The Debt Didn’t Move: Burford’s Leverage After YPF
Between December and June, Burford’s borrowings rose 13% and its leverage ratio nearly quadrupled, from 0.9x tangible book to 3.5x. Almost none of that came from borrowing. The denominator collapsed, because a court reversed a judgment and $2.4 billion of carrying value went with it. In the same six months the firm’s realized gains went up.
I don’t own Burford shares and I’m not proposing to. My interest is narrower. I’m a limited partner in three private litigation-finance funds, and one of them has taken $1,000,000 of my money and distributed nothing back yet — everything I know about that position is a mark somebody else computed. So when the one company in this asset class that has to publish its marks, and file them with the SEC, shows what happens to a balance sheet when a single mark reverses, that is the closest thing I get to a look inside my own statements.
What Six Months Did
Burford reported its second quarter on August 6 — the first full set of results since the Second Circuit reversed the $16.1 billion YPF judgment against Argentina in March, and the roughly $2.4 billion write-down of YPF-related assets that followed. The balance sheet either side of that event:
| Dec 31, 2025 | Jun 30, 2026 | |
|---|---|---|
| Shareholders’ equity | $2,448M | $819M |
| Tangible book value per share | $10.57 | $3.12 |
| Debt payable | $2,128M | $2,403M |
| Cash and marketable securities | $621M | $733M |
| Debt to tangible book value | 0.9x | 3.5x |
| Net debt to tangible book value | 0.7x | 2.4x |
Equity fell by $1,629M. The year-to-date net loss attributable to shareholders was $1,630M. Those two numbers are the same number: the entire decline in the firm’s equity is the loss, and effectively all of the loss is an unrealized change in what the YPF assets are estimated to be worth. Nothing was repaid, called, or defaulted on.
The Denominator Did the Work
Leverage ratios put a contractual number on top of an estimated one. The debt is exact — it is written in an indenture, it accrues interest on stated dates, and no accountant’s judgment enters into it. Tangible book value is the residual after every capital provision asset has been marked to a fair value, and for a litigation funder those marks are probability-weighted opinions about what courts will do.
So the ratio can move violently while nothing about the firm’s obligations changes at all, and here is the split. Burford borrowed $275M more over the six months, a 13% increase. Against December’s tangible book, that new borrowing alone would have taken leverage from 0.92x to about 1.04x. The move from 1.04x to 3.52x is entirely the write-down.
Of roughly 2.6 turns of leverage added in six months, about a tenth of a turn came from raising debt. The rest came from a three-judge panel in Manhattan.
The uncomfortable half of that is the reverse case, which nobody discusses while it is happening. When the YPF asset was carried at its peak, the same arithmetic ran the other way: the largest position in the portfolio was inflating the equity that the borrowings were being measured against. The 0.9x that looked prudent in December was prudent on the assumption that the mark was right. A ratio built on a fair value is at its most reassuring exactly when the most aggressive mark in the book is at its highest, which is the point at which it is least informative.
Two Measures of the Same Book, Pointing Opposite Ways
What makes this a useful specimen rather than a disaster story is that Burford published both measures side by side, and they disagree completely.
Year to date, capital provision income was negative $1,584M, against positive $246M in the same period of 2025. Over the same six months, net realized gains were $63M, against $61M — slightly up. In the second quarter alone, realized gains were $35M against $27M a year earlier. Cash receipts were $157M in the quarter, more than three times the $48M of the prior-year quarter and the strongest since the first quarter of 2025, with $733M of cash and securities on hand at the end of June.
Read the realized column and this was an ordinary-to-good half year: cases resolved, money arrived, slightly more of it than last time. Read the fair-value column and it was the worst in the company’s history. Both are honest. They are answering different questions. Realized gains tell you what happened to matters that finished. Fair value tells you what the firm currently estimates about matters that haven’t, and the estimate on one enormous unfinished matter changed.
The lesson I take is not that fair value is fake. It is that the two columns have different error bars, and only one of them is capable of moving $1.6 billion in a quarter without any cash changing hands.
