on settlement; but where a funder has a commercial interest, the requisite connection may be established even absent control: Court House Capital at [35], citing Gore at [64], applied at [239]–[240]. As to the deregistration of the trustee companies, the submission that they were the proper respondents was ‘fatal[ly]’ undermined by Mr Purcell’s own declarations to ASIC that each had assets under $1,000 (at [250]–[253]).
Two features aggravated the position. The $2.55 million advanced under the two mortgage-secured agreements was repayable irrespective of the outcome, rendering that funding ‘entirely risk free’; absent a non-party costs order, the funding vehicles ‘would experience only upside’, which ‘would encourage litigation funders to pursue profit with no consideration for the vindication of legal rights’ (at [247]). And the calling of those mortgages forced the sale of Marri Wood Park, removing Julian’s principal asset from the defendants’ reach (at [248]–[249]).
Mr Trudgian: the investor who introduces other investors
Mr Trudgian neither appeared nor adduced evidence; the court was satisfied the application had come to his attention and proceeded in his absence (at [258]–[259]). He had advanced approximately $425,000 to $475,000, introduced at least eight further investors under an incentive arrangement rewarding him with shared B Class units, and stood to gain up to $37.2 million (at [261], [266]–[267], [273]). His funding was ‘purely with a view to making a commercial profit’, and the connection was sufficient (at [274]–[275]).
The rejected discretionary defences
Four further matters were said to weigh against the orders; none did. First, Mr Thom’s claimed impecuniosity (his only asset a half-share of his home) was held irrelevant, applying by analogy Northern Territory v Sangare [2019] HCA 25; (2019) 265 CLR 164, 175: ‘Whether a party is rich or poor has, generally speaking, no relevant connection with the litigation’ (at [181]–[182]). To hold otherwise ‘would serve to encourage a non-party to take steps to structure their affairs to avoid being subject to a non-party costs order’ (at [183]). Mr Purcell’s assertion that an order would bankrupt him failed for want of any evidence (at [244]–[245]).
Secondly, the defendants’ ability to pursue Julian was no answer: the evidence established a real likelihood that Julian could not pay, which weighed in favour of, not against, the orders (at [185]–[192], [241]–[243]). The court also confirmed that an application made before taxation of costs is ‘entirely appropriate and reasonable’, and that non-parties made liable have the right to be heard on quantum (at [189]).
Thirdly, the failure to seek security for costs carried no weight. Courts do not, absent exceptional circumstances, order security against impecunious individual plaintiffs, and the defendants were ‘well advised’ not to apply (at [194], [197]); Hardingham at [28] and Court House Capital at [18] are to the same effect (at [195]–[196], [198]).
Fourthly, the failure to warn the non-parties failed as a factor because the defendants did not know of the non-parties’ financial interests. The defendants had twice sought disclosure of interested non-parties under O 9A r 2 of the Rules of the Supreme Court 1971 (WA), in 2017 and 2018, and were twice told there were none (at [200]). While the Thom Parties may not have fallen within the O 9A definition, it would be unfair to weigh a want of warning against an applicant kept in ignorance of the non-party’s existence or interest (at [203]–[204]).
5. Assessing the Consequences
The financial architecture of the judgment repays attention. The defendants’ trial costs were estimated at $5.76 million (at [117], [187]). Total litigation funding advanced to Julian was approximately $4.75 million, of which approximately $4.4 million was connected to the respondent non-parties (at [120]). Against those figures sat the non-parties’ potential rewards: Barbirolli’s entitlement to 50% of any settlement up to $400 million together with 15% of future dividends or royalties (at [147]); LFSA’s potential $47.7 million (at [228]); and Mr Trudgian’s potential $37.2 million (at [273]). The asymmetry between stake and risk – particularly the $2.55 million of fully secured, ‘entirely risk free’ funding (at [247]) – was central to the conclusion that it would be unjust for the non-parties to escape liability (at [14], [206], [257], [275]).
