Ligand Pharmaceuticals Incorporated (NASDAQ: LGND) — A Levered Royalty Lender Wearing a Compounder’s Multiple

Report date: 2026-08-15 · Price reference: $294.595 (close, 2026-08-14) · Initiating coverage Sector: Health Care · Biotechnology / Pharmaceutical Royalties

The analytical body of this article (sections 1–15) contains no recommendation and no price target. The single exception is the clearly-labelled Claude's Take block immediately below.

⚡ Claude’s Take

This block is the author’s own independent opinion and is offered as general information only — it is not investment advice, and it is the only part of this article that takes a position. Everything below it (sections 1–15) carries no position, no recommendation and no price target.

Verdict: HOLD — great execution, wrong price. Accumulate on weakness in the ~$210–$250 zone; not a short.

Tag: “The best specialty-finance business in biotech, priced as though it were a biotech.”

Ligand has done something genuinely rare: it has transformed itself. Since 2022 management has stripped out an operating pharma company, spun off two businesses, cut headcount to forty-seven people, and rebuilt around royalty aggregation. Royalty revenue went from $73M to a guided $225–250M. Adjusted EPS went from $2.44 to a guided $9.00–9.50. The XOMA acquisition — $668M of cash for 120+ assets, $110M of usable tax attributes, ~$1.50 of 2027 accretion and a $2.3B milestone pool that management deliberately underwrote at zero — looks like a well-priced deal by a team that knows what it is doing. This is not a promotional story. The execution is real.

What I cannot get comfortable with is the price against the quality of the earnings being capitalised. Three things sit under the headline. First, the FY2025 “60% revenue growth” is ~29% once you remove the $53.1M one-time divestiture gain that was booked inside revenue. Second, GAAP operating cash flow in FY2025 was $49.4M while the metric management is paid on — “core cash flow” — was $215M; when the CFO guides to “>$200M of operating cash flow,” he means the second number, and most of the sell-side note-taking will not make that distinction. Third, adjusted EPS excludes $2.22/share of stock comp and $1.49/share of royalty-asset amortisation — and for a royalty aggregator, amortising an acquired royalty is depletion, not an accounting artifact. Strip both and the 31.8x becomes 41.9x, then 53.2x. Meanwhile the stock sits at the 98.9th percentile of its own ten-year price-to-sales range and the 92nd on price-to-book, and the insiders who bought conviction lots at $59, $97 and $105 have not bought a single share in fifteen months while selling ~$34M into $190–$275.

The framing is an idiosyncratic re-rating on a one-way street — not a momentum bubble, not a falling knife. The factor evidence is unusually clear: a 92% one-year gain with a 14% maximum drawdown and a 2.50 Sharpe, but a factor R² below 30% and 31.9% idiosyncratic volatility. This stock is not being carried by the biotech complex and will not be rescued by it. It trades on Filspari, on XOMA, and on continued access to sub-3.5% convertible money. That is a concentrated bet dressed in the language of diversification: at a 9% royalty rate, Travere’s reiterated >$3B Filspari peak is worth ~$270M a year to Ligand — more than the entire currently guided royalty base, from one drug.

I would own this business. I would not underwrite it at 42x SBC-burdened current-year earnings with the margin of safety fully consumed. A 25–28x multiple on ~$8.95 of SBC-burdened 2027 earnings implies roughly $225–$250; I would start there and want the $210s for size.

Conviction: medium. Flips bullish if FY2027 GAAP operating cash flow converges toward the guided figure — i.e. prints above ~$150M — proving the adjusted earnings are actually cash. Flips bearish if a pivotal readout fails while the convertible window closes, forcing Ligand to fund deals with equity or expensive debt at exactly the wrong moment.

📈 Stock Price Action — Five-Year Event Map

Ligand has round-tripped from $101 (November 2021) to a $46.49 low (June 2022) and then compounded relentlessly to an all-time-window high of $322.61 on 2026-07-09. It closed at $294.595 on 2026-08-14, 8.7% below that high, inside a 52-week range of $151.08–$322.61. Trailing returns: +34.5% over three months, +60.3% over six, +91.6% over one year, +191.6% over two, +342.1% over five. Price sits above its 21-day ($295.07), 50-day ($284.16) and 200-day ($231.18) exponential moving averages, stacked bullishly.

Cycle narrative. (1) The 2021–22 decline was sector-wide and company-specific at once: XBI halved while Ligand’s own COVID-era Captisol revenue (Gilead’s Veklury) rolled off, taking FY2022 revenue down to $196M from $242M. (2) The November 2022 move requires a warning. The largest apparent single-day gain in the split-adjusted series, +28.8% on 2022-11-02, is a data artifact — the price provider applies a 1.603 adjustment factor on that date for the OmniAb spin-off; the unadjusted close actually went $84.61 → $68.00, a −19.6% separation. It is not a price event and is excluded from the analysis. (3) Through 2023 the market correctly saw a shrinking royalty base with Kyprolis maturing and nothing yet replacing it. (4) From late 2023 the pivot began to show in the numbers as Filspari, Capvaxive and Ohtuvayre royalties initiated. (5) 2025 was the year the model was validated — Filspari royalty tripled to $32.0M, the Pelthos spin-out closed, and August’s $460M convertible gave the company real deployment capacity. (6) April 2026 brought Filspari’s second-indication approval in FSGS and the XOMA merger agreement. (7) June 2026 is the most interesting month: +36.3% with no single outsized day — a relentless grind from $229.78 to $316.09 through the pricing of a $700M convertible, which normally pressures a stock as hedge funds short the delta. Management explicitly bought back $60M of stock “to offset the market impact from the convertible hedge activity.” (8) The stock has drifted back since the July 9 high, with Mirum’s FDA setback on PSC disclosed on the August 6 call.

Price moves are Fact; attributed drivers are Interpretation. This section contains no recommendation and no price target.

1. Executive Summary

Ligand Pharmaceuticals is no longer a pharmaceutical company. It is a biopharmaceutical royalty aggregator — a specialty finance business that deploys capital into royalty interests on mid-to-late-stage and commercial drugs, run by forty-seven people against $290.5M of trailing revenue. Management pivoted decisively in 2022, spinning out OmniAb (2022) and Pelthos/LNHC (2025), cutting cash operating expenses to roughly half their 2022 level, and rebuilding around capital deployment. The results are not in dispute: royalty revenue has gone from $73M in 2022 to a guided $225–250M in 2026, and adjusted diluted EPS from $2.44 to a guided $9.00–9.50.

Two transactions define 2026. In June the company issued $700M of 0.00% convertible senior notes due 2031, and in July it closed the acquisition of XOMA Royalty for $39.00 per share in cash plus a contingent value right — roughly $668M, the largest deal in company history — adding 7 commercial programs, ~14 late-stage clinical assets, 100+ earlier programs, $2.3B of disclosed potential milestones and more than $110M of usable Section 174 credits and NOLs. Management guides the deal to ~$0.50 of adjusted EPS accretion in H2 2026 and ~$1.50 in 2027, with XOMA’s ~$30M standalone cost base falling below $5M. On the disclosed facts this is a competently underwritten transaction and the single best evidence of management’s capital-allocation ability.

The competitive position is more contingent than the narrative suggests. Under Greenwald’s framework Ligand has no barrier to entry: capital is a commodity, customers have alternatives (management says so explicitly), and there is no scale-plus-captivity dynamic. What Ligand has is an extremely lean cost structure that converts a modest gross spread into a large net one, accumulated deal-flow relationships, and — for now — cheap capital. Only the cost structure is structural, and it is replicable. Management’s own description of the niche — “very high demand and low supply of capital,” royalty finance under 9% of biopharma funding — is, in Marathon’s capital-cycle language, precisely the configuration that attracts supply.

The financial quality requires care. Three adjustments matter. FY2025’s headline 60% revenue growth becomes ~29% once the $53.1M one-time LNHC/Pelthos divestiture income — booked inside revenue — is removed. GAAP net income is dominated by marks: $111.8M of FY2025’s $159.0M of pretax income was non-operating, and in Q2 2026, $47.5M of $48.5M of net income came from investment fair-value gains against just $8.6M of operating income. And GAAP operating cash flow was $49.4M in FY2025 against the $215M “core cash flow” figure disclosed in the proxy as the executive bonus metric — a $166M gap, and almost certainly the basis for management’s “>$200M in 2026, ~$300M in 2027” cash guidance.

Growth is real but concentrated. Filspari (Travere), at a 9.0% effective royalty rate, went from $12.2M to $32.0M of royalty in one year, with US net sales up 96% year-over-year in Q2 2026 and the partner reiterating peak sales above $3B. At that rate a $3B Filspari alone is worth ~$270M annually to Ligand — more than the entire guided 2026 royalty base. Against it, Kyprolis, still the largest single royalty at $35.5M, is declining (−17% year-over-year in Q2). Seven pivotal readouts land within eighteen months and management concedes not all will succeed; one setback (Mirum in PSC) and one Phase III discontinuation (Agenus BOT/BAL) have already been disclosed.

At $294.595 the shares trade at 31.8x guided 2026 adjusted EPS, 41.9x once stock compensation is charged, and 53.2x once royalty-asset amortisation is also charged. Enterprise value is ~26.6x guided 2026 royalty revenue and ~128x FY2025 GAAP operating cash flow. On the cash-flow-to-the-firm measure Ligand trades at roughly 2.5x Royalty Pharma’s multiple. The AZI own-history percentiles put the stock at the 98.9th percentile on price-to-sales and the 92nd on price-to-book. The market is simultaneously underwriting a >$3B Filspari, full XOMA delivery plus milestone conversion, undiminished deployment spreads, continued sub-3.5% convertible access, and the proposition that $3.71/share of stock comp and royalty depletion are not real costs.

2. Business Overview

What the company does. Ligand describes itself in the Q2 2026 10-Q as “a biopharmaceutical royalty company focused on deploying capital and licensing technologies to acquire and create diversified royalty streams from high-value medicines,” whose “primary business is investing in and structuring royalty interests in mid- to late-stage development and commercial biopharmaceutical products, allowing us to generate long-duration, non-dilutive cash flows supported by a lean corporate cost structure.” That is an accurate self-description and it should be taken literally. This is a finance business.

Revenue segmentation. Ligand reports three revenue lines, and the distinction between the first two components of royalties is important because it drives both the accounting and the cash-flow presentation:

The financial royalty asset category matters more each quarter and will matter far more post-XOMA. As the CFO explained on the Q2 call: “cash received upon achievement of a milestone may not be recognized entirely as revenue in the period received. A portion may instead reduce the carrying value of the underlying asset.” In other words, part of the economic return on a growing share of the portfolio never appears in revenue at all — it appears as a balance-sheet movement and, in the cash flow statement, within investing activities (“Proceeds from financial royalty assets” — $9.5M in H1 2026). Any analyst modelling this company from the revenue line alone will systematically understate cash economics; any analyst taking management’s cash-flow guidance at GAAP face value will systematically overstate it. Both errors are live in this name.