What the Strategy Change Concedes
The interview Burford gave a week after the numbers is more informative than the numbers, because it describes behavior rather than accounting. Chief executive Chris Bogart said the firm had “somewhat reduced our willingness to take on some very large, but only moderately profitable deals,” and described roughly $10 million of annual compensation taken out across a workforce of about 160 people as “a one-off” rather than the start of repeated cuts. The firm is also looking at investing in law firms, a route that alternative business structures and management services organizations have recently opened.
The deployment figures show the same thing without the framing. New definitive commitments were $282M year to date against $520M a year earlier — close to half. Deployments barely moved, $195M against $211M, and realizations were $191M against $225M. So this is not a funder that has stopped writing business; it is one that has stopped signing the biggest tickets while it works through what it already owns.
Notice what that sentence is a concession about. It is not about picking bad cases. YPF was, by the standard measure, a spectacularly good pick — it produced one of the largest judgments in the history of commercial litigation before it was reversed. The retreat is about size: a position large enough to set the firm’s book value is a position whose valuation error sets the firm’s book value too. “Only moderately profitable” is doing a lot of work in that quote, because it concedes that the deals in question were being written for scale rather than for return per dollar of risk.
What a Limited Partner Can Actually See
Here is why I read a listed funder’s filings at all when I hold none of its stock.
Every quarter I receive a net asset value for each of my three funds. That number is produced by the same exercise Burford performs: unresolved claims marked to an estimate of fair value. What I do not receive, and have never received, is any of the machinery that would let me interrogate it. I don’t get the largest position as a percentage of the fund. I don’t get a realized-versus-unrealized split of the reported gain. I don’t get whether the vehicle carries leverage, at what ratio, measured against what. For the 2026 fund the situation is at its starkest: $1,000,000 wired in, nothing distributed out, so the entire reported position is a mark with no cash yet to test it against.
Burford is the one place in the asset class where all of that is public, audited, and filed. It just demonstrated that a fair-value litigation book can lose 70% of tangible book in two quarters while realizing slightly more cash than the year before. I cannot rule that shape out for any fund I’m in, because none of them publish the numbers that would let me look.
I’ve seen the American version of the same event at the case level. When I took apart a $213M verdict that a $60M judgment preservation policy couldn’t rescue, the finding was that a trial court win is not a final number — an appeals court ordering a retrial turned a headline award into nothing recoverable. YPF is that finding at balance-sheet scale. The mechanism is identical and only the zeros are different.
The Honest Handicap
Several things cut against the way I’ve framed this.
Fair value is the right accounting, and the alternative is worse. Carrying litigation assets at cost would have concealed the appreciation and the reversal alike, and would have made the firm less legible, not more. The marks moved because the facts moved: an appellate panel ruled. That is the system working as designed, and a reader who concludes “fair value is fiction” has taken the wrong lesson. My complaint is about what a ratio computed on top of a fair value can be relied on to mean, not about the fair value itself.
Nothing here was hidden. YPF’s size was the single most discussed position in litigation finance for years, disclosed relentlessly by the company and modelled to death by everyone who followed it. Anyone could see that one matter dominated the book. This is not a disclosure failure, and calling it one would be lazy — it is a demonstration that complete disclosure of a fair value still leaves you holding an opinion about an appeal.
The cash position is genuinely fine, and I am not predicting distress. $733M of liquidity against $2,403M of debt, with receipts accelerating, is not a funding crisis. More importantly, I don’t know Burford’s covenant terms. Debt to tangible book is a metric the company chose to report; whether any lender measures anything against it is a different question, and asserting that a covenant is threatened would be inventing a fact. The 3.5x is alarming as a description and may be irrelevant as a constraint.
The reversal may partly reverse. The write-down is unrealized. Burford continues to point to bilateral investment treaty arbitration as a recovery route and has a Supreme Court path in a separate matter. A mark that fell 70% can be written back up, and if it is, the leverage ratio will improve for exactly the same non-reason it deteriorated.
The “best quarter since 1Q25” figure is partly flattery. $157M against $48M is more than triple, but the comparison quarter was unusually weak. Year-to-date receipts were $247M against $306M — down. The cash story is good; it is not as good as the single-quarter framing suggests, and I’d be misusing it if I let it stand as the counterweight without that.