The form of the orders matters. Each non-party is jointly and severally liable, with Julian, for the whole of the defendants’ trial costs pursuant to the existing costs orders (at [277]). The defendants may therefore enforce in full against whichever respondent has assets, leaving the non-parties to pursue contribution from one another or from other funders not pursued (at [117], [119]). Quantum remains to be agreed or taxed, and the non-parties retain the right to be heard on it (at [189]). The non-parties were also exposed to the costs of the application itself, having unsuccessfully opposed it (at [278]).
Systemically, the decision recalibrates the economics of informal litigation funding in Western Australia. A broker’s fee carved out of each advance, B Class trust distributions, success-fee ‘employment’ arrangements and proceeds-sharing deeds will each found exposure to the whole of the successful party’s costs. Deregistering funding vehicles after the event will not assist those who stood behind them, and declarations made to ASIC in that process may be deployed against them. Funders who secure their advances so as to eliminate downside risk should expect that feature to weigh in favour of an order, not against it.
6. Worked Example
Suppose the following. Paula commences Supreme Court proceedings against Mercer Holdings Pty Ltd claiming $60 million for fraudulent misrepresentation in the buy-out of her interest in a family company. Paula is asset-poor. Her long-time acquaintance Donald reviews her old files, introduces her to counsel, collates documents, and procures $200,000 from three of his friends under ‘loan agreements’ repayable only out of any recovery, each lender taking a small percentage of the proceeds. Donald’s shelf company, Slateco Pty Ltd (no assets, no trading), holds a deed entitling it to 35% of any recovery. Separately, a broker, Fiona, through her company LFC Pty Ltd, establishes a unit trust: twelve investors subscribe $1.5 million as A Class unitholders; LFC takes B Class units carrying 20% of net income; the advance is secured by a mortgage over Paula’s investment unit; and the funding agreement disclaims all control but requires Paula to consult the trustee on settlement. The claim fails. Mercer’s costs are $2.1 million. The mortgage is called and the investment unit sold, leaving Paula unable to pay.
From Mercer’s perspective, Wright v Lemon [No 5] maps the application. There is no onus to discharge (at [97]) and no need to prove the litigation would have stopped without the funding (at [102]–[110]). Mercer would lead the funding instruments themselves: Slateco’s 35% entitlement and LFC’s B Class units are commercial interests going well beyond recovery of funds advanced (at [113(1)], [147], [219]); the settlement-consultation clauses engage the second factor (at [113(2)], [238]); Paula’s inability to pay engages the third (at [192]); and the absence of a security for costs application against an impecunious individual carries no weight (at [194]–[198]). Donald’s position mirrors Mr Thom’s: he advanced nothing himself, but sourcing funds from others, introducing counsel and working the file are facilitation enough (Maylord at [52]; at [130], [146]). Fiona’s position mirrors Mr Purcell’s: brokerage fees carved from the advances and B Class entitlements are commercial interests, and neither trusteeship nor an absence of control answers the application (at [212], [229]–[230], [239]–[240]). Mercer may pursue Donald, Slateco, Fiona and LFC alone – ignoring the twelve A Class investors – and seek joint and several orders without apportionment (at [119]–[124]).
From the non-parties’ perspective, the available answers are narrow but real. Donald might seek to prove his assistance was founded in friendship without commercial interest – but the Slateco deed forecloses that characterisation; the Hardingham family-assistance category (at [90]) protects only those with no stake in the outcome. If Slateco’s entitlement had been documented as terminated before trial, contemporaneous records of termination would be essential – uncorroborated assertions failed in Wright v Lemon [No 5] (at [176]–[177]). Fiona cannot resist the order by pointing to her trustee companies, but she retains the right to be heard on quantum at taxation (at [189]) and may seek contribution from Donald, Slateco and, potentially, the A Class investors (at [119]). An impecuniosity plea would require cogent evidence and would, at best, inform the structure of the order rather than prevent it (at [182], [245]).
7. Practitioner Guidance: A Step-by-Step Framework
The following framework, derived from the judgment, is directed principally to practitioners acting for a successful party considering recovery against non-parties, with corresponding notes for those advising funders and facilitators.