Business model — how it makes money. Ligand acquires or creates royalty interests three ways. It buys existing royalty streams from originators (universities, biotechs, other royalty holders). It provides development-stage or project financing to partners in exchange for a royalty on future sales — the Agenus BOT/BAL agreement (2.625% of global net sales) and the Orchestra Biomed transaction are examples. And it licenses proprietary formulation technologies — Captisol, a modified cyclodextrin that improves solubility and stability, and NITRICIL — in exchange for royalties, milestones and material sales. Captisol is the legacy engine: it is embedded in Gilead’s Veklury, Merck’s Noxafil-IV, Baxter’s Nexterone and Sage’s Zulresso, and it still generated $40.2M of material sales in FY2025.

Deal sizing is deliberately small. Management targets sub-$100M transactions, typically $25M–$75M per asset, and states that perpetuity-level growth requires only $100M–$125M of deployment per year, against a team capacity of ~$200M± annually. This is the strategic core: Ligand is fishing where Royalty Pharma’s cost structure will not let it fish.

Customer types and end markets. Ligand’s “customers” are two distinct populations. On the origination side, they are capital-constrained biotech issuers — small-cap public companies, late-stage privates, occasional mid-caps — seeking non-dilutive financing. On the collection side, they are large-cap marketers who owe the royalties: the FY2025 10-K names Amgen, Merck, Jazz, Recordati and Sanofi, to which XOMA adds Roche, Servier, Ipsen and Zevra. The credit quality of the paying counterparties is excellent; the credit and execution quality of the origination counterparties is emphatically not, and that asymmetry is the business.

Recurring versus non-recurring. Royalties ($161.0M in FY2025, 60% of revenue) are genuinely recurring and contractual, though each individual stream has a finite patent life. Captisol ($40.2M, 15%) is recurring in character but lumpy in timing — management guided H2 2026 Captisol to ~40% Q3 / ~60% Q4 purely on order timing. Contract revenue and income ($66.9M, 25%) is the problem line: it is where milestones land, and it is also where $53.1M of one-time gain on the LNHC divestiture landed in FY2025. Guidance for 2026 contract revenue is $10–20M, against $66.9M in 2025 — management’s own numbers confirm the 2025 figure was not repeatable.

Scale and cost structure. Forty-seven employees. FY2025 revenue per employee of $5.70M. Headcount actually fell from 68 to 47 across Q3 2025 to Q4 2025 as the Pelthos spin-out completed. The lean structure is not incidental — it is the whole economic engine, and it is what allows a portfolio of sub-3% royalty rates on other companies’ drugs to produce meaningful per-share earnings.

Verdict: a genuinely capital-light, genuinely recurring-revenue financial business with an unusually clean cost structure, whose reported revenue line materially understates portfolio economics in one direction (financial royalty assets) and materially overstated growth in FY2025 in the other (the Pelthos gain). Understand both or misprice the company.

3. Industry Dynamics

Industry structure. Pharmaceutical royalty financing is a young, concentrated and under-penetrated capital market. Management’s own estimate, offered on the Q2 2026 call and flagged by the speaker as rough: royalty capital represents less than 9% of biopharmaceutical company financing overall, and “well less than 5%” on the development side. The competitive set is short — Royalty Pharma, Blackstone Life Sciences, HealthCare Royalty Partners, Oberland Capital, DRI Healthcare, Sagard Healthcare, plus an expanding fringe of private credit funds — and, in Davis’s words, “they all have their different approaches… we haven’t really been head-to-head on any royalty financings.”

Market size and profit pools. The demand driver is structurally sound and not a fad. Biotech equity issuance has been difficult for several years; a small-cap with a Phase III asset and a depressed share price faces a genuinely unattractive choice between dilutive equity and expensive venture debt. A royalty sale monetises a future cash flow without touching the cap table. That demand is real, is growing with the number of approved and late-stage assets, and is why Ligand’s origination pipeline is “exceptionally active.”

Competitive intensity — and the capital-cycle warning. Here the analysis has to depart from management’s framing. Davis characterises the market as one of “very, very high demand and low supply of capital.” Marathon’s capital-cycle discipline says: that is not a moat, that is a signal. High returns earned in a market with no barrier to entry attract capital until returns normalise. There is no licensing regime, no scale threshold, no network effect and no proprietary distribution protecting a royalty buyer’s spread. The specific mechanism of erosion is easy to name: as the sub-$100M niche is shown to be profitable, credit funds with lower return hurdles and larger balance sheets will price more aggressively for the same assets. Ligand’s protection is that it is early, disciplined and low-cost — not that entry is hard.

The counter-argument deserves weight. Origination in this market is genuinely relationship- and diligence-intensive; the ability to underwrite a Phase III readout is not a commodity skill; and a $30M cheque to a small-cap is uneconomic for a fund that needs to deploy billions. Those frictions are real and could sustain the spread for years. They are frictions, not barriers.

Value-chain position. Ligand sits between the innovator (who needs capital) and the marketer (who pays the royalty). It bears development risk on pre-approval deals and commercial risk on approved ones, but bears no manufacturing, no salesforce, no regulatory and no litigation cost. That is a structurally attractive position in the value chain — the returns accrue with almost none of the operating leverage risk. It is also a position with essentially no pricing power: royalty rates are set at the negotiating table against alternatives, which is why Ligand’s effective rates on the biggest products are so thin.

Regulation. Ligand is not itself regulated as a drug developer for its royalty assets, which is a genuine advantage — no FDA inspection risk, no manufacturing compliance, no DOJ marketing exposure. It is, however, entirely exposed to FDA decisions on its partners’ assets, as the Mirum PSC setback and the Agenus Phase III discontinuation both demonstrate within a single quarter. Downstream, drug pricing policy (IRA negotiation, most-favoured-nation proposals) compresses the very net sales on which the royalties are calculated — a second-order but real and under-discussed exposure for every royalty aggregator.

Comparison to the scale player. Royalty Pharma carries an enterprise value of $59.3B on TTM revenue of $2,536M, with TTM operating income of $1,581M — a 62% operating margin — and TTM cash flow to the firm of $3,338M. Ligand’s EV is ~$6.3B on TTM revenue of $290.5M. RPRX is roughly nine times larger on both measures and enjoys a materially lower cost of capital. The strategic logic of Ligand’s niche is exactly that it is beneath RPRX’s economic floor. That logic holds today. It holds less well every year that Ligand’s own cheque sizes grow.

Verdict: a structurally attractive industry to be in right now, but not a structurally defended one. The demand side is genuine and durable; the supply side has no barrier and management’s own market description is the textbook precondition for competitive entry. Investors should underwrite spread compression, not spread persistence, and should treat the current return on deployed capital as a cyclical high rather than a permanent feature.

4. Competitive Position

Naming the moat — or its absence. The Greenwald taxonomy admits three genuine competitive advantages: supply/cost advantage, demand-side customer captivity, and economies of scale combined with captivity. Ligand has none of them in the strict sense.

Supply/cost advantage. Ligand’s input is capital. Its current cost of capital is genuinely low — the 2031 notes carry a 0.00% coupon and an all-in economic cost of roughly 3.1% per annum (see the Capital Allocation section below). But this is a market condition, not a proprietary asset. It reflects an accommodating convertible market and the equity’s own high implied volatility, which makes the conversion option valuable to buyers. Royalty Pharma funds itself more cheaply still. If the convertible window narrows, Ligand’s advantage evaporates without any change in its own conduct.

Customer captivity. There is none, and management concedes it plainly: partners “usually almost always have alternatives. Those alternatives are — there are other options in the equity markets, access to debt, et cetera. So that keeps everybody honest, and it keeps the market pretty level.” That is an honest and creditable answer, and it is dispositive. There are no switching costs — a royalty sale is a one-time transaction, not a recurring relationship.

Economies of scale plus captivity. Ligand has genuine operating leverage but no captivity to pair with it, which under Greenwald means the scale does not convert into a barrier.

What Ligand actually has. Three things, in descending order of durability:

1. A structurally lean cost base. Forty-seven employees against $290.5M of revenue is the real edge. It means Ligand can accept a lower gross yield on a royalty than a competitor with a larger cost structure and still earn a superior net return. The Q2 call made this concrete in the XOMA context: XOMA’s ~$30M standalone opex becomes <$5M inside Ligand. That is a $25M annual synergy on a $668M purchase price — a 3.7% yield uplift from cost structure alone. This is real, is measurable, and is the most defensible thing about the business. It is also fully replicable by any competitor willing to run lean.

2. Origination relationships and portfolio-management infrastructure. Davis emphasised that “a strategic part of our organizational changes over the last couple of years has been to put in place a sophisticated portfolio management system, which allowed us to absorb the XOMA assets with a de minimis amount of absorption of their existing infrastructure.” Absorbing 120+ assets without absorbing the staff is a genuine capability. It is an accumulating advantage, not a barrier.

3. A reputation as a repeat, non-predatory counterparty. Ligand consistently follows on in partner equity financings — Palvella (“we have followed on in every equity financing they’ve done”), Agenus, Athira — which management frames as signalling insider confidence and strengthening the partner. This builds real deal flow. It also, note, means Ligand takes equity risk in the very companies whose royalties it owns, doubling the exposure.

The pricing-power test. The decisive evidence against a moat is in the royalty rates themselves. If Ligand had negotiating power, it would show up in the rate:

Ligand captured 2.75% of $4.8B of partner product sales. The newly acquired Vabysmo — Roche’s third best-selling product, on track for roughly $7B of peak sales — carries a peak royalty to Ligand of “approximately $35 million,” an effective rate near 0.5%. The high rates (Teriparatide 23.8%, Evomela 20.0%) sit on tiny, declining products. The pattern is unambiguous: where the product is large, Ligand’s rate is small. That is the signature of a price-taker.

Direct comparison to competitors. Against Royalty Pharma, Ligand is smaller, higher-growth, higher-cost-of-capital, and operates below RPRX’s deal-size floor. Against private credit and Blackstone Life Sciences, Ligand is permanent capital with no fund life and no redemption pressure — a genuine structural advantage in a business where the asset duration runs to 2040. Against a hypothetical new entrant, Ligand’s protection is its lean cost base, its accumulated relationships, and the fact that underwriting Phase III readouts well is hard. None of these is a barrier to entry.

Verdict: a narrow, contingent and genuinely valuable competitive position — but not a durable moat. The advantage is real today and is expressed correctly in the financials (the operating leverage is visible). It rests on a cost structure that is replicable, on capital-market conditions that are cyclical, and on underwriting skill that is personal to a small team. If the “moat” claim cannot be tied to a financial outcome that would deteriorate without it, it is not a moat — and here the honest answer is that Ligand’s returns would deteriorate meaningfully if the convertible market closed or if two well-capitalised competitors moved down-market. That is a competitive position, and a good one. It should not be capitalised as a fortress.

