Analysis only. This version carries the sector reading and the levels at which it would fail. It states no view on any security and recommends no action. Figures, sources and the documented defects are unchanged from the full document.
Sector Research
The order books are real. The unit economics are not.
Electrification has stopped being a forecast and become a backlog: it is sitting in tier-1 order books with delivery dates attached. What it has not done is prove it earns money at scale. The distance between those two facts is this sector’s decision, and the companies on either side of it are not the same companies.
Automobile and Auto Components As of 9 August 2026
| Figure | What it is |
|---|---|
| 24.57% | Sona Comstar EBIT margin, Jun 2026 |
| 22.0% | the level at which this reading fails |
| 257bp | what stands between them |
That is the argument in three numbers, and the third is the one to watch: it was 350bp a quarter ago, on the same consolidated series.
- Volume is at a record, and that is not the question
- Margins run from 32.40% to 0.75%, and who is improving cannot yet be stated
- The capital commitments say the participants expect this to persist
- The retirement lines are set in advance; four names pair a line with a print today, the nearest 123 basis points away
- The tier that holds the order books — settled on a margin, not on the books
Section 1 — Volume is at a record, and that is not the question
The cycle is at a record, and the record is broad enough that arguing about direction is arguing about the wrong thing.
Passenger-vehicle production reached a record 5.5 million units in FY26, up 9%, with exports up 17.5%. Underneath that, Ather Energy’s registrations grew 102% year on year in Q1 FY27. Apollo Tyres ran its India and Europe operations at 90% utilisation. Hero MotoCorp’s FY 2025-26 revenue was Rs 48,854 crore on a total-income basis, against Rs 41,968 crore the year before.
| Figure | What it is |
|---|---|
| 5.5 million | PV units produced, FY26 |
| +9% | on the prior year |
| +17.5% | PV exports, FY26 |
| +102% | Ather registrations, Q1 FY27 |
Volume is not the scarce thing here, and a reader looking for the argument in production numbers will not find it. Who converts that volume into margin is where the sector separates, and it separates further than most cycles allow.
Section 2 — Margins run from 32.40% to 0.75%, and who is improving cannot yet be stated
Levels, not changes — because for these companies the two are not equally knowable, and treating them as though they were is how this argument went wrong before.
Operating margins run from 32.40% at Eicher Motors to 0.75% at Ather Energy. Between them: 30.53% at SJS Enterprises, 24.57% at Sona Comstar, 21.75% at Sansera, 20.85% at Mahindra & Mahindra’s auto segment, 15.85% at Amara Raja, 15.75% at ZF Commercial Vehicle Control Systems, 15.20% at TVS Motor, 14.52% at Endurance, 12.70% at Apollo Tyres, 11.54% at Maruti Suzuki, and 7.73% at Motherson Sumi Wiring — that last one on a standalone basis, the others consolidated.
| Figure | What it is |
|---|---|
| 32.40% | Eicher Motors, Mar 2026 |
| 24.57% | Sona Comstar, Jun 2026 |
| 11.54% | Maruti Suzuki, Jun 2026 |
| 0.75% | Ather Energy, Jun 2026 |
Where those margins came from is a harder question than it looks, and on the evidence gathered for this sector it cannot yet be answered. The XBRL feed these filings were read from was retired in December 2024, and the earliest prints gathered here came from it on a standalone basis while every recent one is consolidated. A margin change measured from one to the other is measuring the change of source as much as the business — and for Sona Comstar the direction reverses entirely when both endpoints are taken on a consistent basis, a finding recorded in this sector’s review notes rather than in its evidence base.
Consistent series do exist further back than the gathered evidence reaches, which is the useful part: the spread is stated and the trends are withheld not because the data is unknowable but because the evidence assembled here does not yet carry it. That is a real loss for a reader who wants to know who is improving, and a smaller one than a confident direction that turns out to be an artefact of which filing happened to be quoted.
Section 3 — The capital commitments say the participants expect this to persist
Capital expenditure is the most credible statement a management team makes, because it is expensive to reverse. Read together, these say the participants expect the volume to persist — and they say where.
Eicher Motors approved INR 1,225 crores for a first-phase greenfield expansion at Tada, adding 450,000 motorcycles a year at full utilisation and completing during FY 2029-30. Endurance has begun adding 12 lakh ABS units a year to an existing 6.4 lakh, with start of production expected in September 2026, and is proceeding irrespective of the final regulation. Ather Energy is taking capacity from 4.2 lakh units to 9.2 lakh. Apollo Tyres is directing close to INR 3,000 crores of a INR 3,500 crores programme at Indian truck and car tyre capacity. Exide’s first cell phase is 6 gigawatt hours with provision to reach 12.
Two things follow. The money is going where section 1 found the volume — two-wheelers, electrification, safety content — which is a consistency rather than a coincidence. And the dates are FY28, FY 2029-30, September 2026: anyone who thinks this is a late-cycle top is disagreeing with the people who have to fund it, on the record, with board approvals attached.
But a commitment is a bet on demand that has been contracted, not demand that has been earned, and the difference between those two is what section 4 prices.
