New Delhi [India], September 30: Every project I have built in several years has ended the same way. The client signs off, we raise the final bill, we demobilise, and the site becomes somebody else’s responsibility. Last year, at an approximately 9 MW plant at Phalodi in Jodhpur district, we finished the commissioning tests and then did none of that. We kept the keys. There was no final invoice to raise, because the plant is ours and the discom will pay us for its electricity, unit by unit, for the next twenty-five years.
That is a strange feeling for a contractor. It is also, I think, where a certain kind of Indian infrastructure company is heading, and it is worth being honest about why.
The push came from the discoms, not from us. Under PM-KUSUM’s feeder solarisation route a state utility can get daytime solar power for its agricultural feeders without spending a rupee of capital – it signs a power purchase agreement and someone else builds, funds and runs the plant. For a discom still working its aggregate technical and commercial losses down, that is an easy decision. National AT&C losses have fallen to 16.16% in FY25 from 22.32% four years earlier, and the utilities that got there did it by being careful with their balance sheets. So the capital requirement moved. It moved to us.
We took it because we could see what it does to the shape of our business. An EPC company lives tender to tender. You win, you execute, you look for the next one, and in a bad quarter there is nothing underneath you. Four plants – three at Phalodi under PM-KUSUM and one at IIT (ISM) Dhanbad under a SECI award, approximately 11 MW in all – give us a little over ₹ 6 Cr a year that arrives whether or not we win anything. Each sits in its own subsidiary, so the asset can be seen for what it is and not lost inside a contractor’s books. Against an order book of about ₹ 350 Cr, that ₹ 6 Cr is small. It is also the only part of our revenue that nobody can outbid us for.
And the opportunity does not stop with solar. Apart from PM-KUSUM, we are also bidding for new Battery Energy Storage System (BESS) tenders under the Capex model. The framework is similar to PM-KUSUM: each BESS plant is developed at our own cost and backed by a long-term Power Purchase Agreement (PPA) with the discom. That gives us another route to build owned infrastructure around a contracted, predictable revenue stream and creates the potential for stable returns over the life of the asset.
The second reason is that the two halves feed each other. The same crews, the same machinery and the same NABL-accredited testing lab that serve our discom contracts – close to ₹ 100 Cr of feeder segregation and household electrification work for the Jaipur and Jodhpur utilities under the Revamped Distribution Sector Scheme – also maintain our plants. A developer who has no presence in the district has to send a team. We already have one there.
Now the part I am less comfortable with. This model consumes capital in a way EPC does not. A RESCO plant pays back in about six years before financing cost, and eight to nine years after it. Our EPC contracts turn cash in roughly ninety days. Any company that lets the asset side run ahead of the contracting side will find it has swapped a lumpy business for an illiquid one, and there is no clause anywhere that lets you reprice a twenty-five-year tariff when inverters need replacing in year twelve. That is why we have said publicly that we will add one or two more plants over the next one to two years, and not a dozen. EPC remains the primary business. The plants are ballast, not the ship.
I would say the same to the discoms. The RESCO model works because a contractor is willing to carry risk the utility does not want. That willingness is finite, and it depends on tariffs that leave something behind after twenty-five years of operating cost, and on payment that arrives when it is due. Push the ceiling tariff too low in the bidding and the serious builders will stop turning up, which is not what a country adding 55.3 GW of non-fossil capacity in a single year needs.
There is a picture I keep coming back to from that morning at Phalodi. Two of our engineers were walking the rows with a meter, not because a client had asked them to, but because the reading was now our revenue. Nobody had to supervise them. That is the whole argument for owning what you build, and it took me several years to see it.