5. Growth History and Forward Opportunities

The historical record. Revenue tells a story of transformation, not of smooth compounding:

The 2021–2023 decline reflects the COVID-era Captisol unwind and the OmniAb separation; the 2024–2026 acceleration reflects the royalty portfolio inflecting. Management’s preferred framing — royalty revenue from $73M in 2022 to $225–250M in 2026, adjusted EPS from $2.44 to $9.00–9.50 — is fair and is the right lens for judging the pivot. On that basis the compound growth is genuine and impressive.

The one required correction. FY2025’s headline +60% revenue growth includes $53.1M of income recognised on the LNHC divestiture in connection with the Pelthos Transaction, booked within contract revenue and income. Ex that item, FY2025 revenue was ~$215.0M, up ~29%. Twenty-nine percent is still very good. Sixty percent is not the number, and 2026 contract-revenue guidance of $10–20M against 2025’s $66.9M confirms management’s own view that it was not repeatable.

Organic versus acquired. Almost all of the growth is acquired in the sense that royalty streams were purchased or structured rather than internally generated — that is the business model, not a criticism. But it means growth is a direct function of capital deployed, and capital deployed is a function of capital access. In 2025 Ligand invested $131M against a stated plan of $150–200M, and added 10 new programs against a target of ~5. Deployment, not demand, is the constraint.

The forward drivers.

Filspari (Travere) is the thesis. US net sales reached $141M in Q2 2026, up 96% year-over-year. The drug secured a second FDA approval in FSGS in April 2026 — the first and only approved treatment in that indication. On 2026-08-04 Travere reiterated confidence in peak sales exceeding $3 billion across IgAN and FSGS. Ligand’s disclosed effective rate is 9.0%. At a $3B peak that is ~$270M of annual royalty to Ligand from one product — greater than the entire 2026 guided royalty base of $225–250M. No other asset in the portfolio is remotely comparable in importance, and management’s repeated emphasis on diversification should be read against that arithmetic.

Ohtuvayre (Merck) is the second engine. Q2 2026 partner sales of $204M, up 98%, at a 3% total royalty rate (2% intangible, 1% financial).

Kyprolis (Amgen) is the melting ice cube. Still the largest single royalty at $35.5M in FY2025, but partner sales fell from $1,627M to $1,529M in FY2025 and −17% year-over-year in Q2 2026 on lower volume. Management says the decline was “anticipated to some degree and… remains within our royalty revenue guidance.” Roughly a quarter of the intangible royalty base is in structural decline.

XOMA adds breadth and duration. Seven commercial programs (notably Vabysmo, Roche, ~$2.6B of H1 2026 sales; Ojemda, Servier/Ipsen, mid-single-digit royalty, with Day One acquired by Servier for $2.5B; Miplyffa, Zevra), ~14 late-stage clinical assets, 100+ earlier programs, $2.3B of disclosed potential milestones, and IP rights that “in many cases extend beyond the next decade” — in some cases to 2040. Duration extension is genuinely valuable for a royalty portfolio and is under-appreciated relative to the headline accretion.

The catalyst calendar. Management describes “the most catalyst-rich period in our company’s nearly 40-year history” — as many as seven pivotal readouts in eighteen months: Orchestra Biomed’s AVIM (BACKBEAT), Ojemda Phase III in frontline pLGG (data mid-2027), LeonaBio’s lasofoxifene, Takeda’s mezagitamab, osavampator, volixibat in PBC, plus a Vabysmo BLA filing in myopic CNV expected 2027 after the Phase III met its primary endpoint in July 2026.

The counter-evidence, disclosed in the same quarter. Two setbacks landed inside one call. Mirum: at the pre-NDA meeting the FDA recommended a Phase III in PSC, pushing submission from H2 2026 to H1 2027 — though breakthrough therapy designation was granted afterwards, which management fairly called “mixed signals.” Agenus BOT/BAL: the current Phase III in relapsed/refractory colorectal cancer was discontinued, with a new neoadjuvant trial not dosing its first patient until Q1 2027 — a multi-year deferral of a 2.625% royalty. Management’s own guidance is appropriately honest: “we do not expect that every one of these pivotal studies will be positive.”

Verdict: high-quality growth in composition, but far more concentrated than presented. The revenue is contractual, the counterparties are investment-grade marketers, the duration is extending, and the operating leverage means incremental royalty dollars fall through at very high margins. Those are the hallmarks of quality. But the growth rests disproportionately on one asset — Filspari — with a second (XOMA’s contribution) still unproven under Ligand’s ownership and a third (Kyprolis) actively declining. The seven pivotal readouts are optionality, not a plan; management says so. This is high-quality growth with a concentrated spine, and it should be underwritten as such.

6. Financial Quality

This is the section that determines whether the valuation discussion below is defensible. Three questions matter: are the reported earnings real, does cash follow them, and do economics improve with scale?

Margin structure and operating leverage — genuinely excellent. Gross margin has expanded from 73.1% (FY2022) to 91.9% (FY2023), 93.4% (FY2024) and 94.6% (FY2025) as low-margin Captisol material sales shrank as a share of the mix. With only forty-seven employees, incremental royalty revenue drops through at very high rates: FY2025 incremental operating margin was 23.9% on a GAAP basis but far higher on a cash basis once non-cash charges are isolated. The operating leverage claim is true and is visible in the financials — which is the only test that matters for a claimed advantage.

But GAAP operating income is small, and GAAP net income is not earned from royalties. This is the central quality-of-earnings finding:

*sign conventions differ by source; the pretax reconciliation is the reliable read.

In FY2025, $111.8M of $159.0M of pretax income — 70% — was non-operating, overwhelmingly fair-value movements on the investment portfolio. Q2 2026 is starker still: operating income of $8.6M produced net income of $48.5M because of an $11.8M gain from short-term investments and a $35.7M gain from changes in fair value of equity-method and other investments. Approximately $47.5M of $48.5M of quarterly GAAP net income came from marks, not from royalties. These marks are two-way: H1 2026 as a whole recorded a $13.5M loss on the same line, and H1 2025 a $11.4M loss on short-term investments. GAAP EPS for this company is close to unanalysable, and the AZI P/E percentile (a benign-looking 55.8th) should be disregarded for exactly this reason — read the P/B (92nd) and P/S (98.9th) instead.

Cash conversion — the gap that matters most.

FY2025 GAAP operating cash flow was $49.4M — barely above FY2023’s $49.6M, and below FY2024’s $97.0M despite revenue rising 60%. Free cash flow was $38.0M. Cash conversion of net income was 0.40x.

Set against that, the 2026 proxy discloses that FY2025 “core cash flow” was $215 million, against a bonus target of $120M — a 179% achievement that drove a meaningful part of executive compensation. The proxy defines the metric in footnote (1): “Core cash flow for annual bonus plan purposes is a non-GAAP measure and is generally calculated as revenue less 1) cost of sales, and 2) cash operating expenses.”

The $166M gap between $49.4M of GAAP operating cash flow and $215M of “core cash flow” is bridged by adding back stock-based compensation ($46.8M), by excluding royalty-asset amortisation, and by treating R&D-financing outlays that GAAP requires to be expensed immediately as investments rather than costs. Each add-back has a defensible argument. Together they produce a number 4.4x the audited one.

This directly conditions how to read management’s cash guidance. On the Q2 2026 call the CFO stated the company is “generating strong operating cash flow in excess of $200 million in 2026, growing to an expected $300 million in 2027.” H1 2026 GAAP operating cash flow was $72.7M. Reaching $200M of GAAP operating cash flow would require ~$127M in H2 — a 75% sequential step-up. It is far more likely that the guidance refers to the core-cash-flow construct. An investor who capitalises “$300M of 2027 operating cash flow” as a GAAP figure against a ~$5.9B market capitalisation will reach a materially different — and wrong — conclusion about the multiple. This is the single most important disclosure trap in the name.

Stock-based compensation — large, and excluded from the headline metric.

At 17.5% of revenue and roughly $2.22 per diluted share on an annualised H1 2026 basis, SBC is not a rounding item — it is approximately 24% of guided 2026 adjusted EPS. Adjusted EPS excludes it entirely. Shareholders approved a further 600,000-share increase to the equity plan in June 2026, on top of 2,403,395 shares already subject to outstanding awards at 2026-03-31 — 12.1% of shares outstanding.

Royalty-asset amortisation — a real cost, also excluded. Adjusted EPS also excludes “amortization of intangible assets” and “amortization or impairment of financial assets.” For most companies, adding back intangible amortisation is standard and reasonable. For a royalty aggregator it is not: an acquired royalty is a depleting asset. Every dollar of royalty collected on a purchased stream is partly a return of capital, not a return on it, and amortisation is the accounting recognition of that depletion — economically analogous to a mining company’s depletion charge. Adding it back overstates sustainable earnings unless the base is continuously replaced, which requires the very capital deployment that is itself uncertain. FY2025 amortisation of intangibles was ~$32.7M; H1 2026 $16.2M (~$32.4M annualised, ~$1.49 per share).

The counter-argument is fair and should be stated: Ligand buys royalties at a discount to expected undiscounted cash flows, so cash yield exceeds book amortisation, and mechanically deducting book amortisation understates economics. That is true. It does not make the charge zero, which is what the adjusted presentation implies.

Returns on capital. FY2025 ROE 13.5%, ROA 9.9% — respectable. ROIC of 2.77% — poor. The divergence is arithmetic: GAAP operating income is suppressed by royalty-asset amortisation while invested capital carries the assets’ full purchase price. This does not prove the business earns bad returns. It does prove that the returns management claims are not observable anywhere in the audited operating statements — they exist only in the adjusted presentation and in management’s core-cash-flow metric. For a business whose entire investment case is “we deploy capital at attractive returns,” the inability to see those returns in GAAP is a significant analytical handicap and a legitimate reason for scepticism.

Balance sheet (2026-06-30, pre-XOMA-close). Cash $1,006.7M plus short-term investments $364.0M = $1,370.7M; long-term investments $351.1M; goodwill $101.5M; other intangibles $209.2M; total assets $2,185.9M. Long-term borrowings $1,130.4M; total liabilities $1,218.4M; total equity $967.5M. Shares outstanding 19.920M. Net debt $119.9M at quarter-end — but the picture changes materially after the XOMA close consumed ~$668M of cash in July. Management states ~$700M of deployable capital remains. Liquidity is strong; the current ratio of 31x is meaningless for a business of this shape.

Verdict: economics do improve with scale — the operating leverage is genuine and measurable — but the reported earnings are of materially lower quality than the adjusted presentation implies. Revenue growth was overstated by a one-time gain; GAAP net income is dominated by two-way investment marks; GAAP cash conversion is weak and running far below the metric management is paid on and guides to; and adjusted EPS excludes ~$3.71 per share of stock compensation and royalty depletion, both of which are real economic costs. The business is good. The earnings are softer than they look.

