QBE INSURANCE (AUSTRALIA) LIMITED -v- SURE PEOPLE SOLUTIONS PTY LTD [2025] WASC 391 (19 September 2025)

(a) all documents relating to WorkCover WA's 'proper assessment' of QBE's premiums (within the meaning of s 154(4) of the Worker's Compensation and Injury Management Act 1981 (WA) (rep) (former Act) and / or the reasonableness of such premiums proposed by QBE (within the meaning of WorkCover WA's policies) and the extent to which WorkCover WA's assessments and determination were required to reasonably compensate QBE in relation to the Policy in respect of the relevant Period; and

(b) all documents comprising any actuarial model used by WorkCover WA when considering Loading Applications in relation to the Policy in respect of the Relevant Period or when making the assessments or determinations referred to in the preceding paragraph.

1. WorkCover has produced 22 documents in response to the subpoena. There is no objection to inspection and copying of documents 3, 4, 7 and 12. I will refer to the remainder of the documents as the subpoena documents.
2. The subpoena documents concern the actuarial model used by WorkCover to assess applications made under the former Act by the plaintiff for approval to charge workers compensation premiums more than 75% above the recommended premium rate published by WorkCover. The subpoena documents produced largely consist of excel spreadsheets which contain the outputs from the actuarial model and apply the formulas comprising the actuarial model when new data is inserted into the spreadsheet.

1. Before providing an overview of the relevant aspects of the pleadings, it is relevant to note that this case concerns workers compensation insurance premiums. The statutory framework governing workers compensation insurance is relevant to the pleadings and the claimed legitimate forensic purpose of the subpoena documents. It is therefore necessary to give an overview of the relevant aspects of that regulatory framework and the role played by the actuarial model in that framework.
2. On 1 July 2024, the Workers Compensation and Injury Management Act 2023 (WA) (the 2023 Act) came into effect.
3. The period of time relevant to the subpoena documents is between 2017 and 2022, which is when the former Act was in existence. For this reason, the focus of this summary is on the relevant provisions of the former Act, with any relevant differences under the 2023 Act identified.
4. There was no dispute as to the relevant elements of the statutory framework, and the role played by the actuarial model in the framework. Each were helpfully summarised in the affidavits filed by WorkCover, being:

(a) affidavit of Rebecca Harris, General Manager, Regulatory Services at WorkCover affirmed 23 June 2025;

(b) affidavit of Rebecca Harris, affirmed 8 July 2025; and

(c) affidavit of Kevin Gillingham, Manager of Policy and Legislative Services at WorkCover, affirmed 23 June 2025.

1. The below summary is largely taken from these affidavits.
2. WorkCover is a statutory authority, currently constituted under the 2023 Act and, previously, the former Act.
3. Under s 160(1) of the former Act (and also under the 2023 Act), all employers were legally required to hold workers compensation insurance with an approved insurer for which they pay an insurance premium. An insurer was also statutorily required to insure any employer who requests a workers compensation insurance policy under s 160(3) of the former Act.
4. Each financial year, WorkCover sets and publishes recommended premium rates for workers compensation insurance which operate as a guide for insurers and employers when pricing the risk and assessing the premium payable by a particular employer. The purpose of the recommended premium rates is to ensure sufficient premium is collected by insurers to meet the costs of workers compensation claims, without compromising the stability of the scheme, but ensuring that rates are broadly equitable across different industry classes.
5. To assist the WorkCover Board, WorkCover contracts an independent scheme actuary to provide actuarial assessments of recommended premium rates for 517 individual industry classifications. These classifications are identified by reference to industry codes drawn from the Australian Bureau of Statistics' Australian and New Zealand Standard Industrial Classification 2006.
6. The annual change to each recommended premium rate is capped by WorkCover to limit excessive variations.
7. While WorkCover sets and publishes recommended premium rates, the workers compensation scheme is privately underwritten by private insurers. This means that private insurers charge premiums to employers and pay out on insurance claims and WorkCover has no direct role in setting the premiums charged by insurers, subject to the following relevant caveats.
8. The insurer will assign an employer to an industry classification which best reflects the activities of the employer. Where an employer's operations span multiple industries, the insurer will assign a classification which represents the most dominant industry at a specific location.
9. Where the employer is a labour hire company, as is the case with the defendant, the insurer will assign multiple classifications, each reflecting the industry to which workers are placed for work by the labour hire company.
10. Under the former Act, insurers could discount or surcharge recommended premium rates. The only insurance premium pricing restriction placed on insurers under the former Act related to charging premiums with 'loadings' that are at least 75% higher than the recommended premium rate set and published by WorkCover. Section 152 of the former Act provided that unless permitted by WorkCover to do so, an insurer could not charge a loading on a recommended premium rate of more than 75% of that rate unless first obtaining approval from WorkCover.
11. Further, s 154 of the former Act provided for an appeal mechanism for an employer to WorkCover if an employer was dissatisfied with the amount of the premium which an insurer had assessed at the time of the issue or renewal of their policy.
12. Often premium loading applications under s 152 were dealt with by WorkCover at the same time as any appeal by the employer under s 154 of the former Act.
13. Under s 154 of the former Act, WorkCover had a statutory discretion to lay down its own procedure to hear and determine any appeal, and to decide the proper classification or the proper assessment of the premium not exceeding the assessment initially sought by the insurer.
14. The former Act did not prescribe the matters to be considered by WorkCover in making a determination on a premium loading or appeal matter.
15. The process adopted by WorkCover in considering premium loading applications and appeals under the former Act was guided by the Board's Assessment of Premium Rates and Industry Classifications: Loading and Appeal Policy.
16. When considering whether a loading application is approved, the WorkCover Board considers whether a premium sought by an insurer is fair and reasonable.
17. The WorkCover Board considers many factors including information submitted by the parties, the actuarially calculated premium rate using the actuarial model, agency advice, internal policy documents, and the nature of the entity. The WorkCover staff prepared recommendations for the WorkCover Board which then made a decision on the loading application or the appeal. The Board considers the results produced by the actuarial model along with other relevant information in accordance with its policy when determining an outcome for the premium loading or review application.
18. The actuarial model has been prepared by specialist actuaries to assess and produce an indicative premium amount based on a number of factors including claims experience, risk profile of the relevant industry, and size of the employer. The actuarial model uses up to four different actuarial methods to calculate an estimated premium for the employer. The models estimated premium is equal to the average of these methods and considers exclusion if a method results in a significantly different result to the others.
19. However, the affidavit evidence filed by WorkCover explains that the actuarial model was built primarily to reflect the more common scenario in which an employer has one industry classification assigned to it and has a single recommended premium rate. The actuarial model is best fit for that purpose.
20. In the present case, the defendant is a labour hire company. Labour hire companies supply on‑hire workers to different business and therefore have multiple industry classes and a more complex risk profile.
21. Where the results from the actuarial model are not appropriate to be used, such as where an employer is a labour hire company, WorkCover's recommendation to the WorkCover Board is based on an assessment of the proposed premium by considering:

