By Aditya Sinha, The author writes on macroeconomic issues
In 1879, Thomas Edison lit a carbon filament bulb in his Menlo Park laboratory for 14.5 hours. By 1882, his Pearl Street station was supplying 85 buildings in lower Manhattan at a commercial tariff of $0.24 per kilowatt-hour (eyewatering by any standards), yet every customer paid it, because the alternative was candles and lamp oil. There were no state distribution companies in 1882. There was also no accumulated loss of Rs 7.08 lakh crore.
India’s electricity distribution sector has managed that figure (compiled by the Prayas Energy Group for the Sixteenth Finance Commission or FC) through state-owned utilities that have been bailed out in 2001, restructured in 2012, and bailed out again in 2015, emerging from each exercise approximately worse than before. The Electricity (Supply) Act of 1948, itself modelled on Britain’s Electricity (Supply) Act of 1926 and designed for an administrative structure that has since dissolved, created the vertically integrated state electricity boards that proved impossible to reform and difficult to kill. The Electricity Act of 2003 was supposed to change all this. It has now been 22 years.
Begin with the bailout record, because it is instructive. The 2001 settlement identified Rs 41,473 crore of dues for restructuring; states issued tax-free bonds and signed commitments to reform. The Financial Restructuring Plan of 2012 followed, covering short-term distribution company (discom) liabilities, with participating states bound to eliminate their cost-revenue gap and enact a Model State Electricity Distribution Responsibility Bill. The 16th FC notes, with considerable understatement, that none of the participating states met their improvement conditions, and none enacted the Model Bill. The Assurance Yojana (UDAY) arrived in 2015: Rs 2.32 lakh crore in bonds, 75% of discom liabilities absorbed by state governments, another set of mandatory targets. Accumulated losses at the time of UDAY stood at `3.78 lakh crore. They stand today at `7.08 lakh crore. Three bailouts, yet the number has nearly doubled.
Why does this keep recurring? Prayas offers an explanation. Between 44% and 86% of discom borrowing in eight major states goes not to capital investment but to working capital, paying generators for power already consumed, covering subsidy arrears that state governments owe but defer, and servicing old debt with new debt. Interest expenses alone account for roughly 5% of annual discom expenditure. Tamil Nadu’s discoms have accumulated `1.66 lakh crore in losses; in case of Rajasthan it is Rs 91,565 crore; and Rs 69,301 crore in Madhya Pradesh. These are not utilities in financial difficulty. They are utilities sustained by the implicit guarantee that the state government will eventually absorb whatever they cannot pay, which removes every incentive to pay.
The tariff question is where the analysis turns genuinely damning. Prayas modelled what would have happened had State Electricity Regulatory Commissions simply enforced inflation-linked tariff revisions of 3.9% annually, from FY16 onwards, rather than the erratic pattern of non-revision followed by shock correction that most states prefer. The answer was `4.41 lakh crore in revenue gap reduction over eight years, with Rs 1.6 lakh crore in interest savings. The political cost of 3.9% per year would have been a mildly unpopular annual announcement. The fiscal cost of avoiding it has been borne by every taxpayer who will eventually fund the next bailout.
What makes the Prayas report more than another diagnosis is the scale of the opportunity it documents. Battery energy storage costs have fallen from `11.25 lakh per megawatt per month (Kerala tender, April 2022) to `2.80 lakh per megawatt per month (January 2025), a 75% reduction in under three years, with no viability gap funding required in the more recent tenders. Solar combined with storage is now cost-competitive with discom industrial tariffs in 21 of 28 states. Agricultural feeder solarisation (aligning farm power supply with daylight hours, replacing subsidised grid electricity with solar) could save `77,558 crore annually across 12 states by FY32, with cumulative savings of `2.61 lakh crore between FY27 and FY32. Maharashtra contracted 18.7 gigawatts of agricultural solar in roughly nine months once it decided to move.
There is a further irony. Cross-subsidies from commercial and industrial consumers, the traditional mechanism by which discoms effectively taxed industry to fund cheap power for farmers, now account for less than 10% of discom revenues, because large consumers have been moving to captive generation and open access. The argument for protecting the cross-subsidy model no longer has a model to protect. The FC recommends privatisation as the long-term answer, proposing a special purpose vehicle to warehouse accumulated working capital debt, stripped out so a private buyer might consider what remains. (In fairness, the private discoms that already exist (operating under the same tariff environment) post aggregate technical and commercial losses of 12.12% against 16.37% for state-owned ones. State-sector power projects, meanwhile, average delays of 42 months in Uttar Pradesh and 87 months in Andhra Pradesh. Private projects are commissioned on time, or close to it.
Tacitus recorded that the Romans made a desolation and called it peace. India’s power sector has made a restructuring and called it reform. The recommendations are good. They were good in 2001, in 2012, and in 2015. At some point the question is not what needs to be done but who will bear the political cost of doing it, and whether we reach the answer before the fourth bailout.