Step 1: Map the funding network early and in writing. Issue requests under O 9A r 2 of the Rules of the Supreme Court 1971 (WA) at commencement and renew them as the matter progresses, and keep the answers (at [200]–[201]). Even where the answers deny any interested non-party, the correspondence will later defeat a ‘failure to warn’ submission (at [204]). Watch the courtroom: the conduct of a non-party at trial – where they sit, with whom they confer – proved significant evidence of connection (at [140]–[144]).
Step 2: Do not regard the absence of a security for costs application as an obstacle. Where the plaintiff is an individual within the jurisdiction, security is generally unavailable on grounds of impecuniosity, and the failure to apply will not weigh against a later non-party costs application (at [194]–[198]).
Step 3: Frame the application on sufficient connection, not causation. Plead and prove what did happen – who sourced what, on what terms, for what reward – rather than speculating on what would have happened without the funding (at [102]–[110], citing Gore at [61]). The five factors at [113] supply the organising structure, but they are neither a checklist nor closed (at [114]).
Step 4: Prove the commercial interest from the instruments. Obtain the funding agreements, deeds and trust documents by pre-action discovery or orders in the application, and put the percentages in evidence. The decisive findings were drawn from the Barbirolli Deed’s percentages (at [147]), the B Class unit entitlements (at [219], [223], [227]) and the incentive arrangements for introducers (at [266]–[267]).
Step 5: Look through corporate vehicles and trust structures. Identify the natural persons standing behind funding vehicles, their directorships and shareholdings, and any fees flowing to associated entities. Trusteeship is no shield where the trustee’s associates take a share of the proceeds (at [229]–[230]), and post-judgment deregistration of funding vehicles may yield admissions – here, ASIC declarations of assets under $1,000 (at [252]).
Step 6: Establish the unsuccessful party’s inability to pay. Direct evidence of statements by the unsuccessful party, property searches against the party and related entities, and tracing of asset disposals (here, the forced sale of Marri Wood Park and the discharge of the funders’ mortgages) all weighed in favour of the orders (at [187]–[192], [248]–[249]).
Step 7: Seek joint and several orders and resist apportionment. The applicant need neither pursue every funder nor apportion between those pursued; contribution is a matter for the non-parties inter se (at [117]–[124]). Bring the application before taxation – that course is ‘entirely appropriate’, and the non-parties’ right to be heard on quantum answers any prejudice (at [189]).
Step 8: For those advising funders, brokers and facilitators – price the risk and paper the file. Advise that adverse costs exposure attaches to any commercial stake in the outcome, however informal: success-fee consultancies, proceeds-sharing deeds and brokerage structures all sufficed (at [146], [179], [213], [228]). Risk-free structures (fully secured advances) aggravate rather than mitigate (at [247]). If an entitlement is genuinely terminated, document the termination contemporaneously; an uncorroborated account of an oral termination, advanced for the first time in the costs application, will likely fail (at [176]–[177]). Sound document-retention practices matter: the routine deletion of emails and absence of records told against Mr Thom (at [177(5)]).
8. Evidence and Arguments Available to Each Side
For the successful party seeking the order
The applicant’s case is built from documents and observed conduct: the funding agreements, deeds and trust instruments showing entitlements beyond repayment (at [127], [147], [216]–[228]); correspondence evidencing introductions, sourcing of funds and involvement in funding decisions (at [78], [138]); evidence of the non-party’s presence and role at trial (at [140]–[144]); O 9A correspondence and its answers (at [200]); company and property searches tracing structures, deregistrations and asset disposals (at [188], [250]–[252]); and evidence of the unsuccessful party’s inability to pay, including the party’s own statements (at [187]–[188]). Affidavits from the solicitors who conducted the trial proved central, and their contemporaneous file notes and observation evidence withstood challenge (at [67]–[72]).