(a) the nature and context of the entity;

(b) the recommended premium rate for the industry or industries of the employer;

(c) the insurer's underwriting process, methodology and rationale in determining the proposed premium;

(d) whether the employer was afforded reasonable opportunity to source alternative quotes from other insurers, and the outcome of the quotes; and/or

(e) any information relevant to the claim experience of the employer.

1. The affidavit evidence explains that whilst the actuarial model is still used during the assessment of premium loading applications and appeals with respect to labour hire companies, WorkCover may place more weight on the other relevant factors. At all times, the actuarial model is a tool only, used by WorkCover in the exercise of its statutory functions.
2. Under the 2023 Act, there is no longer any requirement for insurers to obtain the approval of WorkCover to charge a loading of more than 75% above the recommended premium rates. However, under the 2023 Act employers still have a right of appeal in relation to the premiums charged by their insurer, and one ground of appeal is where the insurer has charged more than 75% above the recommended premium rates. Accordingly, the affidavit evidence provides that the actuarial model continues to be used by WorkCover when undertaking its statutory functions under the 2023 Act. The WorkCover affidavit evidence goes on to explain that the actuarial model is not developed afresh each year, but rather is updated over time.
3. In terms of costs, it costs WorkCover approximately $130,000 in annual fees to the independent scheme actuary to produce the recommended premium rates, and a further $5,500 per annum to create the actuarial model.
4. In terms of the excel spreadsheets that have been produced, each spreadsheet contains underlying formulas and methods which should the outcomes of the actuarial methods that are averaged to provide an actuarially based premium loading. Each spreadsheet is locked against editing, so that WorkCover can only input data in relation to the following limited fields:

(a) the number of cumulative claim reports for the employer;

(b) the number of active claims for the employer;

(c) the value of cumulative claim payments for the employer;

(d) the insurer's value of outstanding case estimates for the employer;

(e) the value of declared wages for employees within that ANZSIC code with the employer for that policy period; and

(f) the value of premium charged by the insurer.

1. WorkCover collates the information, received from the insurer, to insert into the spreadsheet fields and both the insurer and the employer are provided with the opportunity to verify the data.
2. The affidavit evidence deposes that the actuarial model is not a generally accessible tool within WorkCover. It is kept separate from other WorkCover records and is only available to a small group of staff who have the authority given to them on the basis of their duties and responsibilities. Mr Gillingham deposed that he was not aware of any instance where the actuarial model had been copied, distributed or sent electronically or in hard copy externally to any person other than relevant WorkCover employees or professional advisors (including actuaries) and under strict conditions. Ms Harris is not aware of the underlying formulas in the spreadsheets being shared with anyone at WorkCover.
3. Mr Gillingham deposes of the following concerns if the actuarial model is disclosed to the parties in this matter directly:

1. If the Actuarial Model was disclosed to the parties directly, particularly the Plaintiff as one of the 6 approved insurers in the Western Australian workers compensation scheme, the parties would have access to key inputs which determine how premium loading applications and appeals are dealt with and there is a risk that they might reverse engineer the workings and calculations of the model and use this to forecast outcomes to be used in future premium negotiations. 2. If only QBE had access to the Actuarial Model, QBE would have a competitive advantage over the other insurers in the scheme, as it would be able to determine what information to input into the Actuarial Model to increase the premiums that it may charge. 3. If all insurers had access to the Actuarial Model, I believe that it would have a significant inflationary effect on premium negotiations as parties would have information on the likely result of a premium review and could manipulate the system. 4. There is also considerable scope for misuse if the Actuarial Model is disclosed. Assessment of a 'proper premium' using the model relies on correct information from the parties. Information relating to claims experience, number of workers employed, and total remuneration is often complex and contested and requires independent scrutiny, as employers and insurers have vested interests. There are likely to be significant variations in the calculated 'proper premium' if parties with vested interests are inputting their own unverified and unchecked data into the Actuarial Model and know the peculiar sensitivities in the Actuarial Model which might produce the highest incentives for inputting incorrect data. 5. If the Actuarial Model was accessible to the market, in my opinion:

(a) the WorkCover WA Board would likely be faced with a significant increase in premium review applications arising from misuse or miscalculation by parties and the input of incorrect data; and
>
> (b) this scenario would, in turn, undermine the intent of WorkCover WA's role in determining what a fair and reasonable premium is in an employer's specific circumstances, and capacity to modify the effect of market forces.

1. Ultimately, if the Actuarial Model were disclosed without a confidentiality undertaking, I would be advising WorkCover WA that:

(a) the integrity of WorkCover WA's legislated premium loading and review process can no longer be relied upon to produce independent outcomes for the Western Australian community;

(b) WorkCover WA should change its approach to premium reviews (noting the premium loading process has been discontinued under the 2023 Act and the premium review process and criteria has changed); and

(c) WorkCover WA needs to urgently brief a new Scheme Actuary to prepare a replacement actuarial model with the attendant cost in time and resources involved in doing so to the Western Australian public.

1. Ms Harris deposes to the following concerns:
   1. While the Excel spreadsheets contain 'locked down' formulas, based on my use of those documents I believe the ability to enter different data sets and parameters would enable the results to be used to 'reverse-engineer' the specific actuarial assessment methods and predict the results from the Actuarial Model.
   2. I also am aware that the Excel spreadsheets are only protected using in-built protections within Microsoft Excel. The spreadsheets do not have any more advanced encryption than those in-built protections.
   3. An openly accessible Actuarial Model could have a significant inflationary effect on premium negotiations, with information accessible on the likely result of a premium review process. An insurer could set premiums with the knowledge of the likely information to be considered by the WorkCover WA Board if the premium were appealed.
   4. A stakeholder in possession of the Actuarial Model would have a significant competitive advantage. Ultimately, disclosure of the Actuarial Model would compromise the integrity of WorkCover WA's legislated premium review process, deeming it redundant and unable to produce independent outcomes for the Western Australian community.
   5. This in turn would undermine the intent of the WorkCover WA Board's role in determining what a fair and reasonable premium is in an employer's specific circumstances, and capacity to modify the effect of market forces.
   6. In my view, if the Actuarial Model was provided to the parties in these proceedings (rather than the solicitors and approved experts bound by the proposed Undertaking), the WorkCover WA Board would likely have to change its approach to premium reviews and replace the Actuarial Model. In my view, this change would be necessary even if QBE only provided copies of the Actuarial Model to a small number of staff within QBE.
   7. I do not believe there is any way to continue using the Actuarial Model if one of the licensed Insurers competing in the market is able to charge a premium based on the expected results of WorkCover WA's Actuarial Model.
   8. Whilst I do not know how much it would cost to have Scyne, or another Scheme Actuary, create a wholly new actuarial model to replace WorkCover WA's Actuarial Model, the cost to update the current model annually is significant to both Scyne and WorkCover WA.

Relevant legal principles The court's jurisdiction

1. The court's jurisdiction in relation to subpoenas is governed, in part, by O 36B r 8A(2) of the RSC which provides that:

On a request by the addressee, a party or any other person with a sufficient interest, the Court may, by order -

(a) set aside a subpoena or part of it; or

(b) make or vary directions in relation to removing from and returning to the Court, and the inspection, copying and disposal, of any document or thing that has been or is to be produced under it; or

(c) grant other relief in respect of it.

1. The court is also able to set aside a subpoena in the exercise of its inherent jurisdiction.[1]

1. It is well established that a subpoena may be set aside where it does not serve a legitimate forensic purpose.
2. The relevant principles concerning applications to set aside subpoenas, with particular regard to the asserted lack of a legitimate forensic purpose, were outlined by Hill J in Hongkong Xinhe International Investment Company Ltd v Bullseye Mining Ltd [No 4],[2] which I adopt for the purposes of the present application:

(a) Order 36B r 8A(2) ... provides that on a request by a party, the court may set aside a subpoena or part of it;

(b) the court will set aside a subpoena if it has no legitimate forensic purpose. In determining whether a subpoena has a legitimate forensic purpose, the court will consider whether there is a reasonable possibility or 'it is on the cards' that the documents sought in the subpoena will materially assist the claim or the defence of the proceedings or are relevant to an issue in the proceedings. A document may provide material assistance even if it is not admissible in the proceedings;

(c) a legitimate forensic purpose extends to material which may relate to the cross-examination of a witness, including documents which relate only to credit;

(d) in determining whether the documents have apparent relevance to the issues in the proceedings, the court must consider the issues that potentially arise on the pleadings, including the particulars that have been provided;

(e) the court will set aside a subpoena where the subpoena is an abuse of the process of the court. A subpoena may be an abuse of process where it may be characterised as 'fishing', being used for the purpose of obtaining discovery against a third party, or oppressive;

(f) in determining whether a subpoena is properly regarded as fishing, I have had regard to the statement of Owen J in Associated Dominions Assurance Society Pty Ltd v John Fairfax & Sons Pty Ltd:

A 'fishing expedition', in the sense in which the phrase has been used in the law, means, as I understand it, that a person who has no evidence that fish of a particular kind are in a pool desires to be at liberty to drag it for the purpose of finding out whether there are any there or not. If, however, there is material before the court pointing to the probability that a party to litigation has in his possession documents tending to destroy his case or support the case of his opponent and that privilege from inspection of such documents has been wrongly claimed, an application by that opponent to be allowed to inspect them cannot properly be described as a mere 'fishing expedition'.

