There is a particular genius to the way India’s power sector manages risk. It does not eliminate it. Nor does it absorb the risk. It simply moves it systematically, backed by decades of policy and contract.

By the time risk completes its journey through the system, it has passed through half a dozen hands, changed shape several times, and arrived at its final destination: the electricity bill of a household in Nagpur or the unpaid dues of a small factory in Coimbatore.

India's power generation capacity has grown from nearly 250 GW in March 2014 to 515 GW by late 2025. Per capita electricity consumption has risen from 960 units a year in 2013-14 to 1,460 units in 2024-25. Enormous sums have been invested. Yet the question of who pays when things go wrong has never been honestly answered. Follow the money, and the answer is almost always the same person.

How Costs Get Passed Along

The standard tool for passing costs is the ‘pass-through’ clause, buried in long-term supply contracts between power plants and electricity companies. When coal prices rise, generators pass the extra cost to distribution companies. Distribution companies pass it to consumers through higher bills.

These contracts typically run for 20 to 25 years. By March 2018, around 72% of India's power capacity, 620 plants generating 197 GW, was locked into such arrangements, the bulk of it coal-fired. By 2020, the share had risen to roughly 78.5%, or 291 GW, and nearly nine in ten units of power that distribution companies buy today come through such contracts.

After a decade of pause, new long-term coal contracts have revived: Maharashtra, Madhya Pradesh and West Bengal awarded 4.5 GW of new coal projects in 2024-25 at tariffs above Rs 5 a unit, even as recent renewable-plus-storage contracts have cleared at Rs 4.37. A procurement decision made in 2025 will ride on consumers' bills until 2050. Maharashtra and Tamil Nadu have also contracted over 40% above peak demand. They thus pay twice: once for the power they use, once for the power they over-bought.

The logic is understandable: power plants should not be left exposed to commodity price swings they cannot control. The problem is that this same logic, applied all the way down the chain, means every risk eventually lands on the consumer. At the same time, electricity companies have been allowed for decades to recover their costs regardless of how efficiently they operate. A distribution company that wastes 15% of the electricity it carries bills consumers for those losses rather than fixing them.

State governments now pay 98.9% of the subsidies they owe, up from 97% the year before, but the arrears were cleared largely by fresh 10-year loans from two public lenders at 10 to 11% interest. Distribution companies posted their first-ever combined profit of Rs 2,701 crore in 2024-25, but accumulated losses still sit at Rs 6.47 lakh crore. The numbers look better. The bills do not.

The distribution company sits at the tightest point in this funnel. It buys expensive power at the top and tries to sell it cheaply at the bottom. When its finances finally collapse, state governments absorb the debt. Taxpayers, who are also electricity consumers, pay for that.

The Bill That Was Never Sent

When electricity companies cannot recover their full costs through prices, the shortfall is not written off. It is recorded as a debt that consumers will repay later, with interest. These deferred amounts have been accumulating for years.

Rating agency ICRA puts these deferred costs at close to Rs 3 lakh crore across seven major state-owned distribution companies, with Tamil Nadu, Uttar Pradesh and Rajasthan accounting for the bulk. This sits on top of Rs 7.26 lakh crore in distribution-company borrowings as of March 2025, of which Rs 2.74 lakh crore has been disallowed by regulators as unsustainable, compounding at 10 to 11% a year. None of it has been written off. It is a bill waiting to be sent.

The Supreme Court stepped in with its 6 August 2025 judgment in BSES Rajdhani Power Ltd v. Union of India; it directed electricity regulators to clear these deferred costs within four years and capping the creation of new ones at 3% of a company's yearly revenue. In Delhi, where tariffs were not formally revised for nearly a decade, the three private distribution companies are owed over Rs 27,200 crore, of which Rs 21,413 crore is the recoverable share of the two BSES companies and Rs 5,788 crore is owed to Tata Power Delhi Distribution.

Paying this back within the Court's seven-year window means recovering Rs 16,580 crore a year, or roughly Rs 5.50 extra on every unit consumed across the city. The consumer did not vote to freeze prices for a decade. They will pay the cost of that decision regardless.