7. Capital Allocation

Capital allocation is the bridge between business value and shareholder value, and for a royalty aggregator it is not merely one input — it is the entire operating activity. Judged on deal-making, management’s record is good. Judged on financing communication and on incentive design, it is weaker.

The XOMA acquisition — the strongest evidence for management. Announced 2026-04-27, closed 2026-07-14. Consideration: $39.00 per share in cash plus one contingent value right per share. Against 12,129,405 common shares plus 5,003,000 shares issuable on conversion of Series X Preferred, that is roughly $668M of cash, plus redemption at par of XOMA’s 8.625% Series A and 8.375% Series B Cumulative Perpetual Preferred. Funded entirely from cash on hand, with no financing condition. Voting and support agreements covering 47.0% of XOMA’s voting power — including BVF Partners funds — were signed at announcement, and the break fee was $40M.

What was acquired: 7 commercial programs, ~14 late-stage clinical assets, 100+ earlier clinical and preclinical programs, ~$2.3B of publicly disclosed potential milestones, and more than $110M of Section 174 R&D tax credits and NOLs usable over three to five years. XOMA’s ~$30M of standalone annual operating expense drops to under $5M. Guided accretion: ~$0.50 of adjusted EPS in H2 2026 and ~$1.50 in 2027.

The underwriting test: $668M for ~$1.50 per share of 2027 adjusted earnings on ~21.8M shares implies roughly $32.7M of incremental adjusted earnings, a ~4.9% year-one earnings yield — before the tax attributes (worth ~16% of the purchase price in recoverable cash taxes) and before any milestone conversion. Critically, Davis stated that the great majority of the acquired assets were valued at zero in the underwriting: “a significant majority of the assets that we either do not have enough information about or they were just very early. So we valued those at 0.” Underwriting to the assets you can see and treating the rest as free optionality is exactly the discipline one wants. This looks like a good deal, competently priced. It is the best single piece of evidence for management’s capital-allocation skill.

The convertible financing — good economics, poor disclosure hygiene. On 2026-06-25 Ligand completed $700.0M of 0.00% Convertible Senior Notes due 2031-09-15, including the full $75M greenshoe. Initial conversion price $334.27. Alongside it, Ligand paid $195.79M for convertible note hedges struck at $334.27 and received $114.10M for warrants on 2,094,120 shares struck at $524.34. Debt discount and issuance costs were $20.175M.

Management’s framing was emphatic and repeated: “exceptionally attractive long-term capital,” a “0 coupon” offering, “very low cost of capital,” a transaction that “lowers our long-term cost of capital.” Run the actual cash:

The net call-spread premium of $81.7M equals 11.67% of principal. The all-in economic cost is approximately 3.1% per annum — before counting the equity upside surrendered above $524.34 on 2.09M shares. That is genuinely attractive financing for a levered royalty business, and materially better than the alternatives. But it is not 0%, and the gap between “0 coupon” and 3.1% is roughly $21M a year of real economic cost that no headline figure captures. Analysts should model the 3.1%.

The 2030 notes are deep in the money and represent unrecognised claim on the equity. The August 2025 issue of $460.0M of 0.75% notes due 2030-10-01 converts at ~$194.79 — 5.1338 shares per $1,000. At $294.595 the 2.362M underlying shares are worth ~$696M against $460M of principal: ~$236M of value has already transferred to noteholders. The 10-Q carries the 2030 notes at a Level 2 fair value of $746.4M against $447.6M of carrying value. Standard enterprise-value calculations that use carrying value understate the claim on Ligand’s equity by roughly $282M across both issues. Notably, the 2031 notes were structured with a call spread to limit dilution; the 2030 notes were not.

The buyback. Ligand repurchased ~229,000 shares for $60.0M in Q2 2026, an average of roughly $262 per share. The CFO’s stated rationale is revealing: it reflected “our confidence in the intrinsic value of the company while helping offset the market impact from the convertible hedge activity.” At ~1% of market capitalisation, executed for a technical reason around a convertible pricing, this is not a valuation-driven return of capital. It is also internally consistent only if management believes fair value lies between $262 (where it bought) and $524.34 (where it sold upside via warrants) — a wide band that tells the reader little.

Deployment discipline. The 2025 plan was to invest $150–200M and add ~5 new programs. Actual: $131M invested, 10 new programs added. Under-deploying against plan while over-delivering on program count is the right kind of miss — it suggests price discipline rather than deal hunger. Management targets $100–125M per year for perpetuity-level growth and ~$200M± of capacity, and states explicitly that the lower cost of capital has not changed underwriting hurdles: “Our strategy and our underwriting criteria as well as target returns remain unchanged. Lower cost of capital just allows us to deliver greater alpha and a spread for our investors.” That is the correct answer. It is also unverifiable from outside, and it is the answer every levered acquirer gives.

Incentive alignment — the weakest part of the file.

Three problems.

First, the annual targets were soft. The 2025 Financial objective (30% weight) was “Revenue of $190 million” and “core cash flow of $120 million.” Actual: revenue $268.1M (141% of target) and core cash flow $215M (179% of target), scoring 160%. Beating a revenue target by 41% and a cash target by 79% in a single year is more plausibly evidence that the targets were undemanding than evidence of extraordinary performance. Note also that the $268.1M “achievement” includes the $53.1M one-time Pelthos divestiture gain — a divestiture gain counted toward a revenue target. Total weighted achievement across the three objectives (BD 137%, Financial 160%, Operations 126%) drove the CEO’s cash bonus to $764,400 against a $507,150 target.

Second, the PSUs vest on projections, not results. The 2025 PSUs are measured on two equally weighted metrics: “the level of projected 2029 revenue derived from acquired assets over a two year period ending December 31, 2026” and “the projected adjusted EPS achieved through December 31, 2027.” Paying management on projected revenue from acquired assets rewards the underwriting assumptions used to justify acquisitions — precisely the variable management most controls and most benefits from inflating, in precisely the business where that variable is hardest for outsiders to audit. There is no relative-TSR metric anywhere in the program.

Third, the metric itself is management-defined. Both the annual bonus and the PSUs key off “core cash flow” and “adjusted EPS” — non-GAAP measures whose construction management controls, and which in FY2025 differed from audited operating cash flow by $166M.

Insider behaviour — a clean and unflattering signal. Across the entire five-year Form 4 corpus there are only eight open-market purchases, totalling ~20,010 shares: Foehr (2,500 @ $82.76, May 2022); CEO Davis (4,000 @ $59.38, Sept 2023; 2,500 @ $97.39, Aug 2024; 9,510 @ ~$105, May 2025); CFO Espinoza (1,500 @ ~$104, May 2025). Those were genuine conviction purchases at genuine lows, and they were right.

There has not been a single open-market purchase by any insider since 2025-05-09 — fifteen months during which the stock rose from ~$105 to $294.595. Over the same period, sales:

Two qualifications are owed. CEO Davis does not appear in the sales table at all — he bought and has held, which is the most important single data point and a genuine mitigant. And with 60 Form 144s in the corpus, a substantial share of these sales is likely 10b5-1 plan-driven diversification rather than discretionary conviction; individual plan status was not verified filing-by-filing and no single sale should be over-read. With those caveats, the aggregate pattern is still what it is: the people closest to the assets accumulated at $59–$105 and have distributed ~$34M into $190–$275. The CFO in particular bought 1,500 shares at ~$104 in May 2025 and has since sold 54,925 shares at an average of $247.73.

No dividend has ever been paid. Given the reinvestment opportunity set, that is correct.

Verdict: management has allocated capital intelligently on the deals, and has communicated about the financing in a way that flatters it. XOMA looks well-priced and well-structured; deployment discipline in 2025 was real; the convertible market was used opportunistically and at a genuinely low cost. Against that, the “0 coupon” framing understates the true ~3.1% cost by omitting an $81.7M call-spread premium; the 2030 notes have already transferred ~$236M of value; annual incentive targets were set at levels beaten by 41–79%; and PSUs vest on management’s own projections with no relative-TSR discipline. On the substance, positive. On the governance and the disclosure hygiene, mixed — and the insider distribution pattern deserves to be weighed.

8. Changes and Headwinds — Last Two Years

The strategic arc. The two years to August 2026 completed a transformation begun in 2022. The sequence:

The one accounting/structural change that matters. Post-XOMA, financial royalty assets become a much larger share of the portfolio. The CFO flagged the consequence directly: milestone cash received on those assets “may not be recognized entirely as revenue in the period received. A portion may instead reduce the carrying value of the underlying asset.” Reported revenue will therefore become a progressively worse proxy for portfolio cash economics, and the divergence between GAAP presentation and management’s adjusted measures will widen rather than narrow. Management has committed to “providing additional transparency around both the accounting presentation and the underlying economics,” and the December 2026 Investor Day is the stated venue. That commitment should be held to.

Headwinds and disappointments, honestly stated.

Kyprolis is in structural decline. Amgen’s Q2 2026 net sales of $314M were down 17% year-over-year on lower volume, following a full-year decline from $1,627M to $1,529M. This is Ligand’s largest single royalty at $35.5M.

Mirum’s PSC pathway slipped. At the pre-NDA meeting the FDA recommended a Phase III, moving submission from H2 2026 to H1 2027. Breakthrough therapy designation was granted after the meeting — management fairly called this “mixed signals” and said the company believes approval remains achievable without another Phase III. Ligand made “no major changes in terms of our projections.”

Agenus discontinued a Phase III. The BOT/BAL trial in relapsed/refractory colorectal cancer was discontinued; a new neoadjuvant trial will not dose its first patient until Q1 2027. Agenus raised $85M upfront in July 2026 (up to $225M more on warrant exercise), with Ligand participating. Ligand holds a 2.625% royalty on future global net sales. This is a multi-year deferral.

Leverage is new and material. Ligand carried effectively no long-term debt at 2025-06-30 ($4.6M of finance leases). Fourteen months later it carries $1.16B of convertible principal. The Amended & Restated Credit Agreement imposes a consolidated senior secured net leverage ratio of ≤2.50x (3.00x for four quarters around a material permitted acquisition) and a minimum consolidated EBITDA (as defined) of $100M through Q1 2027, stepping to $150M from Q2 2027. TTM GAAP EBITDA is $127.6M and FY2025 GAAP EBITDA was $74.8M; the covenant is measured on an add-back-inclusive “as defined” basis and is likely to be met, but the step-up is a real, dated constraint that bites precisely in the scenario where readouts disappoint.

Guidance quality. The Q2 raise lifted only the floor of adjusted EPS guidance, to $9.00–9.50, and the CFO attributed it to “the incremental earnings contribution from the proceeds of our 0 coupon convertible note offering, together with the benefit of the share repurchase.” All revenue guidance was merely reaffirmed. That is a financial raise, not an operational one, and is lower quality than the headline suggests.