(g) the fact that documents which are the subject of a subpoena may also be sought in discovery does not, of itself, mean that the subpoena is an abuse of process.

1. In Simon Dirk Kenworthy-Groen as Administrator of the Estate of William Grove v Grove,[3] the Court of Appeal approved the following propositions in connection with whether there as a legitimate forensic purpose, by reference to the Full Court decision in Stanley v Layne Christensen Co:[4]
   1. A legitimate forensic purpose will be established if a document gives rise to a line of enquiry which is relevant to the issues before the trier of fact.
   2. In assessing whether a legitimate forensic purpose exists it must be borne in mind that the necessity for having a document to fairly dispose of the issues at trial might well not become apparent before trial. Thus, whether a document is necessary to fairly dispose of proceedings is to be understood in the broad sense of embracing any document which has apparent relevance even if it might not readily be seen, at the pre-inspection stage, necessarily to be admissible in evidence.
   3. No narrow view ought to be taken as to the legitimate purposes of a subpoena.
   4. In the interests of a fair trial, litigation should be conducted on the footing that all relevant documentary evidence is available.
2. Accordingly, 'apparent relevance' is a low threshold.[5]

1. The principles concerning the application of public interest immunity in the context of a subpoena were outlined by the Court of Appeal in The Australian Statistician v Leighton Contractors Pty Ltd as follows:[6]
   1. The general law is that a court will not order the production of information or a document, although relevant and otherwise admissible, if it would be injurious to the public interest to disclose it: Sankey v Whitlam (at 38).
   2. An objection may be made to the production of information or a document because it would be against the public interest to disclose its contents or because it belongs to a class of documents which in the public interest ought not to be produced: Sankey v Whitlam (at 39); Commonwealth v Northern Land Council [1993] HCA 24; (1993) 176 CLR 604 at 616. The parties accepted that public interest immunity cannot be waived: Air Canada v Secretary of State for Trade (No 2) [1983] 2 AC 394 at 436.
   3. Public interest immunity is not absolute. It has to be weighed in the balance with the competing public interest in the due administration of justice: Sankey v Whitlam (at 38-39).
2. Public interest immunity protects information which must be kept secret for the protection of government at the highest levels and in sensitive areas of executive responsibility.[7]
3. A claim for public interest immunity is usually supported by an affidavit outlining the reason and basis for the claim.[8]
4. In assessing a claim for public interest immunity, the court must therefore consider the following matters:[9]

(a) would disclosure be contrary to the public interest?;

(b) is the document relevant to a contentious issue in the case?; and

(c) (if the answer to both (a) and (b) is yes) whether the competing interests nonetheless require disclosure?

1. In answering the first question, the court must form a view as a matter of real possibility, as opposed to a matter of probability. The court must also assess the effect of disclosure on the basis of unrestricted production or disclosure for the purposes of the litigation, and not assuming some form of limited disclosure.[10]
2. In answering the third question, the public interest in withholding the documents must be weighed against the public interest in the administration of justice.[11]

1. It is only necessary to provide an overview of the key elements of the pleadings that are relevant to the application.
2. In broad terms, the plaintiff's case against the defendant is for payment by the defendant of a liquidated debt consisting of unpaid policy premiums due under various workers compensation insurance contracts between 2017 and 2022. The plaintiff submits that their case is a debt collection matter. The total amount claimed is in the vicinity of $3.5 million. The key relevant pleaded elements of the plaintiff's case are as follows:

(a) the plaintiff and the defendant entered into various workers compensation insurance contracts dated between 2017 and 2022;

(b) a term of each of the policies was that the defendant would pay to the plaintiff a premium, including any adjustment premium, due to the plaintiff and calculated in accordance with the terms of the policies;

(c) the defendant was required to pay an initial renewal premium which would be calculated based on an estimate provided by the defendant of the remuneration that it expected to pay to its workers during the relevant policy period;

(d) the defendant was then required to provide statements of the aggregate amount of all remuneration actually paid during the relevant policy period within one month from the expiry of that relevant period;

(e) if the total remuneration paid differed from the estimated information provided by the defendant, the defendant was required to pay any further adjusted premium to the plaintiff;

(f) the defendant provided the required remuneration estimates to the plaintiff;

(g) the plaintiff applied to WorkCover for approval to load the defendant's premiums in excess of 75% of the recommended premium rate in respect of the policy years 2017, 2019, 2020, 2021 and 2022. Those approvals were granted by WorkCover in respect of certain industry classifications;

(h) in purported compliance with its obligation to provide the aggregate amount of all remuneration actually paid, the defendant provided the plaintiff with aggregate information for the 2017, 2018 and 2021 policy years. However, the plaintiff pleads that, in breach of the contracts, the defendant did not in fact provide any of the aggregate amount of remuneration paid for each of the 2017, 2018, 2019 or 2020 policy years;

