MTAR Technologies — the year the guidance kept climbing

The State of Play

MTAR closed FY26 with record revenue of ₹876 crore, up 30%, and net profit of ₹94 crore — finally back near the FY23 peak after a two-year margin slump. The year inside that number was a rollercoaster: a tariff-disrupted September quarter where profit fell to ₹4 crore, followed by the two biggest quarters in the company’s history. Management’s response to each twist was to raise guidance — FY26 growth went from 25% to 30–35% mid-year, and FY27 guidance went from 50% in January to “80% plus or minus 5%” by May, at a 24% EBITDA margin. The market has paid up for the story: the stock trades at a P/E of 239, while promoters have trimmed from 39% to 30.4% and FIIs have nearly tripled their stake.

The Company

Founded in 1970 by the Reddy family (P.R., K.S.N. and P.J. Reddy) to serve the Indian government’s post-embargo engineering needs, MTAR is a Hyderabad precision-engineering house — seven units within a 4 km radius plus an export facility — that machines mission-critical, high-tolerance hardware: ball screws, roller screws, water-lubricated bearings, electro-mechanical actuators, and the guts of nuclear reactors and rocket engines. It is one of the top three suppliers to India’s civil nuclear, space and defence programs, with 35–40 years of NPCIL history.

But the business that actually moves the numbers is clean energy: MTAR builds “hot boxes” — the core fuel-cell assemblies — for Bloom Energy, the world’s only solid-oxide fuel-cell maker, whose product is having a moment powering AI data centers. Bloom-linked work is roughly 60–70% of revenue, which makes MTAR effectively a leveraged bet on one customer’s supercycle, with nuclear, aerospace, space (ISRO’s Vikas and cryogenic engines), and a new oil & gas vertical (Weatherford) stacked behind it. Exports are now over 70% of revenue.

The financial arc: revenue quadrupled from ₹214 crore (FY20) to ₹876 crore (FY26), but profit took a detour — FY23’s ₹103 crore peak halved through FY24–FY25 (Bloom platform transitions, a deferred ISRO engine) before FY26’s recovery. Dividends stopped in FY23 and haven’t returned. Debt doubled in FY26 to ₹370–377 crore funding expansion. Promoter holding has fallen 16 percentage points in three years while FIIs went from 6.7% to 17.3% in twelve months — institutions concentrating the float as retail thinned from ~3 lakh to ~2.1 lakh shareholders.

The Story So Far

Q1 FY26 (call: August 6, 2025) — steady quarter, a ₹1,000 crore nuclear promise

The June quarter was solid: revenue ₹156.6 crore (+22%), EBITDA up 71%. Management reaffirmed the year’s frame — 25% revenue growth, 21% EBITDA margin ±100 bps — and laid out the claims the rest of the year would test. The biggest: roughly ₹1,000 crore of nuclear orders within 3–6 months, with Kaiga Units 5 & 6 (two new 700 MW reactors) to be “frozen by end-August/September” and refurbishment-reactor L1 status by September–October. Working capital, at a bloated 267 days, would come down to 200 by year-end. Bloom’s quarterly run-rate (~₹100 crore) would step up to ₹140–150 crore — in FY27. Aerospace would grow ~80% to ₹100–120 crore. A Weatherford oil-and-gas SEZ plant would be commissioned by June 2026. On US tariffs, the line was breezy: Bloom had given its highest-ever forecast after the tariff announcement, and the Taiwanese rival supplier was “not going to make any difference.”

Q2 FY26 (call: November 6, 2025) — the bad quarter, and a raised guidance anyway

Then the tariff confidence got tested. September-quarter revenue fell to ₹135.6 crore, EBITDA margin collapsed to 12.5%, and PAT to ₹4.2 crore — the second-worst quarter in three years. The explanation: a 3–3.5 week standoff with US customers over who absorbs the new tariffs (“if we absorb the tariff then we don’t stand a chance” — Srinivas Reddy held the line and dispatches stalled), plus deliberate inventory build for the second half.

The counterintuitive move came next: guidance went up, not down —

“A revised guidance of 30% to 35% increase in revenues for FY ‘26 compared to our initial guidance of 25%, which is driven by robust order inflows scheduled for execution within the fiscal year.” — Srinivas Reddy, Q2 FY26 call

— implying ₹900 crore for the year and an H2 running at double the H1. The order book backed the swagger: ₹930 crore at June had become ₹1,296 crore by September and ₹1,703 crore by the call date, with a year-end target of ₹2,800 crore. The Kaiga order (₹500 crore) was now promised “anytime during this month itself.” The Bloom capacity roadmap got specific: hot-box capacity from 8,000 units to 12,000 by March 2026, 20,000 by March 2027 — against the “hardly 3,500 a year” MTAR did historically. Quietly, the working-capital target slipped from 200 days to 220. And MTAR signed an EOI with Adani Aerospace (as 50% non-lead partner) for the AMCA fighter program.