The long-term framework is stale. Management stated that “the business today is materially different than the one reflected in the long-term framework we introduced last December,” and will re-baseline at the December 2026 Investor Day. Investors are therefore currently valuing the company without a current management long-range plan.

Verdict: on balance these changes strengthen the thesis, but they also raise the stakes. The portfolio is genuinely broader, longer-duration and better diversified than two years ago; Filspari’s FSGS approval materially expanded the single most important asset; and XOMA added scale, tax attributes and optionality at a defensible price. Against that, the company has taken on $1.16B of debt, made its accounting harder to read, seen its largest legacy royalty decline 17%, absorbed two clinical setbacks in a single quarter, and is operating without a current long-range framework. The transformation is real and largely favourable. The margin for error is smaller than it was.

9. Risk Analysis

Risk of catastrophic or total loss: low. This is a diversified portfolio of contractual claims on approved drugs sold by investment-grade counterparties, held with a strong liquidity position. Even a severe adverse scenario — a failed pivotal, a Filspari disappointment, a closed financing window — impairs growth and compresses the multiple rather than threatening solvency. The convertible notes do not mature until October 2030 and September 2031, and the company holds ~$700M of deployable capital. The realistic bear case is a large multiple de-rating, not a wipeout.

The two risks that dominate. Numbers 1 and 2 are the ones that actually determine outcomes. Filspari concentration is a business risk; adjusted-earnings quality is a perception risk that requires no business deterioration at all to express itself. An investor should be clear which of the two they are being paid to bear — at the current multiple, both.

10. Valuation Discussion

No price target and no recommendation appear in this section; valuation is discussed only as embedded expectations and scenarios.

Capital structure at 2026-08-14. Price $294.595. Shares outstanding 19.920M (2026-06-30); Q2 2026 diluted share count 21.837M. Basic market capitalisation $5,868M; on the diluted count $6,433M. Convertible principal $1,160M ($700M 2031 notes + $460M 2030 notes), disclosed Level 2 fair value $1,408M. Management states approximately $700M of deployable capital remains after the convertible and the XOMA close. Taking net debt at roughly $1,160M less ~$700M gives ~$460M, and an enterprise value of approximately $6,328M on the basic count. Long-term investments of $351.1M are additional non-core value not netted here; marking the converts to fair value rather than carrying value would add ~$282M.

The multiple ladder. This is the heart of the valuation question. Using 2026 guidance midpoints — royalty revenue $237.5M, total revenue $290M, adjusted EPS $9.25:

The spread between 31.8x and 53.2x is not an accounting quibble — it is the difference between a demanding-but-defensible growth multiple and an extreme one, and it turns entirely on whether stock compensation and royalty depletion are costs. The honest answer is that SBC certainly is, and royalty amortisation partly is (Ligand buys at a discount to expected cash flows, so book amortisation overstates true depletion, but it is not zero). A defensible middle reading sits near the 42x line.

Own-history context. LGND’s EV/TTM sales has run 10.7x (FY2021), 5.1x (FY2022), 8.3x (FY2023), 10.7x (FY2024), 12.8x (FY2025) — and stands at 20.8x today. EV/TTM EBITDA was 45.9x at FY2025 and is 47.5x now. The AZI own-history percentile ranks, which compare the current multiple only against the stock’s own decade, are unambiguous on the two metrics not distorted by mark-to-market accounting:

The P/E percentile should be disregarded: TTM GAAP EPS of $7.505 is inflated by non-operating investment marks, which flatters the earnings and drags the percentile down to a misleadingly benign level. On price-to-sales and price-to-book, Ligand is at or near the most expensive it has ever been relative to its own history.

Peer context. Royalty Pharma is the correct comparator — not the XBI biotech complex that the factor model statistically associates Ligand with:

On the sales multiple Ligand actually looks cheaper than RPRX — but that comparison is misleading, because RPRX’s revenue line understates its cash economics far more than Ligand’s does. On the cash-flow measure, which is the right one for a royalty business, Ligand trades at roughly 2.5x Royalty Pharma’s multiple. Ligand’s superior growth rate justifies a premium; whether it justifies 2.5x is the question, and a growth premium of that magnitude requires the growth to persist for a long time.

Embedded expectations — what the price requires. At $294.595, the market is simultaneously underwriting five propositions:

1. Filspari reaches or approaches Travere’s reiterated >$3B peak with the 9.0% rate intact — worth ~$270M of annual royalty alone, more than today’s entire guided royalty base.
2. XOMA delivers its guided ~$1.50 of 2027 accretion and converts a meaningful portion of the $2.3B milestone pool that management explicitly valued at zero.
3. Deployment continues at $100–200M per year at unchanged spreads, notwithstanding management’s own description of a high-demand, low-supply market — the classic precondition for competitive entry.
4. Continued access to sub-3.5% convertible capital, without which the growth algorithm requires dilutive equity.
5. The adjusted-earnings presentation is the economically correct one — that $2.22 per share of stock compensation and $1.49 per share of royalty depletion are genuinely not costs.

All five must broadly hold. Proposition (5) is the one most investors have not consciously priced, and it is the one that requires no business deterioration whatsoever to go wrong.

Scenario analysis (illustrative; assumptions stated; no price target implied):

The distribution is genuinely wide and genuinely right-skewed in the tail — the seven pivotal readouts and the $2.3B milestone pool are real, unpriced optionality that management underwrote at zero. But the base case does not offer a margin of safety at 41.9x SBC-burdened current-year earnings and a 98.9th-percentile price-to-sales.

What the market is pricing correctly, and what it may not be. Correctly: the quality of the transformation, the durability of the contractual revenue, the value of the operating leverage, and the competence of the XOMA underwriting. Possibly not: the degree of Filspari concentration hidden inside a diversification narrative; the ~3.1% (not 0%) true cost of the 2031 convertible plus the ~$236M already transferred via the 2030 notes; the $166M gap between GAAP operating cash flow and the core-cash-flow metric on which guidance is framed; and the ~$3.71 per share of real costs that the adjusted presentation removes.

11. Variant Perception

The consensus belief. Ligand is a successfully transformed, capital-light royalty compounder with a proven management team, a diversified and lengthening portfolio, a fortress balance sheet with $700M of dry powder, an accretive transformational acquisition just closed, seven pivotal catalysts in eighteen months, and a five-year record of compounding adjusted EPS from $2.44 to $9.50. At ~32x current-year adjusted earnings for that growth profile, the stock is reasonably priced for quality. The sell-side questions on the Q2 call were uniformly congratulatory — “congrats on a very good momentum,” “congratulations on this integration,” “adding my congratulations” — and focused on upside sizing, not on the accounting.

The strongest bull case. It is genuinely strong and should not be dismissed. Filspari at Travere’s reiterated >$3B peak generates ~$270M of annual royalty at the 9% rate — by itself larger than the entire current royalty base, from an asset already growing 96% year-over-year with a second approval secured in April 2026. XOMA was bought for $668M with more than $110M of tax attributes attached and with the great majority of its 120+ assets underwritten at zero, meaning the $2.3B milestone pool and the 100+ early programs are free options; Davis says of the unpartnered assets that “for very small levels of investment, in some cases, a couple to $3 million, you can validate some of these earlier-stage assets in the clinic and make them very licensable.” Duration now extends in some cases to 2040. The company has $700M to deploy into a market it describes as high-demand and low-supply, at a ~3.1% cost of capital, with a forty-seven-person cost structure that converts thin gross royalty rates into fat net per-share earnings. If even two of seven pivotal readouts hit, the 2029 earnings base looks nothing like today’s.

The strongest bear case. The bear case is not that the business is bad — it is that the earnings being capitalised are softer than they appear and the concentration is greater than disclosed. Strip the FY2025 headline: 60% revenue growth becomes ~29% ex the $53.1M divestiture gain booked inside revenue. Strip GAAP net income: 70% of FY2025 pretax income and 87% of Q2 2026 pretax income were non-operating investment marks. Strip the cash guidance: GAAP operating cash flow was $49.4M in FY2025 against the $215M “core cash flow” management is paid on, and the “>$200M in 2026” guidance almost certainly refers to the latter. Strip adjusted EPS: $2.22 of stock compensation and $1.49 of royalty depletion per share take 31.8x to 41.9x and then 53.2x. Then observe that the diversification narrative conceals a single-asset thesis — one drug worth more than the entire current royalty base — while the largest legacy royalty declines 17%. Then observe that the “moat” is a lean cost structure in a niche with no barrier to entry, in a market management itself describes in the exact terms that attract competing capital. Then observe that the stock sits at the 98.9th percentile of its own price-to-sales history, and that insiders who bought at $59, $97 and $105 have bought nothing in fifteen months while selling ~$34M into $190–$275.

The three to five assumptions that actually matter.

1. Filspari’s peak and rate. Everything downstream flows from whether >$3B at 9% is achievable. Falsified by: a competitive entrant in IgAN, a rate renegotiation, a patent challenge, or two consecutive quarters of decelerating Travere net sales. Confirmed by: sustained >60% growth through 2027 and FSGS uptake tracking to management’s framing.
2. Whether GAAP cash converges on core cash flow. Falsified by: FY2026 GAAP operating cash flow materially below ~$150M, or FY2027 below ~$200M. Confirmed by: FY2027 GAAP CFO above ~$200M, which would prove the add-backs are economically real.
3. Whether deployment spreads hold. Falsified by: falling incremental royalty yields on new deals, larger cheque sizes at the same return, or management widening the deal-size band. Confirmed by: continued $100–200M annual deployment at unchanged disclosed underwriting hurdles through a full cycle.
4. Whether XOMA delivers. Falsified by: 2027 accretion below ~$1.00, or the cost base failing to reach <$5M. Confirmed by: H2 2026 accretion of ~$0.50 landing on schedule and the December Investor Day quantifying milestone conversion.
5. Whether the market continues to capitalise adjusted EPS. Falsified by: any broad market shift toward charging SBC — this requires no company-specific deterioration at all.

Where consensus may be offside — the positioning evidence. The factor data is unusually informative here and cuts against the lazy reading in both directions. Ligand has delivered a 92% one-year return with a maximum drawdown of only 14% and a Sharpe ratio of 2.50, sitting 8.7% below an all-time-window high, above all three major moving averages. That is textbook momentum. But the factor R² is only 0.22–0.29 and idiosyncratic volatility is 31.9% annualised — meaning roughly 70%+ of the stock’s variance is company-specific. Loadings are modest and unremarkable: Market 0.91, SmallSize 0.89 (a known upward bias for biotech-classified issuers), Biotech SPDR industry 0.78.

The correct interpretation is that this is not a crowded factor trade. Ligand is not being carried by the momentum factor or by a biotech beta rally, and it will not be dragged down by one. It is an idiosyncratic re-rating — a stock the market has independently re-priced on company-specific news flow. That materially changes the risk profile from the usual “crowded momentum name” framing: there is no factor unwind waiting to happen, but equally there is no sector recovery to bail out a disappointment. The stock will trade on Filspari, XOMA and the convertible market, and on nothing else.

