Key Insights: Storage is Getting Quite Cheap – But Measuring its System-level Cost is Complex

India’s solar-plus-storage tenders have cleared at prices as low as ₹2.70–3.52 per kWh,  yet within months, a storage-backed tender with a stricter evening supply obligation cleared near ₹6.45 per kWh.

An ongoing study at CSEP aims to understand what causes the differences in bid prices seen, and what is a reasonable framework for comparing different storage bids.  The Key Insights presentation digs into the study details and initial findings (undergoing peer review), while a Blog walks through the main findings.

This analysis explains the gap: bid prices reflect what tenders oblige, not the cost of firm power. Using an hourly optimisation model built on India’s actual demand and solar profiles, we estimate truly firm supply, meeting demand in every time block at ₹8.30–9.07 per kWh, rising to ₹11.82 across bad resource years. One cannot simply take today’s bid prices and linearly scale them to 24×7 reliable power.

If this appears too stringent an ask, one can find cheaper power by only providing, say 95% of demand through solar+storage.  But which 5% of power is unserved? That matters measurably for the economics.  Two systems with a 95% annual capacity utilisation factor (aka plant load factor) cost ₹7.88 per kWh when firm in every block, but only ₹6.41 when the worst 5% of blocks are exempted (a “relax” case), an 18.7% discount that hands the worst time block shortfall back to the distribution company.  Serving those leftover is a portfolio cost passed on to the discom, who can either procure power from the power exchange (specifically at periods when prices are high) or rely on expensive peaker plants which will have a low annual utilisation. The “peakiness” of the unmet demand from the solar+storage system matters.  Grid based solutions would cost ₹8.28–8.79 per kWh at 2023 exchange prices, and measurably more from dedicated backup run at low utilisation (42.65% for (hourly) Firm tender design and 4.66% for Relax tender design (the latter being the norm for storage bids today).

A portfolio view accordingly shows the “cheap” flexible design costing more overall than its firm counterpart. The message is not against storage; India needs much more, both for grid stability and also scaling up renewable energy (RE). But when the race to L1 is run without the right cost signals, the gap surfaces as risk to the reliability and security of the power system. Procurement must therefore do two things at once: add resources at scale, and price firmness correctly as current cheap bids do not eliminate the cost of reliability, they move it.