Long-Term Thesis
Long-Term Thesis — Onyx Biotec Limited
1. Long-Term Thesis in One Page
This is not a long-duration compounder unless Onyx evolves from a single-site, paisa-priced sterile-water ampoule shop into a multi-format CDMO with at least one regulated-market accreditation (EU-GMP or USFDA) by FY2030. Without that evolution, the 5-to-10-year case is a tangible-asset floor plus optionality on Unit II utilisation — a ₹58 Cr Solan factory earning a negative spread over its ~12-14% cost of capital, too small to outcompete Innova Captab (₹1,630 Cr revenue, EU-GMP, 30 km away in Baddi) and too large to win on price against the ~1,000 unlisted Solan/Baddi SMEs that Schedule M is supposed to thin. The durable variable is whether the promoter team — running the plant since 2008, holding 65.10% through a 47.5% drawdown — ever signs the multi-crore capex cheque for a regulated-market dossier. The single highest-value multi-year signal is an EU-GMP or USFDA dossier filing announcement at Unit II; every other variable (margins, working capital, ROCE) is downstream of whether the company climbs the regulatory ladder or stays on the conversion-fee escalator.
The 5-to-10-year conclusion in one line. Onyx is best understood as a 20-year-old private factory that issued public shares in the IPO window — its decade-out fate is binary on a single capital-allocation decision the promoters have not yet signalled they will make. Asset-floor downside is real (₹39 Cr net fixed assets plus ₹52 Cr working capital against ₹58 Cr market cap); compounder upside requires a regulatory-and-capital pivot the company has so far avoided.
2. The 5-to-10-Year Underwriting Map
The map below converts each long-term driver into the underwritable form: what has to be true, what we see today, why it could last, and what would break it. The cells are deliberately specific so the next reader of this page in FY2028 can re-grade each row against new evidence.
The single driver that matters most is #4 — regulated-market accreditation. Every other row in this table is either a shared industry tailwind that benefits Innova and Akums equally (rows 1, 2), an operating-leverage call on already-built capacity (row 3), a behavioural-signal driver that is necessary but not sufficient (row 5), or a downstream consequence of the others (rows 6, 7). A 10-year compounder thesis on Onyx cannot survive without a regulated-market dossier — and that dossier is the one row in the table with the lowest confidence today. Read the rest of this page through that lens.
3. Compounding Path
The compounding question for a sub-scale CDMO is simpler than for a branded pharma: it is the product of three numbers — revenue per unit of installed capacity, fixed-cost absorption (utilisation), and capital efficiency on top of the existing asset base. The 5-year history is a stable-growth top line over which margins compounded and then collapsed. The 10-year forward path is one of three scenarios, each tied to a specific capital-allocation choice.
The chart is the compounding story in one frame. Revenue compounded at 11.6% CAGR — respectable but unspectacular for an SME pharma CMO and inadequate to escape the cost-of-capital trap on its own. Operating profit compounded at -12% CAGR over the same five years, because FY26 erased all the gains of FY24 and FY25 in a single year. Cash from operations is the line that matters: five-year cumulative CFO of ₹3.9 Cr against cumulative net income of ₹13.6 Cr means seventy paise of every reported rupee of profit never showed up in the bank account. A business compounds value only if reported earnings convert to spendable cash; Onyx has not yet demonstrated this conversion mechanism over a full cycle.
The ROCE arc tells the rest of the story: a two-year window (FY24, FY25) where Onyx briefly approached the peer-respectable 12-13% level — the IPO window — followed by a 1,170 bps collapse in a single year. Peer median is 15%; Indian SME pharma cost of capital is roughly 12-14%. Onyx has earned a positive spread over its cost of capital in exactly two of the last five years, and one of those was flattered by a 9× one-time spike in other income. The compounding base case is no compounding — the equity tracks book value plus inflation unless one of the scenarios below plays out.