One further tell: the FactorsToday statistical peer set for Ligand is ARWR, DNLI, PGEN, MGTX, SYRE, NRIX, RARE, KNSA and the XBI/SBIO/LABU ETFs — clinical-stage biotechs, not specialty finance companies. The market is behaving as though Ligand is a biotech. Its economics are those of a levered royalty lender. Whichever way that mismatch resolves, it argues strongly for benchmarking against Royalty Pharma rather than against XBI — and on the cash-flow measure that comparison shows Ligand at ~2.5x RPRX’s multiple.

The variant view, stated plainly. Consensus is right about the business and insufficiently curious about the earnings. The three genuinely non-consensus observations in this file are: (i) the $166M gap between FY2025 GAAP operating cash flow of $49.4M and the proxy-disclosed “core cash flow” of $215M, which conditions how every forward cash guidance figure should be read; (ii) the $53.1M one-time divestiture gain booked inside FY2025 revenue, which turns 60% growth into ~29%; and (iii) the ~3.1% true all-in cost of the “0% coupon” convertible, plus ~$236M of value already transferred through the un-hedged 2030 notes. None of these makes Ligand a bad business. All three make it a more expensive one than the headline multiple suggests.

12. Fact vs. Interpretation

13. Open Questions

1. What is the actual GAAP-to-core-cash-flow bridge? The company has never published one. Without it, the $166M FY2025 gap cannot be independently decomposed, and every forward cash guidance figure is uninterpretable. This should be the first question at the December 2026 Investor Day.
2. Does management’s “>$200M in 2026, ~$300M in 2027” cash guidance refer to GAAP operating cash flow or to core cash flow? The interpretation here (core cash flow) is well-supported but is an inference. A direct answer would materially change how the stock should be modelled.
3. What is the XOMA purchase-price allocation? Goodwill, intangible royalty assets, financial royalty assets and the fair value assigned to the acquired pipeline will not be visible until the Q3 2026 10-Q. The split between intangible and financial royalty assets determines how much of XOMA’s economics will ever appear in revenue.
4. What proportion of the $2.3B XOMA milestone pool is near-term and probability-weighted meaningful? Management declined to quantify: “We’re not prepared to give further visibility to the quantum or the partners.” Deferred to the December Investor Day.
5. What is Filspari’s royalty rate structure at scale? The 9.0% effective rate is disclosed at current volumes. Whether it tiers down at higher sales — common in royalty agreements — would materially change the >$3B peak arithmetic. Not disclosed.
6. How much of the insider selling is 10b5-1 planned? Sixty Form 144s suggest a substantial share, but plan status was not verified filing-by-filing. This materially affects how much weight the distribution pattern deserves.
7. What is the remaining weighted-average duration of the royalty portfolio? Management cites IP “in many cases” extending beyond the next decade and “in some cases through 2040,” but no portfolio-weighted duration figure is disclosed. For a royalty aggregator this is arguably the single most important undisclosed metric.
8. What return does management actually underwrite to? Davis says underwriting criteria and target returns “remain unchanged” but never states them. Without a disclosed hurdle rate, spread compression cannot be detected from outside until it appears in results.
9. How replaceable is the investment team? With 47 employees and an investment function resting on a handful of people, succession for the underwriting capability — not the executive suite — is the relevant question, and nothing is disclosed.
10. Will the December 2026 Investor Day re-baseline the long-term framework upward, and on which metric? Management has said the December 2025 framework is stale. Whether the new framework is anchored on adjusted EPS, core cash flow or GAAP cash flow will reveal how the company intends to be valued.

14. What Must Be True

Bull case

The claim: Ligand compounds royalty revenue toward $500M+ by 2029–30, with adjusted EPS well above $15, as Filspari scales toward $3B, XOMA’s underwritten-at-zero assets convert, and $150–250M of annual deployment continues at unchanged spreads funded by cheap convertible capital.

What must be true:

1. Filspari reaches $2.5–3B+ in peak sales with the 9.0% rate substantially intact.
2. At least two of the seven pivotal readouts in the next eighteen months succeed and initiate meaningful new royalties.
3. XOMA delivers the guided ~$1.50 of 2027 accretion and converts a material share of the $2.3B milestone pool.
4. Annual deployment of $100–200M continues at unchanged underwriting returns, without spread compression from competing capital.
5. Convertible or similarly cheap capital remains available at or below ~4%.
6. GAAP cash conversion improves materially, validating the adjusted presentation.

Falsification test: If, by the FY2027 10-K, GAAP operating cash flow has not exceeded ~$200M — versus $49.4M in FY2025 and management’s guided ~$300M — then the adjusted earnings are not converting to cash, the core-cash-flow metric is an accounting construct rather than an economic one, and the bull case is broken regardless of what the adjusted EPS line shows. A secondary, faster test: if Travere’s Filspari net sales growth decelerates below 40% year-over-year for two consecutive quarters, the >$3B peak assumption is in question and with it the entire growth spine.

Bear case

The claim: Ligand is a well-run levered specialty-finance company trading at a compounder’s multiple on adjusted earnings that exclude ~$3.71 per share of real costs; a single pivotal disappointment or a closed financing window triggers a de-rating from 42x SBC-burdened earnings toward the high teens or low twenties.

What must be true:

1. SBC and royalty-asset amortisation are genuinely economic costs that the market eventually charges.
2. Filspari peaks materially below $3B, or its rate tiers down at scale.
3. Competing capital compresses deployment spreads as returns become visible.
4. GAAP cash conversion stays well below the core-cash-flow metric.
5. The convertible market tightens, forcing dilutive equity or expensive debt.

Falsification test: If, over the next four quarters, Filspari sustains >60% year-over-year growth, XOMA’s ~$0.50 of H2 2026 accretion lands on schedule, and GAAP operating cash flow for FY2026 prints above ~$150M, then the earnings quality concern is substantially answered and the bear case collapses — the adjusted numbers would be demonstrably converting to cash and the growth spine would be intact. A secondary test: if Ligand executes two or more new royalty transactions at disclosed returns equal to or better than its historical underwriting, the spread-compression thesis is disproved for that cycle.

15. Source Appendix

See Appendix B below.

The body of this article (sections 1–15) contains no investment recommendation and no price target; the Claude's Take block is explicitly labelled as the author’s own subjective view. This article is general information, not investment advice, and the author may or may not hold a position in any security discussed. Do your own research.

APPENDIX A — Standard Diligence Questionnaire

Ligand Pharmaceuticals Incorporated (NASDAQ: LGND) — 2026-08-15

A supplemental due-diligence questionnaire. Answers are labelled Fact / Interpretation / Assumption where the distinction matters. Where a question does not map to a royalty-aggregation business model, the correct sector analog is given instead.

General

What thoughtful questions have other investors asked about this company?

The Q2 2026 call (2026-08-06) drew questions from Piper Sandler-adjacent coverage, Stifel (Annabel Samimy), Citigroup (Yigal Nochomovitz), Bank of America (for Jason Zemansky), Oppenheimer (Leland Gershell), H.C. Wainwright (Joseph Pantginis) and RBC (Sahil Dhingra). The recurring themes were:

• “Isn’t XOMA a concentrated bet given the size?” — Davis pre-empted this in prepared remarks, arguing the breadth of the acquired portfolio reduces rather than concentrates risk. Interpretation: true at the asset-count level, false at the earnings level, where Filspari remains dominant.
• Deal-size drift. Samimy asked whether, “with your coffers refilled,” Ligand would need bigger, portfolio-scale transactions. Davis said no — the sub-$100M market remains target-rich and deal sizes will stay at $25–75M. This is the single most important discipline question in the file and deserves ongoing monitoring.
• Does cheaper capital lower the hurdle rate? Nochomovitz asked directly whether the $700M convertible changes underwriting discipline. Davis: “Our strategy and our underwriting criteria as well as target returns remain unchanged. Lower cost of capital just allows us to deliver greater alpha and a spread.” The right answer; unverifiable from outside.
• How much of XOMA’s $1.50 of 2027 accretion is organic versus synergies and tax? BofA pressed this; Espinoza said it is “100% tied to the top line contribution from XOMA plus the significant synergies.” He did not break out the split.
• Timing and probability of the $2.3B milestone pool. RBC asked; management declined to quantify, deferring to the December 2026 Investor Day.
• The XOMA “zombie strategy.” Hewitt asked about XOMA’s practice of acquiring distressed shells for their tax assets. Davis said Ligand has looked at it, would not do very small versions, but is “looking at a number of companies that do provide some tax benefits right now in our active pipeline.”

Notably absent from every analyst question: the gap between GAAP operating cash flow and “core cash flow”; the treatment of SBC and royalty amortisation in adjusted EPS; the true all-in cost of the “0% coupon” convertible; and the $53.1M one-time gain inside FY2025 revenue. Interpretation: the sell-side conversation is entirely about upside sizing, not about earnings quality. That is where the variant perception lives.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: at a cyclical high on two dimensions. First, non-operating investment marks contributed $111.8M of FY2025’s $159.0M pretax income and ~$47.5M of Q2 2026’s $48.5M net income — these are two-way and mean-reverting (H1 2026 recorded a $13.5M loss on the same line). Second, the cost of capital is at a cyclical low: a 0.00% coupon convertible reflects an accommodating market, not a permanent condition. The royalty earnings themselves are early-cycle, not late — Filspari and Ohtuvayre are ramping steeply.

Driven by the external environment or internal actions? Both, and the split matters. Internal: the 2022 pivot, the OmniAb and Pelthos separations, headcount cut to 47, and the XOMA acquisition are genuine management actions. External: the accommodating convertible market, the difficult biotech equity environment that creates royalty demand, and the partners’ commercial execution (Travere, Merck) are all outside Ligand’s control. Interpretation: management deserves credit for the internal half; the external half should not be extrapolated.

How stable are revenues? Royalties ($161.0M FY2025, 60% of revenue) are contractual and stable, paid by investment-grade marketers (Amgen, Merck, Jazz, Recordati, Sanofi, and post-XOMA Roche, Servier, Ipsen, Zevra). Captisol ($40.2M, 15%) is stable in character but lumpy in timing — management guided H2 2026 to ~40% Q3 / ~60% Q4 purely on order timing. Contract revenue ($66.9M, 25%) is not stable: 2026 guidance is $10–20M against $66.9M in 2025, because 2025 included the $53.1M one-time Pelthos gain. Fact. Top-three partner concentration was 51% of total revenue in FY2025, improving from 63% in FY2023.

Outlook for products/services? Mixed by asset. Growing: Filspari (+96% partner sales YoY in Q2 2026, second indication approved April 2026, partner guiding >$3B peak), Ohtuvayre (+98%), Capvaxive, Vabysmo (newly acquired). Declining: Kyprolis (−17% YoY, still the largest single royalty at $35.5M), Evomela, Teriparatide. Deferred: Agenus BOT/BAL (Phase III discontinued, restart Q1 2027), Mirum in PSC (FDA recommended a Phase III; submission slipped to H1 2027).

