Follow the grinding wheel, not the value pool
Indian component makers are being told to move into electric axles and battery systems because that is where an electric vehicle's value sits. I think most should invest where their process skill transfers, and pay only for options where it does not.
Most strategy presentations about electric vehicles that I have seen in the last five years contain the same slide. It divides the value of a petrol car's powertrain among its parts, does the same for an electric car, and shows the engine and gearbox shrinking while the battery, motor and power electronics swell. The conclusion is drawn for the audience: a supplier that wants to stay relevant must move towards the parts that grow. In India that slide has launched joint ventures, technology licences and new subsidiaries at component makers of every size.
I do not dispute the slide's arithmetic. I dispute its conclusion. A value pool tells you where the money in an industry will be. It does not tell you who will earn it, and for most component makers the honest answer is that someone else will.
The slide that shows where the money goes
Kavirath makes transmission gears, shafts and driveline assemblies. About 63 per cent of our revenue still comes from parts for petrol and diesel vehicles, and an electric car needs no multi-speed gearbox. In 2022 our strategy team brought the board a version of that slide, together with a ₹1,400 crore proposal to build complete electric axles, with motor, inverter and reduction gearbox in one housing, through a joint venture with a European motor specialist. The proposal was serious and carefully argued, and our executive chairman supported it. I argued against it, and I want to set out why, because the reasoning applies well beyond our company.
The slide leaves out three things. First, it says nothing about who owns the system. Indian vehicle makers have mostly chosen to keep the electric drivetrain's architecture in-house or with a few chosen partners, because it decides how their vehicles feel and what they cost. A supplier offering a complete axle is offering to take a decision the customer regards as its own. Second, the slide ignores competence. Motors and power electronics are disciplines with their own engineers, their own failure modes and their own suppliers of magnets and semiconductors. A gear company buying that competence through a joint venture is buying a dependence. Third, the slide hides the conflict with customers. If we sold complete axles, we would be competing with vehicle makers and their partners for the very programmes in which we hoped to supply them gears.
None of this makes the large value pool unreal. It makes it somebody else's. The question a component maker's board should ask is not where the value will be but which part of it the company has a right to win.
What a gear maker actually knows
Our right to win lies in a narrower but deeper place. An electric vehicle still needs a reduction gear between the motor and the wheels, usually a single or two-stage set, along with a rotor shaft and a housing for the motor. These parts are simpler in layout than a six-speed gearbox but harder in one important respect. In a petrol car the engine masks gear noise. In an electric car nothing does, and the input gear turns at the motor's speed, which can exceed 15,000 revolutions a minute. A whine no driver would have noticed in 2015 is now a warranty complaint.
That makes the electric reduction gear a precision problem: tooth geometry ground to microns, small corrections to the tooth profile designed with the customer, and noise testing on every unit at the end of the line. Those are skills a good gear maker has spent decades building, often without seeing them as strategic. When we looked at our own capabilities honestly, we found that our hard-finishing, measurement and noise-testing knowledge transferred almost entirely, while our knowledge of motors and inverters was close to nothing.
So we invested where the grinding wheel goes. Our ₹620 crore programme, approved in two stages, covers reduction gear sets, rotor shafts and aluminium motor housings. By March 2026 we had spent ₹540 crore of it. Electric-vehicle revenue rose from ₹38 crore in 2021–22 to ₹560 crore in 2025–26, from nine vehicle makers. The figure is modest beside the slide's value pool, and I would like it to be larger. But every rupee of it rests on a skill we can defend, and none of it puts us across the table from a customer.
Stages a board can hold management to
The second half of the argument concerns how to commit capital when the speed of the transition is uncertain. Indian electric two-wheeler volumes have moved with subsidy decisions, battery prices and the fortunes of new manufacturers. When the central government reduced its purchase subsidy for electric two-wheelers in mid-2023, two of our customers cut their schedules by a third for several months. A supplier that had built capacity on the slide's forecast would have spent that year explaining idle lines.
We use a simple rule. The board releases each stage of electric-vehicle investment only when customer nominations cover at least 60 per cent of its capacity. Nominations are not orders, and customers do revise them, but they are evidence that a vehicle maker has chosen our part for a specific programme. Below the 60 per cent line we spend on options instead: samples, joint development work and deposits on the long-lead machines we may need later.
The rule has costs. In 2024 we staged capacity too tightly at Hosur, and a shortage of grinding and noise-testing capacity delayed two launches, deferring ₹45 crore of revenue. Grinding machines take more than a year to arrive, and I should have ordered them before the second set of nominations was confirmed. We also declined a start-up vehicle maker's request for a complete axle, worth about ₹300 crore over its life, and that vehicle has since sold well. Return on capital in our electric lines was 11 per cent in 2025–26, below our 15 per cent hurdle, though rising as volumes build.
The rule also has a benefit that is easy to underestimate: it makes the board's role concrete. Directors cannot judge how fast India will electrify, and nobody should ask them to. They can judge whether management has the nominations it said it would have before it spends the money it asked for. That is a question a board can put to us every quarter.
I should be fair to the other view. Some Indian component makers have built real competence in motors or power electronics, through long investment or acquisition, and for them a move towards systems is natural. A few vehicle makers do want complete axles from suppliers, especially new entrants without engineering depth. And the gear in an electric drivetrain may become a commodity faster than I expect, leaving specialists like us with a smaller business than our plan assumes. Our board reviewed the question again in 2025 and concluded that we should remain a component specialist, while funding a small team to understand the systems our parts go into. I underfunded that team at first, and I regret it.
My advice to a component maker's board facing the slide is modest. Ask management which parts of the electric vehicle the company already knows how to make better than its customers could. Invest there in stages tied to customer evidence. Pay for options, not commitments, everywhere else. The value pool will be earned by someone. A supplier's task is to earn the part of it that nobody can take away.