Q3 FY26 (call: January 30, 2026) — the promised quarter arrives

The December quarter delivered exactly the inflection the November call had promised: record revenue of ₹278 crore (+59%), EBITDA of ₹64 crore at ~23% margin, PAT up 117%. Order inflows in the quarter were a remarkable ₹1,370 crore — about ₹645 crore of fuel-cell orders plus the Kaiga 5 & 6 order of ₹500 crore-plus, now actually in hand (a quarter later than the August telling, a month or two later than the November one). The order book stood at ₹2,394 crore. Bloom had signed a $2.65 billion fuel-cell agreement with utility AEP; the capacity roadmap stretched again — 12,000 boxes by March, 20,000 by December, then 30,000, with a new SEZ plant consolidating all Bloom work.

This is also where FY27 guidance was born: “we’re expecting growth of about 50% revenue growth for FY ‘27.” Asked to confirm — “5-0, 50%?” — Reddy answered, “That’s right.” The soft spots: nine-month EBITDA margin was 19.2% against the 21% full-year promise (the CFO leaned on a strong Q4), operating cash flow was negative ₹22 crore on swollen receivables, and working capital sat at 260 days against the original 200-day target.

Q4 FY26 (call: May 13, 2026) — records, a beat on cash, and 80%

The March quarter set another record — ₹306 crore of revenue, PAT of ₹44 crore (up ~222% YoY) — closing FY26 at ₹876 crore, up 30%: inside the raised 30–35% band, though notably not the original 25% plus much room. EBITDA margin came in at 19.5%, a miss against 21% ±100 bps, attributed to input-cost and freight inflation and expansion headcount. The cash story, though, flipped impressively: operating cash flow of ₹196.9 crore (double the prior year), and net working capital crashing from 278 days to 172 days — beating even the original 200-day target, on what management insists are structural, renegotiated payment terms rather than one-off advances. The order book ended at ₹2,580 crore versus the ₹2,800 crore target, with ~₹250 crore of nuclear/defence orders deferred a quarter.

And then the headline:

“We are raising our guidance for FY ‘27 from 50% revenue growth to 80% plus - revenue growth, plus/minus 5%, with clear EBITDA margins of around 24% for the year.” — Srinivas Reddy, Q4 FY26 call

That implies roughly ₹1,500–1,650 crore of FY27 revenue, ~70% from clean energy, with a ₹5,000 crore closing order book targeted (implying ~₹4,000 crore of inflows) and a “clear road map” to ₹5,000 crore of revenue by FY30. New threads: a first-article order from SLB for AI-data-center assemblies (₹35 crore now, ₹400–500 crore potential), nuclear order book at a record ₹650 crore-plus, the LCA Tejas actuator order (₹130–150 crore) expected in the current quarter, and the Weatherford plant now slated for full operation by September — its third date, after “June 2026” twice. One quiet retreat: Fluence, the battery-storage customer promised a long-term agreement “by Q4 FY26,” was dropped from the customer presentation — parked over battery duties, “not a closed door.”

The ledger: said vs. delivered

Kept: the raised 30–35% FY26 growth (delivered 30%); H2 at ~2x H1, exactly as promised in November; the working-capital turnaround (172 days vs. a 200-day original target that had been softened to 220 — a genuine beat, and the year’s most under-appreciated delivery); Kaiga ₹500 crore landed, one quarter late. Missed, narrowly: the 21% ±1% EBITDA margin (19.5%); the ₹2,800 crore order book (₹2,580, with ₹250 crore rolled forward). Slipped: the ₹1,000 crore of nuclear orders in 3–6 months became ~₹650 crore of nuclear order book by year-end with the rest “expected”; refurbishment orders kept moving right; Weatherford’s commissioning moved June → June → September; Fluence’s promised agreement became a parked prototype. The pattern to watch: guidance has only ever moved up — 25% → 30–35% → 50% → 80% — and each raise has so far been roughly vindicated by the next quarter’s prints, but the FY27 bar is now extraordinarily high, leaning ~70% on a single customer’s capacity blitz, with capex of ₹250–300 crore and debt already at ₹370 crore against a 0.5 debt-to-equity target.

Where Things Stand

FY27, by management’s own construction, needs to be the biggest year in MTAR’s history by a wide margin: ~80% growth at a 24% margin, clean energy at 70% of mix on Bloom’s “multifold” expansion (specifics now under NDA), nuclear scaling “in a big way” from Q2 as Kaiga and refurbishment work flow, IAI aerospace volumes from October, the oil & gas plant live by September, and the product division crossing ₹200 crore. The checkable near-term list is long: ₹250 crore of deferred orders and the ₹130–150 crore Tejas actuator order in Q1 FY27, Mahi Banswara/ASHVINI nuclear tenders during the year, the ₹5,000 crore order book by March 2027. The structural tension is unchanged: a genuinely scarce engineering franchise riding a real demand wave, financed by rising debt and a brutal (if improving) working-capital cycle, valued at a multiple that assumes the 80% year happens — and that the one after it does too.

