Eighteen Thousand a Head: When the Book Build Is the Asset

Eighteen Thousand a Head: When the Book Build Is the Asset

The Commonwealth agreed to pay A$18,000 for every eligible claimant who registered in the Northern Territory stolen wages class action, so the size of the settlement was set by the outreach program rather than by the case. The funder earned A$3,600 on each person found. The firm that did the finding billed by the hour.

---

I have no exposure to this and never will. My litigation-finance money sits in commercial funds backing single high-value claims — no mass torts, no class actions, nothing that resolves by asking thousands of people to fill in a form. I read McDonald v Commonwealth of Australia anyway because it is 146 pages of a court doing the thing my own reporting never does: taking apart who was paid on what basis, and asking whether the basis made sense given what the payment was for.

The answer in this case is unusually clear, because the settlement had no fixed size.

A settlement that had to be built before it existed

Minnie McDonald sued the Commonwealth in 2021 on behalf of Aboriginal and Torres Strait Islander people who worked in the Northern Territory between 1933 and 1971 under wage control legislation and were paid little or nothing. The parties mediated and settled in 2024. Chief Justice Mortimer approved the settlement in April 2025.

The principal term is worth reading slowly. The Commonwealth agreed to pay “a sum of up to $180,000,000 to be calculated by multiplying $18,000 by the number of eligible claimants up to 10,000 eligible claimants.” Separately and on top of that, it agreed to pay up to A$15 million toward the applicant’s legal costs, up to A$6 million toward the administrator’s costs and up to A$1 million for a costs assessor. A$202 million is the sum of those four numbers, and it is a ceiling rather than a price.

So this is not a fund to be divided. There is no pot sitting in an account waiting for a distribution formula. Every eligible claimant who registers causes A$18,000 to be created; every eligible claimant who is never found causes nothing. The money does not go to a cy-près recipient or revert to a residue or get shared among the people who did register. It is simply never paid by anyone to anyone.

Which means the outreach program is not an administrative cost of the settlement. It is the thing that produces the settlement. A class action of this shape is worth what its book build makes it worth.

Two people were paid to find claimants, on opposite bases

LLS Fund Services funded the case. It asked the court to approve a commission of 20% of the gross A$202 million — the settlement sum plus all three costs components. Mortimer CJ declined and approved 20% of the net settlement sum instead, capped at 8,750 eligible claimants, which works out to a maximum commission of A$31.5 million. She also refused the funder’s attempt to charge commission on the fees of the amici curiae the court had itself appointed to scrutinize the funder’s claim: “There is no justification for this.”

Strip that down and the funder earned 20% of A$18,000, or A$3,600, for each eligible claimant located, up to 8,750 of them. Above that number it earned nothing further. Below it, every person not found cost the funder A$3,600 of its own return.

Shine Lawyers ran the outreach. It used its own lawyers, law clerks and paralegals, charged an hourly rate, and sought an uplift fee on those costs. Its revenue from the exercise was a function of hours worked, not of people located.

Three of those six lines come out of the claimants’ money and three do not, which makes this a better-structured deal for a class than most of what I read. The problem is not the split. It is that the party whose fee scaled with hours was the party deciding how the hours were spent, and the party whose return scaled with people found had bought that work rather than performed it.

What the court found about the hours

Mortimer CJ appointed independent counsel to examine three specific questions: whether it was fair and reasonable to use Shine’s own lawyers and clerks for the outreach rather than people from locally based organizations, whether it was fair and reasonable to charge an hourly rate for that work, and whether it was fair and reasonable to charge an uplift on it.

Her findings were direct. There had been “an excessive level of human resources applied to the conduct of this proceeding by Shine, and applied in a way which was not conducive to the most effective and efficient conduct of the proceeding.” Work categorized and charged as legal work was “logistical or administrative in nature,” so the rates charged for it were excessive. She recorded a failure to even consider using experienced local organizations. On the costs assessor’s figures for the work to 30 October 2024, Shine’s professional fees came down from A$8,118,203.70 incurred to A$7,631,111.48 allowed, and the uplift from A$699,159.23 to A$577,990.67. A further A$8 million sought for the registration process was refused, with a direction to come back once the outreach had concluded so the court could assess what more was reasonable.

The comparison she drew is the one that matters. Three local organizations — Urapuntja Corporation, ARDS and Lutheran Care — had submitted proposals covering roughly a third of the communities on the court’s list. Extrapolating to full coverage, she found “it is probable that the local Aboriginal organisations could have undertaken this work for well under $1 million and possibly closer to $500,000.”

Set that against the arithmetic of the book build. At A$3,600 of funder commission per person located, the difference between the two approaches is not a rounding item on a costs assessment. It is the difference between a settlement of one size and a settlement of another.

