Why India Wasted 4% of Its Solar Power While Running Short

For two months this summer, India threw away solar electricity at lunchtime and then ran short of electricity after dinner. Roughly 4% of the country's solar generation was curtailed in July and August, according to an analysis by the Centre for Research on Energy and Clean Air (CREA), during the same stretch in which the grid recorded supply shortfalls. That sounds like a contradiction. It is actually a timetable problem, and it is the same one waiting for every grid that builds solar faster than it builds the things solar needs.

What the numbers say

Fact: India has added solar at a remarkable pace. Installed solar capacity passed roughly 120 GW in 2025, according to the Ministry of New and Renewable Energy, up from under 40 GW in 2019. The government's target is 500 GW of non-fossil capacity by 2030, and solar is expected to carry most of that load.

Fact: Curtailment means a grid operator tells a working solar plant to produce less than it could. The panels are fine and the sun is out. The electricity simply has nowhere to go, either because demand at that moment is already met or because the wires between the plant and the customers are full.

Fact: India's demand peaks have shifted. All-India peak demand hit a record of about 250 GW on 30 May 2024, and the tightest hours increasingly fall in the evening, when air conditioners are still running and solar output has dropped to zero. Monsoon months complicate the picture further, with cloudy days, lower wind in some regions, and hydro output that is not always available when needed.

Why surplus and shortage happen on the same day

Interpretation: A daily electricity shortage and a midday surplus are not opposites. They are two ends of the same curve. Solar produces a large block of energy between roughly 10 a.m. and 4 p.m., and very little of it survives past sunset unless something stores it. A grid can be oversupplied at 1 p.m. and undersupplied at 9 p.m. without anyone making an arithmetic error.

The main culprit is coal, which still supplies around three quarters of India's electricity. Coal units cannot be switched off at noon and back on at 7 p.m. without cost and wear, so operators keep them running at a minimum stable output. Indian regulations have long treated about 55% of rated capacity as the technical minimum, and the Central Electricity Authority has been pushing plants toward 40% flexibility. Every megawatt of coal that cannot turn down at noon is a megawatt of solar that has to.

The second is transmission. Much of India's large-scale solar sits in Rajasthan and Gujarat, far from the biggest load centres, and new lines have lagged behind new panels. Some routes in Rajasthan were delayed for years by litigation over overhead lines in the habitat of the endangered Great Indian Bustard. A solar park cannot sell power through a line that does not exist yet, a problem familiar to anyone who has watched grid connections turn into a queue for the physical world.

India is not unusual, just early at scale

Fact: California has been curtailing solar for years. The state's grid operator, CAISO, reported more than 3 million MWh of curtailed wind and solar in 2024, most of it solar in spring, when demand is mild and panels are productive. Germany, Chile and parts of China have their own versions of the same story.

Interpretation: What makes India's case more uncomfortable is the coincidence with shortfalls. In California, curtailment mostly signals abundance. In India, it sits next to evening scarcity, which means the waste has a visible price: power that was generated for free at noon had to be replaced, or simply not served, a few hours later. A 4% loss rate is not catastrophic. It is, however, the kind of number that tends to grow if build rates stay lopsided.

The fix is boring and already partly ordered

Fact: The Indian government approved a viability gap funding scheme for 4,000 MWh of battery storage in 2024 and has kept adding to it since, with further viability gap funding for storage in the works. In early 2025 the Central Electricity Authority advised that new solar tenders include co-located storage equal to at least 10% of plant capacity, with two hours of duration.

Interpretation: That advisory is the most important line in this story. Two hours of storage on 10% of capacity will not turn a solar park into a round-the-clock plant, but it does exactly what the curtailment data asks for: it takes the top off the noon peak and moves it into the early evening ramp. Batteries are now cheap enough to do this job against conventional alternatives: four-hour batteries now beat gas peakers on cost, and India, with limited domestic gas, has fewer alternatives than the US.

Pumped hydro is the other half of the plan. India has a substantial pipeline of pumped storage projects, which suit multi-hour shifting, but they take years to build and face the same land and environmental friction as transmission lines. Batteries can be installed in months. The arithmetic of timing favours batteries now and pumped hydro later.

What to watch next

Prediction: Curtailment in India will probably rise before it falls. Solar capacity is being added faster than either storage or transmission, and the storage tenders awarded in 2024 and 2025 will take until roughly 2026 and 2027 to show up as operating assets. Expect higher curtailment figures during the spring and post-monsoon months, when demand is softer and panels are productive.

Prediction: Electricity prices will do some of the persuading. India's power exchanges already show wide gaps between cheap midday prices and expensive evening prices. As that spread widens, the case for storage shifts from policy mandate to plain commercial logic, which tends to be a more reliable motivator than an advisory letter.

Prediction, with less confidence: Coal flexibility will be the slowest piece to move. Retrofitting older units to run reliably at 40% load costs money, and plant operators are paid for availability, not for graciously stepping aside at noon. Without tariff changes that reward flexibility, coal will keep crowding out solar at midday.

The general lesson

India's solar waste is not evidence that it built too many panels. It is evidence that panels are the easiest part of the system to build, and that the expensive, slow parts (wires, storage, flexible backup) set the real pace. The broader mechanics are set out in the complete guide to energy storage and the grid.

The irony is that the cheapest electricity India produces is also the electricity it most often discards. Fixing that does not require a technological breakthrough. It requires batteries, transmission lines and regulators who can count to two hours past sunset.