(i) the actual remuneration paid by the defendant to its employees for each of the 2017, 2018, 2019 and 2020 policy years was higher than the estimate previously provided. The plaintiff then, in accordance with the terms of the policies, calculated an adjustment premium ($1,305,866.34) due to the plaintiff for each of those policy years. The defendant has refused to pay the adjustment premiums despite demand;

(j) following the provision by the defendant of the aggregate remuneration for the 2021 policy year, the plaintiff calculated the two adjustment premiums for the 2021 policy year (a total of $2,063,248.51). The defendant has refused to pay the adjustment premiums despite demand; and

(k) following the provision by the defendant of the aggregate remuneration for the 2022 policy year, the plaintiff calculated the adjustment premium for the 2022 policy year ($213,253.54). The defendant has refused to pay the adjustment premium despite demand.

1. The plaintiff also alleges that the defendant made misrepresentations in its proposals for policy renewals and breached the terms of the policies by failing to keep or provide certain required information and records relevant to the calculation of the premiums.
2. The defendant has filed a defence and counterclaim. An aspect of that counterclaim is a plea that the plaintiff's conduct contravened s 21 of the Australian Consumer Law, and constituted unconscionable conduct.
3. The key relevant pleaded elements of the unconscionable conduct counterclaim are as follows:

(a) the defendant accepts that the plaintiff and the defendant entered into various workers compensation insurance contracts dated between 2017 and 2022;

(b) a term of each of the policies was that the plaintiff would charge a premium calculated using a premium rate not more than 75% in excess of the premium rate recommended by WorkCover for the industry or industries in which the defendant was engaged in business;

(c) it was also a term of the policies that the premium payable by the defendant under that agreement would be calculated on the amount estimated by the defendant as being payable by the defendant to workers and contractors engaged or to be engaged by the defendant during the term of that policy;

(d) the defendant was then obliged to provide a statement of the total amounts paid by the defendant to workers and contractors engaged by the defendant during the term of that agreement within one month of the expiration of that agreement and the premium payable by the defendant under the agreement would thereafter be adjusted to reflect the amounts actually paid by the defendant to workers and contractors during the term of that agreement;

(e) the defendant admits that it neglected to provide statements of the total amounts paid by the defendant to workers and contractors within one month of the expiration of the 2017, 2018, 2019, 2020 and 2021 policy periods;

(f) the defendant pleads that the plaintiff knew of the defendant's failure to provide these statements shortly after each statement was due;

(g) the plaintiff then charged the defendant an increased premium above the recommended premium rate. That increased premium was sometimes less than an amount in excess of 75% of the recommended premium rate. However, on some occasions it was a premium loading of an amount in excess of 75% of the recommended premium rate, and it was charged without the prior approval of WorkCover, in contravention of s 152 of the former Act;

(h) the plaintiff then applied to WorkCover for approval to increase the premium rate in excess of 75% of the premium rate recommended by WorkCover applicable to the defendant in respect of the 2018, 2019, 2020, 2021 and 2022 policy agreements;

(i) when the plaintiff applied to WorkCover, the plaintiff did so knowing that the defendant had not provided the statements within one month of the expiration of each of the relevant agreements; without having exercised its right under the contract to review the defendants' documents; and without having told WorkCover of these two matters;

(j) further, when the plaintiff applied to WorkCover, the plaintiff, in support of its applications asserted, in effect:

that the defendant's claims history and the plaintiff's profit margin on each agreement justified very large increases in the premium rate recommended by WorkCover WA for the industries in which the defendant was engaged, when the plaintiff's assessment of the defendant's claims history and the plaintiff's profit margin on each agreement was detrimentally affected by the understatement of the total amounts paid by the defendant to Workers and Contractors engaged by the defendant during the term of each of those agreements and in circumstances where the plaintiff would subsequently seek to make adjustments for the total amounts paid by the defendant to Workers and Contractors engaged by the defendant during the terms of each of those agreements;

(k) WorkCover then approved the plaintiff charging 'substantially higher' premiums;

(l) the plaintiff's intent in making each of the applications to WorkCover, and WorkCover's purpose in approving each of the applications, was to ensure that the plaintiff received an adequate return on each of the agreements;

(m) the plaintiff then adjusted the premium payable by the defendant in respect of each of the relevant agreements in accordance with the higher premium rate approved by WorkCover;

(n) the above conduct was in trade or commerce;

(o) in those circumstances, the conduct of the plaintiff in obtaining the premium loading approval, and then charging the adjusted premium rates to the defendant based on the updated figures, was conduct that contravened s 21 of the Australian Consumer Law in that it was unconscionable conduct in the supply of services; and

(p) the defendant has suffered loss and damage as a result of this unconscionable conduct as the defendant will be liable to pay amounts in excess of what is fairly due under each of the 2018, 2019, 2020, 2021 and 2022 agreements.

1. The defendant has other defences including that the plaintiff breached the terms of the policies and that the policies are void and also a set off. But these are not relevant to this application.
2. The defendant also has a third party claim against its insurance broker. However, it is not submitted that this third party claim is relevant to this application, and the third party did not appear at the hearing of this application.