How big will this market be — growing, shrinking, domestic or international? Growing. Management estimates royalty capital at <9% of biopharma financing overall and <5% on the development side — a genuinely under-penetrated market. Demand is driven by a difficult biotech equity environment. Predominantly US-centric in origination, but the royalties are on global net sales (Vabysmo, Ojemda in Europe via Ipsen, Pneumosil internationally). Interpretation: the demand side is durable; the supply of competing capital is the variable to watch.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Interpretation: more, and the mechanism is visible. Management describes the market in exactly the terms that attract capital — “very, very high demand and low supply of capital,” royalty finance <9% of biopharma funding, “only a few players that really do this consistently.” Under Marathon’s capital-cycle framework, high returns in an unbarriered niche mean-revert as supply responds. Private credit funds with lower hurdles and larger balance sheets face no structural obstacle to moving down-market. Ligand’s protection is that it is early, disciplined and low-cost — not that entry is hard.

How profitable is the business (ROIC, ROE)? FY2025 ROE 13.5%, ROA 9.9%, ROIC 2.77%. Fact. The ROE/ROIC divergence is arithmetic: GAAP operating income is suppressed by royalty-asset amortisation while invested capital carries the assets at full purchase cost. Interpretation: this is the central analytical handicap in the name. For a business whose entire case is “we deploy capital at attractive returns,” those returns are not observable in the audited operating statements — they exist only in management’s adjusted presentation and its “core cash flow” metric. Gross margin of 94.6% and incremental operating leverage are genuine and measurable; the return on deployed capital is asserted, not demonstrated.

How profitable is the industry — how many competitors, what barriers to entry? Royalty Pharma earns a 62% TTM operating margin on $2,536M of revenue with a $59.3B enterprise value — the industry is highly profitable at scale. Competitors: RPRX, Blackstone Life Sciences, HealthCare Royalty, Oberland, DRI, Sagard, plus an expanding private-credit fringe. Barriers to entry: effectively none. Under Greenwald’s taxonomy there is no supply/cost advantage (capital is a commodity), no customer captivity (management concedes partners “almost always have alternatives”), and no scale-plus-captivity. Interpretation: this is a good business without a moat.

Can the business be easily understood? The model yes — buy royalties, run lean, collect. The accounting no. The split between “intangible royalty assets” (revenue-recognised, amortised) and “financial royalty assets” (effective-interest income, with receipts partly reducing carrying value and flowing through investing cash flow) means reported revenue systematically understates portfolio economics, while adjusted EPS systematically overstates them. This divergence widens post-XOMA, as the CFO explicitly warned. Interpretation: above-average accounting complexity for a business that looks simple.

Can it be undermined by foreign low-cost labour? No. Not applicable to a 47-person capital-allocation business. The relevant analog is undermining by lower-cost capital, which is a live risk.

Do brands matter? Not in the consumer sense. What matters is reputation as a repeat, non-predatory counterparty — the ability to be the first call when a small-cap needs $50M. Ligand builds this by following on in partner equity financings (Palvella “in every equity financing they’ve done,” Agenus, Athira). Interpretation: real and accumulating, but it is a relationship asset, not a brand, and it does not prevent a competitor from outbidding on any given deal.

What is the nature of competition? Price competition on royalty rate and deal structure, mediated by speed, certainty of close and diligence quality. Ligand competes primarily by being willing and able to write cheques too small for RPRX and too specialised for generalist credit. Fact (evidence of price-taking): Ligand captured 2.75% of $4.8B of FY2025 partner product sales; on the largest products the effective rates are 2.3% (Kyprolis), 1.3% (Capvaxive), 0.9% (Vaxneuvance), and ~0.5% on the newly acquired Vabysmo. Where the product is large, Ligand’s rate is small.

Customers’ switching costs? Zero. A royalty sale is a one-time transaction, not a recurring relationship. There is nothing to switch away from.

Financial Condition & Balance Sheet

Assets not fully recognised on the balance sheet? Yes, materially. (i) The $2.3B of disclosed potential XOMA milestones carry no balance-sheet value. (ii) The 100+ early-stage XOMA programs that management explicitly valued at zero in the underwriting — Davis: “a significant majority of the assets that we either do not have enough information about or they were just very early. So we valued those at 0.” (iii) The Tremfya CVR, entitling Ligand to 25% of any net proceeds from the Janssen litigation, with “no governance responsibilities, no obligation to fund litigation costs and no downside financial exposure.” (iv) More than $110M of Section 174 credits and NOLs from XOMA, usable over 3–5 years. Interpretation: genuine unrecognised optionality — the strongest part of the bull case.

Off-balance-sheet liabilities? None disclosed of significance. The material understated liability is on-balance-sheet but carried below fair value: the convertible notes are carried at $1,126.6M against a disclosed Level 2 fair value of $1,408.4M — a ~$282M gap, driven mainly by the 2030 notes (fair value $746.4M vs $447.6M carrying) trading deep in the money at a $194.79 conversion price. Fact. Standard EV calculations using carrying value understate the claim on Ligand’s equity.

How conservative is the accounting? Interpretation: GAAP presentation is conservative; the adjusted presentation is not. GAAP requires immediate expensing of certain R&D-financing transactions that are economically investments — Q2 2026 included a $12M charge on the Orchestra Bio investment for exactly this reason, and H1 2025 R&D of $56.7M (versus $16.8M in H1 2026) reflects similar treatment. That is conservative and works against the company. Working the other way: booking a $53.1M divestiture gain inside revenue, and an adjusted EPS definition that removes SBC, intangible amortisation and financial-asset amortisation. The two presentations differ by ~$3.71 per share.

How CapEx-hungry is the business? Not at all in the conventional sense — FY2025 purchases of property and equipment were $0.45M; H1 2026, $0.25M. The correct sector analog is capital deployment into royalty assets, which is the entire business: $131M invested in 2025 against a $150–200M plan, with management targeting $100–125M annually for perpetuity-level growth and ~$200M± of capacity. Unlike conventional capex, this “maintenance capex” is genuinely required — royalty assets deplete, and the base must be replaced to stand still. Interpretation: this is precisely why adding back royalty amortisation in adjusted EPS is analytically aggressive.

Capital Allocation & Management

How much FCF does the business generate, how does management use it, what is the philosophy? GAAP free cash flow: $119.9M (FY2022), −$4.3M (FY2023), $77.4M (FY2024), $38.0M (FY2025). Fact. Management’s own metric — proxy-disclosed “core cash flow,” defined as “revenue less 1) cost of sales, and 2) cash operating expenses” — was $215M in FY2025, and the CFO guides to “>$200M in 2026, growing to ~$300M in 2027.” The $166M FY2025 gap between $49.4M of GAAP operating cash flow and $215M of core cash flow is the most important single number in this file. Philosophy: reinvest everything into royalty acquisition; no dividend has ever been paid; buybacks are technical and occasional.

Significant acquisitions recently? XOMA Royalty, closed 2026-07-14 — the largest in company history. $39.00 per share cash plus a CVR (~$668M against 17.13M shares including Series X conversion), plus redemption at par of XOMA’s 8.625% Series A and 8.375% Series B perpetual preferred. Funded entirely from cash on hand, with no financing condition; $40M break fee; support agreements covering 47.0% of XOMA’s voting power including BVF Partners. Acquired: 7 commercial programs, ~14 late-stage clinical, 100+ earlier, $2.3B of disclosed milestones, >$110M of tax attributes, cost base cut from ~$30M to <$5M. Guided accretion ~$0.50 (H2 2026) and ~$1.50 (2027). Interpretation: ~4.9% year-one earnings yield plus tax attributes worth ~16% of purchase price plus free optionality — a competently priced deal and the best evidence of management’s capital-allocation skill.

Buying back shares? Yes, but modestly and for technical reasons. ~229,000 shares for $60.0M in Q2 2026 (~$262/share), which the CFO described as reflecting “confidence in the intrinsic value of the company while helping offset the market impact from the convertible hedge activity.” Interpretation: ~1% of market cap, executed around a convertible pricing, not a valuation-driven return of capital. Note the tension: buying stock at ~$262 while selling warrants struck at $524.34.

Issuing large amounts of new shares to insiders? Yes — this is the weakest quantitative fact in the file. FY2025 SBC was $46.8M, 17.5% of revenue, ~$2.31 per diluted share; H1 2026 was $24.2M (~$2.22/share annualised, ~24% of guided 2026 adjusted EPS). At 2026-03-31, 2,403,395 shares were subject to outstanding awards — 12.1% of shares outstanding — with 200,016 available; shareholders approved a further 600,000-share increase on 2026-06-05. All of it is excluded from the adjusted EPS the market capitalises.

Compensation policy of directors/management? CEO Todd Davis total compensation: $6.22M (2023), $12.91M (2024), $11.47M (2025) — 2025 comprising $724,500 salary, $5,520,493 stock awards, $4,441,120 option awards, $764,400 non-equity incentive. CFO Espinoza $4.38M; CLO Reardon $3.75M. Target bonuses 70% / 45% / 40% of salary, 100% tied to corporate performance.

Interpretation — three governance weaknesses. (i) Soft targets: the 2025 Financial objective was revenue of $190M and core cash flow of $120M; actual was $268.1M (141%) and $215M (179%). Beating targets by 41–79% suggests undemanding goal-setting — and the $268.1M “achievement” includes the $53.1M one-time Pelthos divestiture gain, i.e. a divestiture gain counted toward a revenue target. (ii) PSUs vest on projections, not results: the 2025 awards are measured on “the level of projected 2029 revenue derived from acquired assets” and “the projected adjusted EPS achieved through December 31, 2027.” This pays management for underwriting assumptions — the single variable management most controls and that outsiders can least audit. (iii) No relative-TSR metric anywhere in the program, and every metric used is management-defined non-GAAP.

Motivations of management? Fact — the insider record. Across the full five-year Form 4 corpus there are only 8 open-market purchases (~20,010 shares): Foehr 2,500 @ $82.76 (May 2022); CEO Davis 4,000 @ $59.38 (Sept 2023), 2,500 @ $97.39 (Aug 2024), 9,510 @ ~$105 (May 2025); CFO Espinoza 1,500 @ ~$104 (May 2025). Those were genuine conviction purchases near lows and they were right.

There has been no open-market purchase by any insider since 2025-05-09 — fifteen months during which the stock rose from ~$105 to $294.595. Over the same period insiders sold ~145,843 shares for ~$34.3M: CFO Espinoza 54,925 shares at an average of $247.73 ($13.61M); CLO Reardon $7.56M at $236.23; directors Haas ($3.74M at $275.12), Aryeh, LaMattina, Kozarich, Gray and Sabba.