The same pattern plays out in generation. India built far more power plants than it needed during the 2010s. The share of installed capacity actually in use fell from a peak of 85% in 2009 to just 42% during the Covid disruption of April 2020, recovering to around 69% in early 2026.

Yet distribution companies kept paying the fixed charges on plants that sat idle. A study found that the existing long-term contracts inflate total system costs by 26% compared to running the cheapest plants available at any given hour. In Punjab alone, the annual fixed bill for just two private power plants, Rajpura and Talwandi Sabo, totalling 3,380 MW exceeded Rs 4,000 crore during periods when the plants sat largely idle. Those charges sat inside the procurement cost. The procurement cost sat inside the tariff.

Three Ways the Consumer Pays

The most visible route is the tariff hike — sudden, sharp, and usually long overdue. Delhi is the clearest recent case: a decade of frozen prices followed by a steep reckoning.

The same story is playing out across Rajasthan, Tamil Nadu, Uttar Pradesh, Maharashtra, West Bengal, and Karnataka — the states that together hold the bulk of India’s Rs 3 lakh crore in deferred costs. In West Bengal in 2020-21, deferred amounts equalled 15% of the distribution company’s expected revenue for the year. In Delhi, it was 12%. These are debts that real people will repay through future bills.

The second route is less visible: the hidden tax that industrial users pay to subsidise cheaper rates for homes and farms. Large factories and businesses have been paying rates around 50% above what efficiency would justify.

As they find ways to buy power elsewhere, the cross-subsidy shrinks, and the remaining residential consumers absorb the gap. NITI Aayog estimates that losses from theft, meter tampering, and poor collection alone account for nearly 20% of the sector’s total revenue shortfall. That too eventually finds its way into someone’s bill.

The third route is the one least counted but most personally felt: the money households and businesses spend to compensate for unreliable supply. India's inverter market was worth Rs 4,300 crore in 2023 and is growing at 8% a year, heading over Rs 7,300 crore by 2030.

The home UPS market, mostly lead-acid, was Rs 3,000 crore in 2023 and is projected to reach Rs 4,700 crore by 2032. The diesel genset market, the backstop for hospitals, malls, factories and telecom towers, stood at Rs 11,100 crore in 2024 and is heading toward Rs 22,000 crore by 2032. Taken together, this is close to Rs 18,500 crore a year of private spending, after tax, to build the reliability the electricity system failed to provide.

Every inverter on a rooftop in Lucknow or Ludhiana, every diesel genset humming behind a Mumbai hospital, is a receipt for a risk the system passed on and never resolved.

What Discipline Looks Like

Proof exists that things can be done differently. Tata Power turned around Odisha’s electricity companies through better metering, billing, collection, and network maintenance. It shows what can happen when a distributor manages its losses instead of passing them forward. Before that was the more striking Akanksha Power case in the Khaira sub-division of NESCO, Odisha. In this 45,000-consumer rural area losses to leakage, theft and unbilled use were cut by 66% between 2010 and 2021, a rare national award-winning bright spot in a distribution-franchising experiment that has mostly disappointed.

But discipline requires accountability. A framework that lets companies recover costs regardless of performance insulates them from consequences and leaves consumers exposed to whatever those costs happen to be.

A procurement plan that has left 44 GW of awarded renewable capacity sitting idle with no buyer signed up is creating the next wave of deferred costs. In Rajasthan, Gujarat and Tamil Nadu, between 10 and 30% of available renewable power was being switched off the grid in late 2025 because the system could not absorb it.

The choice is not between a financially healthy power sector and one that treats consumers fairly. When costs are recovered honestly, waste is penalised, and investment follows need rather than incentive, bills over time are lower and more stable than the alternative, which defers, obscures, and eventually dumps the unpaid arithmetic on the household at the end of the wire.

The Rs 3 lakh crore sitting on electricity company balance sheets is large enough that it can no longer be quietly rolled forward. The cushion has limits. And when it is fully compressed, there is nothing below it to absorb the imminent heavy shock.