A negative price means sellers pay to deliver electricity in that hour. It happens when output that is hard to turn down meets low demand, or when tax credits make producing worth it even below zero.
Also called negative LMP, negative pricing
Formula
Negative-price share = hours below $0 ÷ hours with a price
Hours below $0
Hours with a price under zero, counted at one node
Hours with a price
Hours with a valid price at that node
With real numbersSPP, 2025: 526,869 negative node-hours ÷ 2,811,960 priced node-hours at 321 plant nodes = 18.7%
In the data
18.7%of SPP plant node hours had a negative real-time price in 2025
SPP had the highest share of the seven markets. Four of them saw fewer negative hours in 2025 than in 2024.
Share of hours with a negative real-time price, plant nodes
ISO real-time prices, 2024 and 2025. Negative node hours over priced hours at plant pricing nodes with a full year.On the map
Built from InfraMap's 2025 data with the same tools the analyst uses.
Common mistakes
Dropping negative prices as bad data. They are real prices and belong in every average and count.
Calling every negative-price hour curtailment. Some plants keep running because tax credits or contracts still pay them.
Assuming a negative price means a loss. Revenue from credits, certificates and contracts sits on top of the market price.
How the InfraMap analyst defines it
A wholesale price below zero: in that interval sellers pay to deliver electricity. Output that is hard to reduce, shutdown costs and per-MWh incentives can make that worthwhile. Negative prices are valid observations, not data errors. Keep them in averages and counts. They do not by themselves prove curtailment or a loss, since credits and contracts can add revenue.
Negative-price share = intervals with price below $0 / intervals with a valid price
Reviewed October 26, 2026 against the references below. The analyst and the MCP server answer with this definition.