And the obvious conclusion is the wrong one. The tempting read is that listed funders are risky and private funds are safer. The opposite is closer to true. Burford’s mark was tested in public by short sellers, analysts, an appellate court and the SEC’s reporting regime. Mine are tested by nobody I can name.
What I’m Asking For Now
The useful output of this is a list of questions, because the position I hold is illiquid and the only live decision is whether to commit more.
- Largest position as a percentage of NAV. One number, and the single most informative thing a funder can disclose about the shape of its risk. If the answer is a range or a policy statement rather than a figure, that is itself the answer.
- Realized versus unrealized, split out. A reported gain that is entirely unrealized is a forecast with a decimal point. Burford’s two columns disagreed by nearly $1.7 billion; the split is the only way to see that coming.
- What moved a mark, not just that it moved. Milestone-based markups mean a trial win can be booked years before a dollar arrives, and an appeal can take it back. I want to know which marks rest on decisions that are still appealable.
- Fund-level leverage, and what it is measured against. If there is a facility, its covenants are computed on somebody’s valuation of illiquid claims. That is the mechanism that turns a bad mark into a forced action, and it is invisible in a NAV statement.
- Concentration limits as written, not as practiced. Whether the documents cap a single position, and at what level. Burford’s constraint turned out to be discretionary, and the discretion was revised after the fact.
Where I Land
I’m not buying the shares, and that isn’t a judgment about the underwriting. It is that I already own concentrated, illiquid, fair-valued litigation risk through three funds, and buying a levered, listed version of the same exposure would be doubling a bet I can’t see the inside of.
What I’ve actually changed is smaller and more useful. I no longer read a NAV line as a number. Burford’s June figure was a real, audited, professionally computed estimate, and it was 70% lower than December’s because one court disagreed with another. My funds’ figures are computed the same way, by people with the same incentives, about claims I can’t inspect, and they arrive without the columns that would let me tell a resolved gain from an anticipated one.
The debt was the honest number on that balance sheet all along. It said $2,128M in December and $2,403M in June, and it meant exactly that on both dates. Everything else on the page was a view about the future, and the ratio joining them was only ever as solid as the view.
Sources
- Burford Capital: Burford Reports 2Q26 and YTD26 Financial Results (Aug 6, 2026) — shareholders’ equity $819M against $2,448M at Dec 31, 2025; book value per share $3.73 and tangible book value per share $3.12, against $11.18 and $10.57; debt payable $2,403M against $2,128M; cash and marketable securities $733M against $621M; debt to tangible book value 3.5x against 0.9x and net debt to tangible book 2.4x against 0.7x; capital provision loss of $1,584M year to date against income of $246M; net realized gains $63M against $61M year to date and $35M against $27M in the quarter; net loss attributable to shareholders of $1,630M year to date and net income of $2M in the quarter; cash receipts $157M in the quarter and $247M year to date, against $48M and $306M; new definitive commitments $282M against $520M, deployments $195M against $211M, realizations $191M against $225M. Quarterly detail also filed on Form 10-Q, Aug 6, 2026 (total revenue $110.7M against $191.3M; diluted earnings per share $0.01 against $0.39)
- Legal Funding Journal — “Burford Reins In Large-Deal Appetite as It Eyes Law Firm Investment” (Aug 10, 2026), reporting Non-Billable: roughly $10M of annual compensation cost removed across a workforce of about 160; Chris Bogart on having “somewhat reduced our willingness to take on some very large, but only moderately profitable deals” and on the compensation restructuring as “a one-off”; the firm examining investment in law firms as alternative business structures and management services organizations open routes to outside capital; the $2.4 billion write-down in March tied to the appellate reversal of the Argentina judgment
- Author’s own records, reconciled against the Portfolio page — three private litigation-finance fund positions, including the 2026 vintage at $1,000,000 contributed and nothing distributed
Commentary and personal experience — not investment, legal, or tax advice. Investing carries risk, including total loss of capital. Always do your own due diligence.