The count

The registration date was fixed by consent at 31 August 2025 and later extended to 3 October. At the August 2025 hearing the court was told that 5,761 people had lodged claims, against the estimate of up to 8,750 eligible claimants that had been used to set the funder’s commission cap. Registrations continued to the extended deadline, and the administrator’s final eligibility determination has not been published, so the closing number is not public and the gap cannot be stated precisely. Registrants also lodge more than one claim each in most cases — a living worker generally registers for a parent and sometimes a deceased spouse as well — so claims lodged and eligible claimants are not the same count.

What is public is the direction of the miss and roughly its scale. The ABC has reported hundreds of people in north-east Arnhem Land who did not know the scheme existed until neighbors received payments, and a local organization saying it was approached by Shine in 2024 because the firm “was finding it difficult to sign Yolŋu up,” was contracted for outreach only to the end of August, and continued past that date mostly unpaid.

The 8,750 cap was written into the orders to limit the funder at the top end. In the event it never bound. The constraint that mattered was at the other end, and nothing in the structure was watching it.

The one thing this settlement got right about timing

Both Shine and the funder submitted that they should be paid ahead of registered group members. The funder argued its commission had priority under the funding agreement, including priority over the A$15 million of agreed costs. Mortimer CJ accepted neither submission. She held that both should wait for a portion of their deductions until after the registration date, so that “receipt of their approved deductions should happen side by side with distribution to group members, because by then the number of eligible claimants will be known.”

She then staged it: once the administrator determines 3,000 eligible claimants, the ATE premiums, Shine’s approved actual costs and commission on the first 3,000 claimants — A$10,800,000 — are released. The sequence for the remaining commission was left to be decided after registration closed.

That is worth noting because the current direction of travel elsewhere is the opposite. The UK consultation on opt-out collective actions proposes a presumption that funders are paid when a settlement is approved rather than after distribution, which is a change I have already written that I would benefit from. Mortimer CJ’s reasoning is the clearest argument against it I have seen, and it is not a fairness argument. It is that paying the funder before the count is known means paying it a number nobody has verified yet.

What the other side gets right

The case was brought. That is not a small thing. These are claims about wage control legislation that ran until 1971, against the Commonwealth, on behalf of people who are now in their seventies, eighties and nineties, with records that are ninety years old. Nobody was going to run it on a conditional fee out of goodwill, and Mortimer CJ said plainly that while the funder is a commercial business seeking to make a profit “like any other business,” its involvement “very much reduces the risk” for claimants. She approved the settlement as fair and reasonable against the alternative, which was a trial with an uncertain outcome and years of appeals.

The outreach is also genuinely hard in a way that an hourly rate is not obviously the wrong answer to. The notice program ran in five languages. The communities are remote and numerous. And the local-organization alternative was not fully formed on the record: the three proposals in front of the court covered about a third of the communities, and the A$500,000 figure is an extrapolation the judge made rather than a quote anyone submitted. A firm choosing the option it could actually staff, in a scheme with a fixed registration deadline, is not the same as a firm choosing the expensive option because it was expensive.

And the mechanism did catch it. The court appointed amici, cut the fees, refused the A$8 million, refused the commission on the gross, refused the commission on the amici’s own costs, and refused the priority. Almost everything I have described above is in the judgment because a judge had the jurisdiction to ask and used it.

What I take from it

The generalizable point is not about class actions, which I do not fund. It is that in any funded matter some part of what the funder recovers depends on work the funder pays for and does not perform, and the basis on which that work is billed determines whether the person doing it is trying to produce the same outcome the funder is. Here the two were pointed in different directions and it was legible in the fee structure from the start, before a single community had been visited.

In my own funds I know the commission rate and I know the case names. I do not know what any counsel is billed at, whether any of the work is on a deferred or conditional basis, or which line items scale with hours rather than with outcome. That is not a complaint about disclosure; it is a description of what an LP report contains. But it means I cannot tell, for any matter I am funded into, whether the person doing the expensive part is paid to finish it or paid to work on it.

The reason I can tell in this case is that an Australian court appointed independent counsel and published 146 pages. There is no amicus curiae in a private fund. The only version of that scrutiny available to me is asking, so the question I am adding to the list for the next annual meeting is a narrow one: across the matters in this fund, what proportion of legal spend is billed hourly and what proportion is contingent on the outcome. If the answer is that it varies by matter, the follow-up is which way it varies on the largest ones.

---

Figures are from McDonald v Commonwealth of Australia [2025] FCA 380 (Mortimer CJ, 17 April 2025) and the court-approved settlement notice. Registration and outreach reporting is from the ABC. The administrator’s final eligibility count had not been published when this was written.

Commentary and personal experience — not investment, legal, or tax advice. Investing carries risk, including total loss of capital. Always do your own due diligence.