113 entrieslast append amendments 0register open

dcentralmind

A public register on decentralization: why power should move from the few to the many
← register
№ 47 · appended

Delhi approves zero-upfront 3 kW rooftop solar for 230,000 households, from a base of about 10,000 systems

Zero upfront cost is the point at which rooftop generation stops being a middle-class capital decision and becomes something closer to a service delivered to a household. Whether that is decentralising turns on one term none of the coverage states: whether title to the asset and the right to the exported units sit with the household or with a vendor recovering its cost from the export stream. If the household holds them, 230,000 Delhi families become generators; if it does not, they become sites, with public money paying the rent on their roofs. Two details point toward the household, since the consumer is paid for surplus and is offered continued maintenance through the discom after the vendor’s five years, but neither is a statement of title. The second live term is the export price. Paying surplus at the discom’s average power purchase cost values the household’s exports at wholesale, structurally below the retail tariff it avoids by self-consuming, which is defensible grid design and also a decision about how much of the value of distributed generation stays at the edge, taken in a cabinet note rather than a tariff proceeding. The third story is arithmetic: 230,000 systems by March 2027 is roughly 30,000 installations a month in a city that has managed about ten thousand in total since 2023, and the binding constraint is installers, surveyors and interconnection approvals rather than money.

Delhi will install a 3 kW rooftop system for free on 230,000 homes by March 2027. The real change is that the subsidy now arrives before the installation rather than five years after it. The target is the part the arithmetic struggles with.

On 1 September 2026 the Delhi cabinet, chaired by Chief Minister Rekha Gupta, approved a second amendment to the Delhi Solar Energy Policy, 2023. Domestic consumers whose average monthly consumption in 2025-26 was up to 400 units can have a rooftop system of up to 3 kW installed at zero upfront cost, subject to technical feasibility. Minister Ashish Sood presented the decision.

The money stacks in three layers. The Centre's PM Surya Ghar central financial assistance contributes up to Rs 78,000 for a 3 kW system, Delhi matches it with a capital subsidy of Rs 78,000, and a further Delhi top-up covers whatever gap remains between the two and the tendered system cost. SaurEnergy's analysis puts the scale of that gap in context with a May 2026 quote from South Delhi of roughly Rs 2 lakh for a 3 kW system against about Rs 1.08 lakh of subsidy then available after installation.

Around that sit the administrative changes, which are less quotable and probably more consequential. The incentive is paid upfront rather than after five years, which the Chief Minister identified as the fix for the scheme's weak uptake in Delhi. Vendors must operate and maintain the system free for five years, after which the consumer may continue maintenance through their distribution company for the system's 25-year life. Registration and application fees are waived for systems up to 10 kW. The residential net metering approval window is cut from 75 days to 25.

Households consuming under 200 units a month keep their existing full electricity subsidy, receive a generation-based incentive, and are paid for exported surplus at the distribution company's average power purchase cost.

The arithmetic

The stated target is 2.30 lakh households and 500 MW of additional rooftop capacity. Those two figures do not sit together comfortably. At 3 kW each, 230,000 systems is 690 MW of nameplate capacity, so either the 500 MW assumes an average system well below the 3 kW ceiling, or the two are separate milestones being reported as one. Coverage compounds the confusion on timing: most reports give March 2027 for the households, while at least one gives February 2027 for the 500 MW. [UNVERIFIED: whether these are one milestone or two.]

Either way the run rate is the problem. Delhi's existing base is about 10,070 installed systems, by India TV's account, or 10,073 by SaurEnergy's, accumulated since the 2023 policy began. Reaching 230,000 by March 2027 means roughly 30,000 installations a month in a city that has managed roughly ten thousand in total. SaurEnergy frames the same point as about 220,000 systems in six to seven months and argues the binding constraint is delivery capacity rather than financing, which is the right diagnosis: the amendment is a well-designed answer to a money problem in a market whose bottleneck is installers, surveyors and interconnection approvals.

For scale on the capacity side, Central Electricity Authority data as at 31 July 2026 gives Delhi 7,549.64 MW of installed generating capacity, of which 546.47 MW sits in the renewable basket. Adding 500 MW of rooftop would roughly double Delhi's renewable capacity inside seven months, in a city SaurEnergy accurately describes as a power importer whose own generation footprint is a rounding error against its consumption.

Why it matters

Zero upfront cost is the point at which rooftop generation stops being a middle-class capital decision and becomes something closer to a service delivered to a household. Whether that is decentralising depends entirely on one term that none of the coverage states: whether title to the asset, and the right to the exported units, sits with the household or with the vendor. If it sits with the household, 230,000 Delhi families become generators. If it sits with a vendor or a RESCO recovering its cost through the export stream, they become sites, and the roof has been leased with public money paying the rent.

Two details in the amendment point toward the household. The consumer is paid for surplus exports, and the consumer is offered the option of continuing maintenance through the distribution company once the vendor's five years lapse, which is an odd right to extend to someone who does not own the equipment. Suggestive is not the same as stated, and no source consulted says it plainly. [UNVERIFIED: who holds title to the system and the right to the exported units.]

The export price is the quiet term worth arguing about. Paying surplus at the distribution company's average power purchase cost prices the household's exports at what the discom pays for wholesale power, which is structurally below the retail tariff the household avoids by consuming its own generation. The amendment therefore points households hard toward self-consumption, which for a 3 kW system on a home that uses up to 400 units a month is largely what will happen anyway. That is defensible policy and arguably good grid design. It is also a decision about how much of the value of distributed generation stays at the edge, made in a cabinet note rather than in a tariff proceeding. [NEEDS DATA: average power purchase cost for BSES Rajdhani, BSES Yamuna and Tata Power-DDL for 2026-27, and the retail slab rate applying to a 400-unit household, so the gap can be stated rather than asserted.]

The DISCOM side

Delhi's three distribution licensees, BSES Rajdhani, BSES Yamuna and Tata Power-DDL, have to carry this. Cutting the net metering approval window from 75 days to 25 is a real commitment imposed on them, and it is the kind of process change that decides whether a subsidy converts into installed capacity, because approval latency rather than panel cost is what usually kills an Indian rooftop pipeline. None of them appears to have said publicly whether they can interconnect 500 MW of rooftop in the time available, or what it does to feeder-level planning in the residential colonies where a 400-unit household actually lives.

What is still unknown

  • Who owns the installed system and the exported units. The whole decentralisation reading of this story turns on it, and no source consulted states it.
  • The tender structure. Whether a single empanelled vendor serves a district, on what contract, and what happens to a household whose assigned vendor fails.
  • Whether the household completion target and the 500 MW target share a deadline, and whether it is February or March 2027.
  • What happens to a household's existing 200-unit free-power subsidy once its own roof is generating and its billed consumption falls.
  • Whether the three distribution companies have accepted the 25-day window and the interconnection volume, and said so on the record.
  • The installed base itself. India TV gives 10,070 systems and SaurEnergy 10,073, neither citing the PM Surya Ghar national dashboard directly. A search summary separately indicates Tata Power-DDL alone had energised over 10,000 rooftop installations by April 2026, which if accurate would mean the citywide figure is considerably higher or the two counts measure different things. [UNVERIFIED: not confirmed against a primary source.]
  • No Delhi government press release, cabinet note or notified amendment text was retrieved. Everything here rests on press reporting of a cabinet briefing. [NEEDS DATA: the notified second amendment to the Delhi Solar Energy Policy, 2023, and the tendered per-kW system cost the top-up is calculated against.]

Sources

Recomputes the content hash in your browser and compares it with the stamp.

Public comments

Loading…