Two qualifications are owed. CEO Davis does not appear in the sales table at all — he bought and held, which is the most important mitigant. And with 60 Form 144s in the corpus, a substantial share of the selling is likely 10b5-1 plan-driven diversification; plan status was not verified filing-by-filing and no individual sale should be over-read. Interpretation, with those caveats: the aggregate pattern is that the people closest to the assets accumulated at $59–$105 and have distributed into $190–$275. That is information, not proof.

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No. Ligand is a Delaware corporation, US domestic filer, common stock on NASDAQ (CUSIP 53220K504, ISIN US53220K5048). Ordinary Form 1099 treatment; no K-1. There is a secondary listing on GETTEX (LGDN).

Dividend policy? No dividend has ever been paid, and none is contemplated. Interpretation: correct, given a reinvestment opportunity set management describes as requiring $100–125M annually for perpetuity growth against ~$200M of deployment capacity.

How profitable is the business? FY2025 gross margin 94.6%, operating margin 17.6%, net margin 46.4% (flattered by investment marks), EBITDA margin 27.9%. TTM operating margin 34.5%. Against Royalty Pharma’s 62.4% TTM operating margin, Ligand’s GAAP operating profitability is roughly half — the gap is amortisation of a younger, more recently acquired royalty base plus a proportionally larger cost structure.

Is net income diverging from cash from operations? Yes, severely and persistently — this is the core quality-of-earnings finding.

FY2025 net income of $124.5M converted to just $49.4M of operating cash — a 0.40x ratio — because 70% of pretax income was non-cash investment marks that are correctly reversed out in the cash flow statement. H1 2026 GAAP CFO of $72.7M implies a required ~75% H2 step-up to reach the guided “>$200M,” which is strong evidence the guidance references core cash flow rather than GAAP. Any investor capitalising “$300M of 2027 operating cash flow” as a GAAP figure against a ~$5.9B market capitalisation will reach a materially wrong conclusion.

Risks & Downside

What factors would cause the stock to decline? In descending order of expected impact: (1) Filspari disappointing — a peak materially below $3B, a rate that tiers down at scale, a patent challenge, or competitive entry in IgAN/FSGS; (2) a market shift toward charging SBC and royalty depletion against earnings, which would take the multiple from 31.8x toward 41.9x–53.2x with no business deterioration whatsoever; (3) FY2026–27 GAAP operating cash flow failing to converge on the guided core-cash-flow figures; (4) failure of one or more of the seven pivotal readouts in the next eighteen months; (5) a closed convertible window forcing dilutive equity; (6) visible spread compression as competing capital enters; (7) XOMA accretion falling short of the guided ~$1.50.

Risk of a catastrophic loss? Low. The realistic severe case is a large multiple de-rating, not an impairment of the asset base. The portfolio is a diversified set of contractual claims on approved drugs sold by investment-grade marketers; the 2030 notes do not mature until October 2030 and the 2031 notes until September 2031; ~$700M of deployable capital remains. The one genuine structural constraint is the Amended & Restated Credit Agreement’s minimum consolidated EBITDA (as defined) of $100M through Q1 2027 stepping to $150M from Q2 2027, alongside a ≤2.50x senior secured net leverage test. Against TTM GAAP EBITDA of $127.6M (FY2025: $74.8M), the “as defined” add-backs make compliance likely — but the step-up removes flexibility precisely in the scenario where readouts disappoint.

Chance of a total loss? Negligible on any reasonable horizon. Equity value would have to be wiped out by $1.16B of convertible debt against a diversified contractual royalty portfolio, $1.37B of pre-XOMA liquidity and ~$700M of remaining deployable capital. Even a compound bear scenario — a failed pivotal, a Filspari halving and a closed financing window — leaves a paying royalty portfolio and a solvent balance sheet. This is a de-rating risk, not a solvency risk.

Recent News & Events

Has the business environment changed recently? Yes, materially, in the last four months. (i) Filspari won a second FDA approval in FSGS in April 2026 — the first and only approved treatment — materially expanding the key royalty’s addressable market, and on 2026-08-04 Travere reiterated peak sales potential above $3B. (ii) The XOMA acquisition closed 2026-07-14, more than doubling the royalty portfolio by asset count and extending duration in some cases to 2040. (iii) The company raised $700M of 0.00% convertible notes on 2026-06-25, taking total convertible principal to $1.16B from effectively zero fourteen months earlier. (iv) Two clinical setbacks were disclosed on the Q2 call: Mirum’s FDA-recommended Phase III in PSC (submission slipping to H1 2027) and Agenus’s discontinuation of its Phase III in relapsed/refractory colorectal cancer (restart Q1 2027). (v) Kyprolis, the largest single royalty, declined 17% year-over-year in Q2 2026.

Significant acquisitions? XOMA Royalty — see above. Also relevant as ongoing pattern: the Orchestra Biomed transaction (closed 2025-08-04, with a $12M one-time R&D charge in Q2 2026), the Agenus BOT/BAL agreement (May 2024, 2.625% royalty), the Athira Pharma private placement (December 2025), and participation in the July 2026 Agenus financing led by Commodore Capital.

Change in accounting policies? No policy change, but a material change in mix with accounting consequences. The CFO flagged that XOMA “increases the number of financial royalty assets within our portfolio,” and that for those assets “cash received upon achievement of a milestone may not be recognized entirely as revenue in the period received. A portion may instead reduce the carrying value of the underlying asset.” Interpretation: reported revenue will become a progressively worse proxy for portfolio cash economics, and the divergence between GAAP and adjusted presentation will widen. Management has committed to more transparency at the December 2026 Investor Day; that commitment should be held to.

Recent changes — new markets, facilities, management? No management changes at the executive level. Headcount fell from 68 to 47 across Q3–Q4 2025 following the Pelthos spin-out — a genuine simplification, not a cut under stress. Portfolio management was formalised under Lauren Hay (VP, Portfolio Strategy and Investments), which management credits for absorbing XOMA’s 120+ assets “with a de minimis amount of absorption of their existing infrastructure.” Geographic expansion is via partners’ launches (Ojemda in Europe with Ipsen, Vabysmo globally with Roche) rather than Ligand’s own footprint. Shareholders approved a 600,000-share increase to the equity plan on 2026-06-05, and the Amended & Restated Credit Agreement of 2026-07-14 replaced the 2023 facility with a $125M revolver maturing 2028-09-12.

A note on news coverage. The ROIC.ai news feed returned no articles for LGND across the period from 2026-03-01. The recent-events record above was therefore constructed entirely from primary sources — the mirrored SEC 8-K corpus, the Schedule 13D, the Q2 2026 10-Q, the FY2025 10-K, the 2026 proxy and the Q2 2026 earnings call. That is the stronger evidentiary basis in any case.

APPENDIX B — Source Appendix

Ligand Pharmaceuticals Incorporated (NASDAQ: LGND) — 2026-08-15

All sources accessed 2026-08-15 unless otherwise noted. Primary sources are listed first. Every material claim in this article and in Appendix A traces to an entry below.

1. SEC Filings — Primary

The trailing 60-month corpus (since 2021-08-01) was enumerated and mirrored in full: 448 filings — 215 Form 4, 67 8-K, 60 Form 144, 26 SC 13D/G, 16 10-Q, 13 SCHEDULE 13G/D, 10 DEFA14A, 6 Form 425, 6 Form 3, 5 S-8, 5 DEF 14A, 5 10-K, 4 ARS, 2 S-3ASR, 2 Form 4/A, 1 8-K/A, 1 Form 5, 1 CT ORDER. CIK 0000886163.

Corpus retrieval: the full filing set was enumerated and downloaded via the SEC EDGAR submissions and archives APIs (public, no authentication required).

2. Management Commentary — Primary

Retrieved via the ROIC.ai MCP get_latest_earnings_call (NASDAQ:LGND); the underlying event is Ligand’s publicly webcast Q2 2026 earnings call, archived in the Investors section of https://www.ligand.com/.

Treatment. All management commentary is treated as hypothesis rather than evidence and was validated against the filings above. Where management’s framing and the filings diverge — the “0% coupon” characterisation, the “operating cash flow” guidance basis, and the FY2025 revenue growth rate — the filing governs and the divergence is flagged explicitly in this article.

3. Quantitative Data Sources — Third-Party Aggregated (not primary)

Authority and reconciliation. ROIC.ai, AZI and FactorsToday are third-party aggregated data and statistical estimates, not primary sources. For this US filer, SEC EDGAR and the 10-K/10-Q remain primary; every material number driving a verdict in this article was reconciled to a filing. FactorsToday loadings and risk statistics are reported as facts; any statement about persistence or mean reversion is labelled interpretation and regime-caveated.

4. Peer and Comparative Data

5. Analytical Frameworks

6. Data-Quality Notes and Caveats

1. Spin-off price artifact (verified and excluded). The largest apparent single-day move in the adjusted price series — +28.8% on 2022-11-02 — is not a price event. The data provider applies a 1.603 adjustment factor on that date for the OmniAb spin-off completed 2022-11-01; the unadjusted close moved $84.61 → $68.00, a real −19.6% separation. Verified against the unadjusted_close and split columns and excluded from the five-year event map as a genuine move.
2. ROIC.ai identifier format. ROIC.ai requires an exchange-qualified identifier (NASDAQ:LGND); the bare ticker returns an “Invalid identifier” error.
3. ROIC.ai news coverage. get_company_news returned an empty array for this issuer across the queried window. The recent-events timeline was built from the primary SEC corpus, the proxy and the earnings call instead.
4. FactorsToday return conventions. All *_return and Sharpe figures on the leaderboard endpoint are annualised, including short windows. The m3 figure of 2.6448 de-annualises to approximately +38.2%, cross-checked against the AZI price CSV (+34.5% over the comparable window); the magnitude is real, not a data error.
5. Non-GAAP reconciliation not published. Ligand does not publish a bridge from GAAP operating cash flow to the “core cash flow” metric disclosed in the proxy. The $166M FY2025 gap is decomposed in this article by inference from identifiable add-backs (stock-based compensation, royalty-asset amortisation, R&D-financing charges treated as investments) and is labelled interpretation accordingly.
6. Insider 10b5-1 status. Rule 10b5-1 plan status was not verified filing-by-filing for individual sales. The presence of 60 Form 144s in the corpus indicates a substantial share is plan-driven; the insider discussion is qualified accordingly and no individual sale is over-read.
7. XOMA purchase-price allocation unavailable. The allocation between goodwill, intangible royalty assets and financial royalty assets will not be published until the Q3 2026 10-Q. All XOMA accretion figures used are management’s guidance, labelled as such.
8. Convertible cost calculation. The ~3.1% all-in cost of the 2031 notes is the analyst’s arithmetic from disclosed figures ($700.0M gross, less $20.175M of discount and issuance costs, less $195.79M for the note hedge, plus $114.10M of warrant proceeds = $598.1M net cash against $700M repayable over 5.22 years) and is labelled interpretation. It excludes the value of equity upside surrendered above the $524.34 warrant strike.