Legitimate forensic purpose The parties' submissions

1. The defendant submits that the subpoena documents are relevant to the unconscionable conduct counterclaim in two ways.
2. First, the defendant submits that the subpoena documents are relevant to its claim for unconscionable conduct because they go to the question of whether the premiums charged by the plaintiff to the defendant were fair and reasonable when all accurate data is considered.
3. Counsel for the defendant submits that, in essence, the defendant's case is that in obtaining approval for premium loading, the plaintiff relied on the estimated remuneration figures from the defendant, and obtained approval to charge a premium above 75% of the recommended premium. The decision by WorkCover to approve the premium loading applications was made by applying the actuarial model to the outdated data (amongst other matters), to arrive at a conclusion as to what represented a fair and reasonable premium for the plaintiff to charge.
4. The defendant submits that the subpoena documents are relevant to the defendant's case in this regard, as the defendant seeks to be able to input the accurate remuneration data into the spreadsheets and to see the conclusion which will be relevant to whether the premiums charged are fair and reasonable.
5. Secondly, the defendant submits that the subpoena documents are relevant to whether the adjustment premiums the plaintiff has charged to the defendant represent a degree of double counting, on the basis that the plaintiff has proceeded to charge an adjustment premium by using the relevant updated final remuneration figures, in circumstances where the WorkCover approved premium rates were based on the estimated remuneration figures and not the actual remuneration figures. The defendant's case is that this involves a degree of double counting, and will result in the defendant being liable to pay amounts in excess of what is fairly due under each of the agreements. This is because, in obtaining premium loading approval, the plaintiff has already been compensated for the extent of underperformance of its contracts, and it now seeks further compensation by applying the approved premiums to the actual remuneration figures.
6. The defendant submits that the subpoena documents are relevant to the defendant's case in this regard, as the defendant seeks to be able to input the accurate remuneration data into the spreadsheets as part of its attempt to assess and demonstrate the extent to which the approved premiums have already compensated the plaintiff with a fair and reasonable premium and the extent to which the application of the adjustment premiums go beyond a fair and reasonable premium.
7. The defendant submits that the allegation of unconscionable conduct is directed towards the conduct of the plaintiff and not of WorkCover. The defendant submits that the statutory process is irreversible. However, in order to demonstrate why the plaintiff's conduct is unconscionable, it is necessary to demonstrate what the plaintiff has already been compensated for in the statutory process, and compare that to what is now being charged under the various contracts.

1. WorkCover submits that as a non-party to the proceedings, and given the nature of the pleadings, it is not in the best position to comprehensively assist the court with the question of whether there is a legitimate forensic purpose for the subpoena documents. Counsel for WorkCover also acknowledged that the defendant's oral submissions had evolved since their written submissions on this issue.
2. Ultimately, WorkCover submits that, based on the affidavit evidence, it was unclear the extent to which the WorkCover model would be critical or relevant evidence given that in the case of labour hire companies, the actuarial model is not (and was not in the case of the defendant) a determinative or even key factor in premium loading applications.

1. Counsel for the plaintiff only made limited submissions in relation to the application, noting that the subpoena is directed to WorkCover.
2. However, the plaintiff submits that it was not clear how a legitimate forensic purpose arose.
3. The plaintiff submits (and has pleaded) that there was no application for judicial review of the decisions of WorkCover in relation the premium loading applications. Accordingly, the plaintiff submits that it is not permissible for the defendant to now attempt to go behind the WorkCover premium loading decisions in the form of a collateral attack on those decisions. The plaintiff submits that to the extent a question arises as to the premium charged by the plaintiff, the plaintiff's actuarial model has been discovered, and the appropriate course would be for the defendant's experts to use that model, as opposed to the WorkCover model.
4. Further, the plaintiff did not accept that the premium loading applications were made by the plaintiff on the basis of the remuneration estimates. Rather, the plaintiff submits (and pleads) that they were made on the basis of inaccurate final remuneration figures provided by the defendant. The plaintiff submits that the figures used were purportedly provided by the defendant in accordance with its contractual obligation, but it turns out that the defendant provided inaccurate figures and declined to allow the plaintiff to inspect the relevant records.