Section 4 — The retirement lines are set in advance; four names pair a line with a print today, the nearest 123 basis points away
For each name we state in advance the consolidated EBIT margin at which we would consider the argument for it retired. The cushion is the distance from the latest available print to that level.
| Name | Latest print | Argument retires | Cushion |
|---|---|---|---|
| Motherson Sumi Wiring | 7.73% Jun 2026, standalone | below 6.5%, standalone | 123bp |
| Maruti Suzuki | 11.54% Jun 2026 | below 10.0% | 154bp |
| Ather Energy | 0.75% Jun 2026 | above 3.0% | 225bp |
| Sona Comstar | 24.57% Jun 2026 | below 22.0% | 257bp |
Convention. An argument is retired only after two consecutive quarters through the level — one print starts the clock, two end it. Levels sit outside the range each business has traded in, so a breach means unusual behaviour rather than a normal quarter. The prints are the latest available for each name and are not all the same quarter, so the cushions are comparable as distances but not as a snapshot of one date. Note one difference from the machine-tracked lines these levels are drawn from: those trip on a single print by their stated date, while an argument here is retired only on the second consecutive one. The line is the same; the patience is not.
The table is short, and the shortness is the disclosure. Lines are set on many more names — Motherson below 8.0%, MRF below 15.0%, Studds below 17.0%, Exide above 13.0%, Bharat Forge above 19.0%, Bajaj Auto above 25.5%, Olectra below 13.0%, ASK Automotive below 10.0%, Banco below 16.0% — and on the macro conditions the whole sector runs against: the rupee at 96.5, Brent at 95.0, the ten-year at 7.2. Only the rows above pair a stated line with a current print on the same basis. Where the two do not meet, the level that would change our mind is stated and the distance to it is not, because estimating that distance is the error section 2 describes.
Note which direction these run. Ather’s line and Exide’s are upward: the argument there is that margin is constrained — by scale and by the lithium transition respectively — so the print that would prove us wrong is an expansion, not a contraction. Read those as the bull case’s trigger.
Section 5 — The tier that holds the order books — settled on a margin, not on the books
Tier-1 systems suppliers
The order books are the evidence and they are unusually specific: Rs 24,000 crore at Sona Comstar of which 64% is electric-vehicle content, and USD 96.0 billion of booked business at Motherson. That is multi-year revenue with delivery dates, in the tier earning 24.57% while the assemblers it supplies earn 11.54%. The risk sits in the same tier: copper inflation and wage pressure left Motherson Sumi Wiring at 7.73% on a standalone basis. And the cushion is thinner than it was — 25.50% in March, 24.57% in June, against a line at 22.0%.
The argument is retired if Sona Comstar’s consolidated EBIT margin prints below 22.0% for two consecutive quarters. That level is the same one carried as a machine-tracked line on this sector’s watch table, checked against the series as each quarter prints.
The qualification comes from the sector’s own record, and it is why order books are the beginning of the argument rather than the end of it. Ola Electric’s revenue fell 57% year on year in the December 2025 quarter and 58.24% in March 2026; Bharat Forge took a 51% revenue decline on North American truck destocking. Both had visibility too. An order book is contracted demand, and contracted demand is what a customer intends before circumstances change their mind — which is why the reading above rests on the tier holding the books, and is retired on a margin rather than on the books themselves.
One reading is carried above and no others are, and the absence is structural rather than a shortage of opinions. Readings that would have run the other way have so far failed the requirement that a bound line be able to refute the reading itself, so carrying only the survivors would report a filter as though it were a finding. Until that is fixed the watch table is the actionable surface, and it should be read as such rather than positions being inferred from the argument’s tone.
Three other things are deliberately not here. There is no valuation work: nothing in the evidence gathered for this sector carries multiples, so a view on price would be a view on nothing. There is no view on the passenger-vehicle assemblers beyond the margin level above — the case against them is a pricing-power argument that the volume figures do not settle either way, and stating it would be reasoning past the evidence. And there is no claim about where any of this sits against consensus: no broker estimate, target or positioning figure was gathered, so “the market expects” would be a sentence with nothing behind it. A reader looking for the differentiated-versus- consensus frame should know it is absent by construction, not by oversight.
Provenance. Written from the same evidence base as the sector’s data set. Two items in that set were withdrawn during review and nothing above rests on either.
Sources and conventions. Every figure above is drawn from a maintained evidence base and is reproducible from it. Margins are consolidated EBIT including other income unless a company basis is named, and a standalone print is labelled as one. Rupee amounts are stated in crore and dollar amounts in billions, as the reporting companies state them; where a source gives a figure without naming its currency, the currency is not supplied here. Data as of 9 August 2026. The reporting source changed in December 2024, so figures spanning that date are not stated on a single basis and no such comparison appears above. One finding in section 2 — that a named company’s margin direction reverses on a consistent basis — traces to this sector’s review notes rather than to its evidence base, and is marked as such where it appears; the consistent series enters the evidence base at the next full revision. The levels at which each argument is retired are stated in advance and tracked to their dates. Written without knowledge of any portfolio. Nothing here is a recommendation to buy or sell.