Business

Know the Business — Onyx Biotec Limited

Onyx is a sub-scale Indian sterile-injectables contract manufacturer running two clean-room units in Solan, Himachal Pradesh — Unit I (Sterile Water for Injection ampoules, ~639k units/day) is the cash-generative base; Unit II (cephalosporin dry-powder injections and syrups, commissioned 2023, WHO-GMP certified May 2024) is what investors are actually buying. FY2026 is the make-or-break tell: revenue grew 12% to ₹69.5 Cr, but operating margin collapsed from 16.8% to 4.9% as Unit II's depreciation and audit cost stack arrived faster than its dry-powder volumes — the classic operating-leverage geometry of an under-utilised clean room. The market is treating this as a failed micro-cap pharma IPO trading near book value; the more honest read is that it is a small, levered bet on a single asset's utilisation curve, and either Unit II fills or this name does not work.

1. How This Business Actually Works

Onyx gets paid a per-unit conversion fee to fill and seal sterile pharmaceutical products that other people own the brand on. Two units, two economic regimes, one combined P&L.

Revenue is volume × unit-price, but profit is unit-price minus a fixed-cost stack that depreciates whether the line runs or not. SWFI at Unit I is a low-price, high-volume base load. Cephalosporin DPI at Unit II is the bet — higher ₹ per unit, but until throughput at Unit II crosses roughly 60-70% utilisation the line burns more cash on depreciation, interest, and QA wages than it generates in conversion margin. That is the entire stock thesis in one sentence.

The collapse is not a slow drift; it is the Unit II cost stack landing on a revenue base that has not yet caught up. H1FY26 OPM fell to 2.7% before recovering to 7.0% in H2. A return toward the FY24-FY25 17-19% band would mean Unit II is finally absorbing its fixed costs; another half year near 3-7% would confirm structural stress.

2. The Playing Field

Onyx sits at the bottom of the listed Indian CDMO ladder — about 1.1% of Akums' revenue, 0.2% of Gland's market cap, and the only loss-maker in the set.

The strongest margins are not from the largest players — they are from the focused ones. Caplin Point (₹2,187 Cr sales, 35% OPM) beats Gland (₹6,431 Cr, 25%) and Akums (₹4,359 Cr, 12%) because Caplin runs a focused LatAm/US export injectables platform on a relatively concentrated SKU set, while Akums runs everything-for-everyone at low conversion margins. The upgrade path for Onyx is not to become Akums — it is to become a smaller version of Caplin: focused sterile injectables, narrow product set, slowly upgraded to export-grade regulatory access. That is a 5-7 year story if it happens at all. J.B. Chemicals at 47x P/E is what the market pays for branded-pharma comfort, not for CDMO economics; comparing Onyx to JB on P/E is meaningless.

3. Is This Business Cyclical?

This is not a commodity cycle business. Underlying prescription demand for cephalosporins and SWFI is anchored to hospital admissions and chronic disease — it grows or stays flat, it rarely contracts. What does cycle, and what hit Onyx hard in FY2026, is the capacity wave: when several mid-sized CDMOs commission new lines simultaneously, fixed-cost burdens land on individual P&Ls before demand absorbs them.

Revenue did not collapse — it grew 12% in FY26. Operating margin collapsed regardless, because Unit II's depreciation, interest on the construction debt, and WHO-GMP QA staffing arrived in the income statement faster than dry-powder injection orders. This is the textbook signature of a CDMO in a fixed-cost lag: top line keeps growing while the bottom line breaks.

The live cycle is the capacity wave plus working-capital stretch, not the demand cycle. Debtor days climbing from 119 to 144 in a single year while operating margin halves and cash from operations turns negative (₹-1.49 Cr in FY26 vs ₹+1.42 Cr in FY25) is a near-classic mid-cycle CDMO squeeze. Onyx absorbed it via the IPO cash; another year of this without operating-margin recovery would consume that buffer.

4. The Metrics That Actually Matter

For this business, the income statement is a lagging indicator and the P/E ratio is misleading. Four operating signals tell you almost everything.

OPM and debtor days are already screaming. Customer concentration is improving (80%→51% top-5 share from FY22 to FY24 per the RHP is genuine derisking as Unit II added cephalosporin clients; FY25/FY26 share has not been separately disclosed). Utilisation is not disclosed but is the swing variable; a reader watching half-yearly results for any operational disclosure on Unit II throughput will know more than the income statement reveals.

One metric consciously not on this list: P/E. In its first operating-loss year after an IPO-driven capacity expansion, trailing P/E is undefined and forward P/E is a guess about whether Unit II fills.

5. What Is This Business Worth?

The right lens is price-to-book against a normalised return on equity at full utilisation, with a heavy haircut for execution risk and SME illiquidity. Trailing earnings are not investable. EV/EBITDA on a TTM basis is misleading because FY26 EBITDA reflects half-utilised Unit II. DCF on a 5-year forecast is precision theatre at this scale. The honest question is: at what normalised return on equity does the market underwrite this asset base, and how confident is the reader that Onyx ever earns that return?

At ₹32 the stock trades at roughly 1.05x book. For a clean-room asset that earned 12.16% ROCE in FY25 and 1.14% in FY26, the 1x book multiple is the market pricing a ~6-8% normalised through-cycle ROE — well below the 12-15% ROCE the company has shown it can earn when Unit II is not crushing the denominator. If the reader believes Unit II eventually runs at 70%+ utilisation and the franchise stabilises at FY25-like 16-17% OPM, the stock is too cheap; if the reader believes FY26 is the new normal and Unit II remains capacity-overbuilt, even book value is generous.

Sum-of-the-parts is not the right frame. Both units share a regulatory umbrella, a single auditor, a single management team, and overlapping working capital. They are one economic engine, not two. Valuing Unit I (a profitable cash cow) separately from Unit II (a depreciating drag) misreads the cost stack — shared QA, audit, HVAC and management overhead means neither unit's standalone economics exist.

6. What I'd Tell a Young Analyst

What the market is most likely getting wrong: this is being priced as a failed pharma IPO, not as an operating-leverage micro-cap inside a structurally fine industry. The premise that FY26's 4.9% OPM is steady state is testable, half by half, against the cost stack the company has already absorbed. The single piece of evidence that would change the view in either direction is H1FY27 OPM. Everything else on this page is anchored to that one number.

What the market may be getting right: there is no moat here. Onyx is one inspection away from a product-line halt, one stretched-payables cycle away from a working-capital squeeze, and one larger competitor's capacity decision away from sustained price pressure. Calling sub-scale WHO-GMP certification a moat is a beginner's mistake; it is a licence to operate, not a competitive advantage. The right framing is "this is a credible operating-leverage trade if and only if you believe Unit II fills" — not "this is a high-quality compounder you should hold forever."