1. I am satisfied in all the circumstances that there is a legitimate forensic purpose for the subpoena documents. There is a reasonable possibility that the subpoena documents sought are relevant to an issue in the proceedings, being aspects of the defendant's counterclaim for unconscionable conduct in two respects.
2. First, the subpoena documents are relevant as to whether the premiums charged by the plaintiff to the defendant, were fair and reasonable. The WorkCover actuarial model is one way in which what is a 'fair and reasonable' premium might be established. Whilst the defendant accepts that there are other ways this could be done (including using the plaintiff's model or the model used by other experts), given the role played by WorkCover in the workers compensation insurance market, I am satisfied that analysing the WorkCover actuarial model in this way is relevant to the counterclaim. Accordingly, whilst in many circumstances what a third party may consider to be a fair and reasonable premium may not be relevant, the position is different in this case.
3. Whilst the evidence before the court is that in the case of a labour hire company the reliance placed by WorkCover on the actuarial model in premium loading applications is not as significant as it is in cases not involving labour hire companies, it is not the case that the actuarial model is not used in such cases, or that it was not used in relation to the plaintiff's applications. The production of the documents demonstrates that it was used in the plaintiff's applications. Therefore, I am satisfied that the subpoena documents are required for a fair disposal of this aspect of the action.
4. Secondly, the subpoena documents are relevant as to whether the adjusted premiums charged by the plaintiff (by using the approved premiums as applied to the accurate remuneration figures) include an element of double counting. The WorkCover actuarial model spreadsheets are relevant to this aspect of the counterclaim as an analysis of the effect of the accurate remuneration figures is part of the defendant's case.
5. Accordingly, I am satisfied there is a reasonable possibility that the subpoena documents may give rise to a line of inquiry in relation to the unconscionable conduct claim, and enable the parties to appraise the strengths and weaknesses of their and their opponent's case.
6. In so far as the plaintiff's case is that the defendant's unconscionable conduct claim involves an impermissible collateral attack on the premium loading decisions of WorkCover, this is not the appropriate stage in the proceedings to make a final determination on this issue. The parties have joined issue on this pleading, and it will be a matter to be determined at trial.
7. However, for the purposes of considering whether the subpoena should be set aside, my preliminary view is that it is not sufficiently clear that the defendant is seeking to challenge or set aside the WorkCover decisions, such that it can be said that there is no legitimate forensic purpose to the subpoena documents. On its face, it appears that the defendant is seeking the subpoena documents to understand the extent to which the WorkCover decisions were impacted by the lower estimated remuneration figures, and the extent to which there has been an element of double counting. That purpose assumes the existence of the WorkCover decisions.
8. I am therefore satisfied there is a legitimate forensic purpose for the subpoena documents.

1. WorkCover relies on the affidavit evidence of Rebecca Harris and Kevin Gillingham in support of its submission that disclosure of the subpoena documents would be contrary to the public interest. WorkCover submits that one of its primary functions is acting as a safety net for employers and volatile market forces to ensure that workers compensation insurance is accessible and sustainable. In broad terms, WorkCover submits that the following matters of public harm may arise if the subpoena documents are disclosed:

(a) there is a risk that if the parties have access to the subpoena documents, the parties might be able to reverse engineer the workings and calculations of the WorkCover actuarial model and use that analysis to forecast outcomes in future premium negotiations or employer appeals;

(b) if the plaintiff itself has access to the subpoena documents they would have a competitive advantage over the other insurers in the scheme;

(c) if all insurers have access to the subpoena documents, there could be a significant inflationary effect on premiums and premium negotiations; and

(d) given the risks identified in (a) - (c) above, and the fact that the actuarial model is still used today by WorkCover, if the subpoena documents were disclosed, the WorkCover Board may need to change its approach to employer appeals and urgently brief a new scheme actuary to prepare a replacement actuarial model, which would cost additional money and have time implications for upcoming reviews and the system generally.

1. The defendant accepts that as the WorkCover actuarial model has never been disclosed previously, it is to be expected that WorkCover's submissions regarding the effect of disclosure will contain a degree of speculation. However, the defendant submits that there is a degree of illogicality to some of WorkCover's submissions. In broad terms, the defendant submits as follows:

(a) to the extent WorkCover is expressing a concern that the disclosure of the subpoena documents may result in a degree of collusion between stakeholders in the market to undercut premium pricing, and that this may push premiums to a point of being unaffordable, the defendant accepts this is a risk that would properly concern WorkCover. The defendant accepts that a stable workers compensation insurance market is of public concern. However, the defendant submits that it is then a matter of balancing this risk with the public interest in documents relevant to litigation being accessible to the parties;

(b) WorkCover has not clearly articulated the extent to which misuse of the actuarial model could occur from the spread sheets alone;

(c) the present legislative regime is very different to that in The Australian Statistician v Leighton Contractors Pty Ltd[12] and there are no similar confidentiality provisions in the former Act or the 2023 Act, and not the same situation arising where information is provided by participants in the market based on statutory confidentiality provisions; and

(d) if there are risks associated with the disclosure of the subpoena documents, those risks only exist if there is unrestricted disclosure of the subpoena documents, and the risks can be managed by a restricted inspection regime consisting of inspection to the parties' lawyers and experts only, in the security of the court registry and subject to a confidentiality regime in similar form to that in Global Smart Cities Pty Ltd v City of Wanneroo [No 3].[13]

1. The plaintiff did not seek to be heard over and above the submissions by WorkCover, and sought only to make submissions on the terms of any order permitting limited disclosure, if the court was minded to make such an order.

1. When considering WorkCover's claim for public interest immunity, I must consider the following three questions:

(a) would disclosure be contrary to the public interest?;

(b) is the document relevant to a contentious issue in the case?; and

(c) (if the answer to both (a) and (b) is yes) whether the competing interests nonetheless require disclosure?

