Energy arbitrage is the least complicated idea in the storage business. Buy electricity when it is cheap. Hold it. Sell it when it is expensive. Keep the difference, minus whatever you lost on the way in and out.
What makes it interesting in India is that the country has accidentally engineered an unusually good version of that trade — by building an enormous amount of solar.
The markets, in one paragraph each
India’s power exchanges — the largest being the Indian Energy Exchange (IEX) — run several segments. The ones that matter for a battery are:
- Day-ahead market (DAM). Bids are placed today for delivery tomorrow, cleared in 15-minute blocks. This is the deepest segment and where most storage arbitrage has been done.
- Real-time market (RTM). Rolling sessions through the day, clearing close to actual delivery. Prices here move more violently than in the day-ahead market, which cuts both ways.
- Green day-ahead market (GDAM). The same day-ahead mechanism, restricted to renewable power. Buyer interest has been strong but the volume available is comparatively small, so liquidity is thinner.
A battery does not have to pick one. Much of the skill in running a merchant asset is deciding which session to buy in and which to sell into.
Why the spread exists
India has more solar than it can use in the middle of the day and none of it at seven in the evening, when demand peaks. That mismatch — the subject of our article on India’s evening peak gap — is what creates the spread a battery lives on.
How wide can it get? In April 2026, surplus solar drove real-time prices on IEX to near-zero levels while evening day-ahead prices approached the ₹10/kWh price cap. That is an extreme day rather than a typical one, but the direction is the everyday pattern.
Volumes are growing alongside. IEX traded 13,527 million units in July 2026, up 7.7% year on year, with higher demand lifting day-ahead prices.
What it has actually been worth
Indicative FY26 arbitrage values, drawn from Indian market analysis:
| System | Cycles assumed | Indicative arbitrage value |
|---|---|---|
| Two-hour battery | 545 per year | about ₹4.81 per kWh |
| Four-hour battery | 365 per year | about ₹4.63 per kWh |
Two things are worth reading out of that table.
First, the two-hour system earns slightly more per unit despite storing less. A shorter-duration battery can cycle more often — it captures the sharpest, narrowest part of the evening peak twice as frequently rather than spreading discharge across a longer, flatter window. Which shape suits your project is exactly the trade-off in our guide to two-hour versus four-hour duration.
Second, these are gross spreads before costs. Round-trip efficiency losses, exchange and transmission charges, and degradation all come off the top. A battery that is 88% efficient buys 100 units to sell 88, so the spread has to cover that shortfall before anything reaches the bottom line — see round-trip efficiency and degradation.
The risks nobody should skip
Spreads compress. This is the structural one. Arbitrage works by buying in the cheap window and selling in the expensive one, which by definition raises the low price and lowers the high one. Every battery that enters the market makes the trade slightly less profitable for the ones already there. Average day-ahead and real-time prices have already fallen year on year as supply liquidity improved.
Revenue is not contracted. There is no offtaker and no floor. A quiet quarter is simply a quiet quarter, which is why lenders fund merchant projects far more conservatively than contracted ones — the arithmetic is in our guide to how lenders look at storage.
Regulatory treatment is still settling. India’s merchant storage market is young — the first merchant BESS trades on IEX came from a single large operating asset — and the rules around scheduling, charges and market participation for storage are still being worked through.
Cycles are finite. Chasing every spread means more cycles, and cycles are what degradation guarantees are written against. The most profitable trade of the week can still be the wrong one if it eats a warranty.
What this means for you
- If you are a developer or IPP: model arbitrage as upside on top of a contracted floor, not as the base case. Run the numbers on compressed spreads as well as today’s, because today’s are the best they may ever be. Arbitrage is one layer of a broader revenue stack.
- If you are considering a merchant project: the operating capability matters as much as the hardware. Somebody has to bid the asset into the right session every single day, and that is a trading function, not a maintenance one. A standalone storage system is the easy half of the problem.
- If you are a C&I buyer: you do not need an exchange account to capture this. The same daily shape appears on your own bill as a time-of-day tariff, and a battery behind your meter arbitrages it with none of the market risk.
- If you want to see the numbers for your site: the savings calculator works the behind-the-meter case, and our team can walk through a grid-connected model with you — get in touch.
Market prices, price caps, exchange rules and storage participation regulations change by regulatory order and notification, and every figure above is a past market observation rather than a forecast. Treat this as an August 2026 snapshot and verify current market design and clearing prices before relying on them.