The Four Checks

1. Quality and moat. A genuinely scarce engineering franchise with a real but narrow moat. The moat is qualification and learning curve: 35–40 years inside India’s civil nuclear program, one of the top three suppliers to NPCIL/ISRO/DRDO, and a 10–15-year Bloom Energy relationship where, in management’s words, “the learning curve is very steep” — a new vendor can’t just show up and make hot boxes or Calandrias. But it is a supplier’s moat, not an owner’s: MTAR is largely a build-to-print contract manufacturer, 60–70% of revenue hangs on one customer’s capacity blitz, and Bloom holds the design, the demand, and ultimately the pricing pen (management insists “there is no question of reduction in prices,” but that’s a negotiation, not a structure). Switching costs are real and durable in nuclear; in clean energy they are real but contestable over time. Call it a defensible niche position with concentrated fragility, not a fortress.

2. Returns on incremental capital and runway. Moderate returns, improving, with a genuinely long runway. Snapshot ROCE is 15.1% and ROE 12.5%, with the three-year ROE a poor 9.5% — the FY24–FY25 margin slump dragged returns down hard before FY26’s recovery lifted ROCE from 11.4% to 17.2% by management’s own count. So the engine currently earns mid-teens on capital, and it is a hungry engine: revenue quadrupled FY20→FY26 (₹214 crore to ₹876 crore) but needed doubled interest and depreciation, debt at ₹370–377 crore, and a working-capital cycle that even after the celebrated fix sits at 172 days. The runway is the attraction — a ₹2,580 crore order book, ₹250–300 crore of capex planned for FY27–28, and a ₹5,000 crore revenue roadmap by FY30 on ₹500–700 crore of cumulative capex. If FY27’s 24%-margin promise lands, incremental returns step up meaningfully; on the record to date, this is a 12–18% returns business with real room to deploy.

3. Capital allocation for the stage. Broadly rational for a build phase, with the FY26 evidence in its favour. Dividends were cut to zero from FY23 (they were 40–43% of profit in FY20–21) and every rupee plus borrowed money has gone into capacity — clean-energy expansion, the oil & gas SEZ plant, nuclear tooling. Expansion is being financed with debt rather than fresh equity, so shareholders haven’t been diluted, and the working-capital turnaround (278 days to 172, beating the original 200-day target on renegotiated terms management calls structural) plus ₹197 crore of operating cash flow shows the balance sheet discipline is more than talk. The quibbles: the FY24–25 capex years produced halved profits before the demand showed up, debt doubled in a single year against a 0.5 debt-to-equity target, and there is no buyback history to judge — though at this valuation a buyback would be the wrong move anyway. Separately, promoters have sold 16 percentage points of the company in three years; that’s their wallet, not the company’s, but it sits oddly beside an 80%-growth story.

4. Price. Demanding does not cover it. As of the June 2026 snapshot, the stock trades at ₹7,152 — a ₹21,973 crore market cap, a P/E of 227 on record FY26 earnings of ₹94 crore, and 27 times book value for a business earning 12.5% on that book. Even granting management the full FY27 promise — 80% growth at a 24% EBITDA margin, implying perhaps ₹190–220 crore of net profit — the stock would still be at roughly 100 times next year’s earnings. The price assumes the 80% year happens, the year after it happens too, and the mid-teens-return economics of the last six years permanently transform. That is priced for perfection on a single-customer supercycle.

Sources

• Concall transcripts (4): Q1 FY26 (Aug 6, 2025), Q2 FY26 (Nov 6, 2025), Q3 FY26 (Jan 30, 2026), Q4 FY26 + full-year (May 13, 2026) — BSE filings, converted to markdown.
• Annual reports (3): FY23, FY24, FY25 high-signal sections — FY24’s was the most useful (the stall-year explanation: Bloom platform transition + deferred semi-cryo engine); FY23’s was thin.
• Screener.in snapshot: consolidated quarterly and annual tables, ratios, shareholding — fetched 2026-06-05 (logged-out session).
• Research files: vault/Sources/Earnings/MTAR Technologies Ltd/ — raw transcripts, AR sections, snapshot, per-document digests (not published).
• Transcript quirk noted: the Q1 FY26 transcript’s “PBT of INR 114.8 crores” is almost certainly a transcription error (inconsistent with its own YoY growth figure and PAT); treated as ₹14.8 crore context.