For the non-party resisting the order
The viable lines of resistance, on the judgment’s own terms, are these. First, absence of commercial interest: assistance ‘founded in family or social ties’ directed at access to justice remains typically insufficient for an order (Hardingham at [19]–[20], quoted at [90]), so a genuine lender or supporter with no share of the proceeds stands apart from the respondents in this case. Secondly, genuine termination of any entitlement before the costs were incurred – but proved by contemporaneous documents, not assertion (at [176]–[177]). Thirdly, public interest: proceedings raising matters extending beyond the parties’ interests engage the fifth factor (at [113(5)]). Fourthly, procedural fairness: a non-party must be afforded a proper hearing, including the opportunity to resist evidence already received (Bischof at 205, quoted at [100]), and retains the right to be heard on quantum at taxation (at [189]). Fifthly, structure: while impecuniosity will not prevent an order, Sangare contemplates that a party’s financial position ‘may inform the structure of a costs order’, such as payment over time (quoted at [182]) – a submission that must be supported by evidence (at [245]). Finally, a non-party held liable may pursue contribution from other funders not joined (at [119]).
9. Key Takeaways for Legal Practice
1. The test in Western Australia is sufficient connection, judged by what is fair and just; causal connection is not a precondition. Kaur does not require a ‘but for’ link between the non-party and the costs incurred; the court will examine what did happen, not speculate about what might have (at [102]–[110], [124]).
2. There is no onus of proof on the applicant. Ballantyne was preferred to Kaur on this point (at [94]–[97]), although applicants should still assemble a complete documentary record, since the facts must persuade the court that an order is fair and just.
3. Facilitators and brokers who advance none of their own money are exposed. Sourcing funds from others, introducing funders, collating documents and working the file on a speculative basis founded orders against both Mr Thom and Mr Purcell (Maylord at [52]; at [130], [146], [213]).
4. Corporate vehicles and trust structures will not shield those standing behind the funding. The court looks to substance: fees carved from advances, B Class distributions and directorships of assetless trustees all pointed to the individuals (at [229]–[230]), consistently with Carter at [55].
5. Control over the proceedings is not necessary where a commercial interest exists. Funding agreements disclaiming control but reserving settlement-consultation rights did not avoid the orders (at [238]–[240], applying Court House Capital at [35]).
6. Timing offers no refuge. Funding provided late in the proceedings – even while judgment is reserved – attracts liability where it secures a share of the outcome (at [231]–[236], applying CPC Patent Technologies at [108]–[109]).
7. The impecuniosity of the non-party is irrelevant, and risk-free funding aggravates. Sangare applies by analogy to non-parties (at [182]–[184]); and a funder whose advance is fully secured, so that it enjoys ‘only upside’, invites rather than avoids an order (at [247]).
8. Joint and several orders, without apportionment, are available against selected non-parties. The applicant chooses whom to pursue; contribution is for the non-parties among themselves; and the application is properly brought before taxation, with quantum protected by the right to be heard (at [117]–[124], [189]).
9. The professional dimension: O 9A answers and document practices will be scrutinised. Denials of third-party funding given on a party’s behalf during the proceedings, and a non-party’s failure to retain documents, each shaped the outcome (at [177], [200]–[204]). Practitioners advising funded plaintiffs should ensure disclosure responses are accurate and kept under review; those advising funders should assume that every instrument, fee and unit entitlement will ultimately be read aloud in court.
10. Conclusion
Wright v Lemon [No 5] settles, at first instance, the two questions that had unsettled non-party costs practice in Western Australia: there is no onus of proof, and there is no causal-connection precondition. What remains is a single, fact-sensitive enquiry – sufficient connection, assessed against what is fair and just – informed by open-ended factors of which a commercial interest in the outcome is the most powerful. On the facts, the decision extends the reach of that enquiry across the full cast of a modern funding network: the friend-facilitator with a proceeds-sharing deed, the broker with B Class units behind corporate trustees, and the investor rewarded for introducing other investors.
The practical message is symmetrical. For successful parties, recovery against those who stood to profit from the litigation is now a well-marked path, available without apportionment and unimpeded by the absence of security for costs or prior warning where the funding was undisclosed. For funders, brokers and facilitators – however informal – the price of a share in the upside is exposure to the whole of the downside. Those who structure, paper and disclose their arrangements accordingly will be best placed; those who rely on labels, corporate shells or after-the-event assertions should expect the courts to look, as Whitby J did, at what was actually done and what was actually promised.