1. I am satisfied in all the circumstances that WorkCover, as an arm of the executive, plays an essential role in regulating and maintaining a functioning and affordable workers compensation scheme. I have outlined the details of that scheme earlier in these reasons. I am also satisfied that it is in the public interest that a functioning and affordable workers compensation scheme exists. It has benefits to workers (and their families), employers and also insurers, and by extension, society as a whole.
2. The evidence before me details the role that the WorkCover actuarial model plays in maintaining that system. The actuarial model is used by WorkCover not only to assist in setting the recommended premiums, but also (currently) to consider employer appeals in relation to the premiums charged by insurers. The WorkCover actuarial model is confidential in that it is not shared with persons outside of WorkCover, and only a limited number of persons within WorkCover have access to it.
3. I am further satisfied, based on the evidence from WorkCover, that if the actuarial model (in the form of the spreadsheets) is disclosed, there is a real possibility that it may enable stakeholders in the market to analyse the spreadsheets (and the information contained therein) and so as to assist in forecasting the outcomes of employer appeals such that there may be an 'undercutting' of the market and inflationary effects on premiums. This could have a negative effect on the workers compensation market in Western Australia and is contrary to the purpose of the statutory role and purpose of having WorkCover and the various statutory processes. In reaching this conclusion, I have had regard to the highly regulated nature of the workers compensation insurance market and the small number of insurers within that market.
4. Further, I am satisfied that as a consequence, there is a real possibility that WorkCover may, given its statutory responsibilities and as part of maintaining a functioning and affordable workers compensation scheme, need to take steps to prevent this happening (or remove the risk if it occurs) by developing a new actuarial model. This will cost both money and time and may in turn impact employer appeals.
5. In all the circumstances, I am therefore satisfied that disclosure of the subpoena documents is contrary to the public interest.

Are the subpoena documents relevant to a contentious issue?

1. For the reasons outlined above when considering whether a legitimate forensic purpose exists, I am satisfied that the subpoena documents are relevant to the defendant's claim of unconscionable conduct and are therefore relevant to a contentious issue.
2. Further, the subpoena documents are of central importance to the defendant's claim, as at its heart, that claim depends upon the WorkCover premium approvals and their effect on the premiums charged by the plaintiff to the defendant under the various contracts.

Do the competing interests nonetheless require disclosure?

1. As I have answered 'yes' to each of the above two questions, it remains for me to balance these two competing interests and determine where the public interest lies.
2. In all the circumstances, I am not satisfied that unrestricted disclosure of the subpoena documents is appropriate. Notwithstanding the centrality of the documents to the defendant's unconscionable conduct case, given the nature of the workers' compensation market in Western Australia, and the continuing statutory role played by WorkCover (and its actuarial model) in that market, I consider that the public interest does not favour the unrestricted disclosure of the subpoena documents. I consider the risks of so doing are contrary to the public interest.
3. However, I am of the view that a restricted form of inspection of the subpoena documents is appropriate. I consider the appropriate balance is to permit only identified counsel, solicitors and expert witnesses to inspect and access (but not uplift) the subpoena documents, subject to each signing appropriate confidentiality orders. I will hear further from the parties as to the precise terms of the access arrangements and confidentiality orders. However, as the documents are largely electronic versions of excel spreadsheets, I consider the confidentiality regime should (as a minimum) permit identified individuals to attend the offices of the Supreme Court registry at pre‑arranged times to inspect the spreadsheets and input data. No person will be permitted to make a copy of the subpoena documents. All persons being granted access will also be required to sign a confidentiality undertaking pursuant to a court order that they will only use the material they inspect for the purposes of the litigation in these proceedings, even after they have prepared reports or given evidence at any further hearings or trial and limiting the persons with whom they can discuss or reveal the information contained in the subpoena documents.
4. I consider this is the appropriate balance as it will ensure that documents which are relevant to, and an essential element of, the defendant's unconscionable conduct claim can be accessed by the legal representatives and expert witnesses to prosecute and defend this case. This process will also ensure that the subpoena documents are not more widely disseminated (including accidentally) and risk the matters identified above.

1. For the above reasons, I am of the view that the subpoena issued to WorkCover on 11 April 2025 should not be set aside.
2. However, I consider that a restricted inspection regime for the subpoena documents is appropriate. I will hear further from the parties as to the appropriate wording of orders to give effect to my reasons, including an appropriate access and confidentiality regime.

I certify that the preceding paragraph(s) comprise the reasons for decision of the Supreme Court of Western Australia.

[1] Jensen v Nationwide News Pty Ltd [No 6] [2018] WASC 415 [26].

[2] Hongkong Xinhe International Investment Co Ltd v Bullseye Mining Ltd [No 4] [2021] WASC 287 [11].

[3] Simon Dirk Kenworthy-Groen as Administrator of the Estate of William Grove v Grove [2022] WASCA 120 [51].

[4] Stanley v Layne Christensen Co [2004] WASCA 50 [9].

[5] Simon Dirk Kenworthy-Groen as Administrator of the Estate of William Grove v Grove [2022] WASCA 120 [53].

[6] The Australian Statistician v Leighton Contractors Pty Ltd [2008] WASCA 34; (2008) 36 WAR 83.

[7] The Australian Statistician v Leighton Contractors Pty Ltd [2008] WASCA 34; (2008) 36 WAR 83 [33].

[8] Kirby v Prisoners Review Board [No 2] [2010] WASC 280 [10] ‑ [14].

[9] Kirby v Prisoners Review Board [No 2] [2010] WASC 280 [5] ‑ [6].

[10] Australian Statistician v Leighton Contractors Pty Ltd [2008] WASCA 34; (2008) 36 WAR 83 [46].

[11] Jensen v Nationwide News Pty Ltd [No 6] [2018] WASC 415 [51].

[12] The Australian Statistician v Leighton Contractors Pty Ltd [2008] WASCA 34; (2008) 36 WAR 83.

[13] Global Smart Cities Pty Ltd v City of Wanneroo [No 3] [2023] WASC 395, Annexure A.