LCOS vs Value

Levelized Cost of Storage (LCOS) tells you what it costs to store and discharge a kilowatt-hour. It does not tell you what that kilowatt-hour is worth. A battery with an LCOS of $150/MWh sounds expensive until you learn it discharges during hours when electricity is worth $250/MWh. The gap between cost and value, not cost alone, determines whether storage is economic.

LCOS is calculated similarly to LCOE: total lifetime costs divided by total lifetime discharge. A 100 MW / 400 MWh lithium-ion battery costing roughly $134 million at NREL's 2024 installed-cost benchmark (a conservative modeled estimate; BNEF's observed 2024 market prices run lower, around $165-220/kWh installed), with 4,000 cycle life, 85% round-trip efficiency, and $5M/year maintenance, has an LCOS of roughly $120-150/MWh. This exceeds the average wholesale price in every US market. By LCOS alone, the battery appears uneconomic.

Add value context. The battery discharges during evening peak when LMPs average $200/MWh in CAISO. It charges during midday when LMPs average $10/MWh. The value of its discharge: $200/MWh. The cost of its discharge: $150/MWh (LCOS including the charge cost). Net margin: $50/MWh.

Why is LCOS misleading when evaluated in isolation?

Because storage is not a generation technology; it is an arbitrage technology. Generators sell undifferentiated kilowatt-hours into a market. Storage buys cheap kilowatt-hours and sells expensive ones. Comparing LCOS to average wholesale prices is like comparing a trader's costs to the average stock price: irrelevant. What matters is the spread between buy and sell prices, not the level of either one.

---