The fair-value column applies a P/B multiple to each scenario's estimated FY2030 book value, calibrated to the operating reality of that scenario (Bear 0.75× book, Base 1.3× book, Bull 2.0× book). On current evidence — margin collapse, CFO negative, FII liquidation, IPO-narrative items going silent — the Bear case is the modal outcome rather than acceleration. The Base case requires only one thing (Unit II ramp without regulatory pivot) and is the consensus bull story; the Bull case requires a capital-allocation decision the promoters have not yet signalled. The 5-to-10-year IRR of holding this equity is governed by which scenario plays out; the probability-weighted IRR on this distribution sits in the mid-single-digit range — below the cost of equity capital, consistent with the stock trading at book.
4. Durability and Moat Tests
A 10-year compounder thesis must survive specific durability tests, not generic "execution risk" hand-waves. The five tests below are competitive, financial, operational and regulatory — each tied to a specific FY27-FY30 evidence threshold.
The pattern across the five tests is consistent: each requires a multi-year accumulation of evidence in a specific direction, and none of the five is currently trending positively. Test 2 (cash conversion) and Test 4 (customer concentration) are the two with the most directly observable evidence over the next 24 months. Test 3 (regulated-market dossier) is the one whose presence or absence by FY2028 most cleanly resolves the bull-vs-bear question — and is the test where management has been most silent.
5. Management and Capital Allocation Over a Cycle
The promoter-family team is the clearest positive in the long-term thesis and the clearest constraint on the most important capital-allocation decision. Sanjay Jain (MD) and Naresh Kumar (WTD) have run this single plant in this single building since 2008. Together they hold 45% of the equity; the broader promoter group holds 65.10% and has not sold a share through a 47.5% drawdown from the IPO price. There are no other listed directorships, no group holding-company stack, no sister concerns visibly being subsidised, no opaque trust structures, no offshore vehicles. By Indian SME standards, the alignment is honest.
What the same evidence cannot tell us is whether this team will make the one capital-allocation decision that matters for the 10-year thesis: filing a regulated-market dossier at Unit II. The historical record on capital allocation since IPO is binary in a different way — the mechanical commitments (debt repayment, Unit II WHO-GMP, promoter holding maintenance, no related-party transactions above 10% of turnover) have all been delivered. The strategic commitments (own-brand LVP launch, export build-out, margin sustainability, profit growth post-IPO) have either failed (exports, FY26 profit) or gone silent (LVP, cartoning line, margin trajectory). The own-brand LVP line — ₹6.07 Cr earmarked at IPO for an Onyx branded large-volume parenteral SKU and the single largest non-debt use of proceeds — does not appear in the FY25 annual report at all. That is the single most informative data point on this management's appetite for moat-building capex: they took the mechanical commitments and let the brand-building one go quiet within six months of listing.
The capital-allocation pattern over the next five years will be the strongest signal on whether this management runs Onyx as a private-asset compounder or as a listed-company growth case. The two are not the same. A private-asset compounder protects what works (Unit I cash flow, promoter dignity, the debt-free balance sheet) and refuses the risky capex. A listed-company growth case takes the multi-crore regulated-market bet at sub-book equity cost, accepts the FCF burn that follows, and earns the right to a 2× book multiple in FY2030. The current evidence is consistent with the first archetype, not the second — which is fine for a 65%-owned family business but inadequate to deliver compounder returns to public shareholders.
6. Failure Modes
These are real thesis-breakers, not generic execution risk. Each is observable in routine filings or competitor disclosures, and each has a specific early-warning signal that lets a future reader update the view before the headline arrives.
The two failure modes with the highest combined severity-and-probability are #1 (customer in-housing) and #3 (sub-book equity raise from working-capital strain). Both are first-order observable in routine filings within the next 18 months. Failure mode #2 (Innova Kathua) is largely out of Onyx's control — it depends on a competitor's capital-allocation choices, not Onyx's own. The lock-in expiry test (#5) is the most binary — it is dated November 2027 and the behavioural signal will be unambiguous when it arrives.
7. What To Watch Over Years, Not Just Quarters
These are the five multi-year milestones that would update — or break — the long-term thesis. Each is a real disclosure, not a guess, and each has a specific validation-and-refutation threshold.
The long-term thesis changes most if Onyx announces a multi-crore capex commitment to file an EU-GMP or USFDA dossier at Unit II — that single decision would move Onyx from a tangible-asset book-value name toward a real CDMO compounder candidate, and its absence through FY2028 is the strongest single refutation of any 10-year bull case.