Binding constraint
A transmission limit the grid has actually hit, forcing the market to reroute dispatch — with a shadow price measuring how costly the limit is.
The grid's dispatch is an optimization run under thousands of limits: line ratings, stability limits, contingency rules. Most of them sit quietly in the background with room to spare. A constraint 'binds' when the system actually runs into it — the flow on some monitored path reaches its limit, and the market must redispatch around it, running more expensive plants on one side and backing down cheaper ones on the other.
Every binding constraint carries a shadow price: what the system would save, per hour, if that limit could be relaxed by one megawatt. It is the market's price tag on the bottleneck. A small shadow price means the workaround is cheap; a large one means the constraint is forcing seriously expensive redispatch — and the largest ones are, in effect, a ranked list of where new transmission would pay for itself.
Binding constraints are where grid physics becomes market price. When prices split between two regions, there is always a specific named constraint doing it, and grid operators publish which ones bound and at what shadow price. For traders and asset operators, that list is the day's actual story: it says which lines shaped prices, and — through each location's exposure to each constraint — why any given node printed the number it did.
For the technical reader
In the SCED formulation, each monitored element/contingency pair contributes a flow constraint; a binding constraint has a nonzero Lagrange multiplier — the shadow price, in $/MW per hour, of its limit. Bus-level congestion components are recovered as −Σ_k (shift factor of bus i on constraint k) × μ_k, which is why one binding west-to-metro constraint simultaneously depresses upstream nodes and lifts downstream ones in proportion to their shift factors.
Shadow prices are typically capped by ISO parameters (constraint-specific maximum shadow prices in ERCOT), which bound how much redispatch cost the market will incur before accepting a violation — so an at-cap shadow price signals the constraint could not be fully resolved at the allowed cost. Analytically, recurring binding constraints plus their shadow-price duration curves are the standard basis-risk and transmission-planning diagnostic: they localize congestion cost to specific hardware.