Transmission Constraints and Congestion
Electricity markets usually behave as though power flows from the cheapest generator to whoever wants it. The physical grid does not work that way. Every line has a limit, power spreads across the network according to physics rather than contracts, and when the cheapest route fills up the grid operator has to pay someone else to generate instead. That is congestion, and in several countries it now costs more than any other part of running the system.
What a transmission limit actually is
Three different limits are usually described with the same word.
Thermal limits. Current heats a conductor, and a hot conductor sags towards whatever is beneath it. Each circuit therefore has a maximum current, which varies with air temperature and wind: the same line carries more on a cold windy day than on a still hot one. Some operators now use dynamic ratings that take live weather into account, which can release meaningful extra capacity from existing wires.
Voltage limits. Pushing power a long way over a heavily loaded network drags voltage down at the far end. Past a point, voltage collapses rather than gently sagging, so operators stop well short.
Stability limits. After a fault, generators at either end of a long corridor must stay in step with each other. On some routes the binding limit is not heat or voltage but how much power can flow before the system would lose synchronism following a fault.
On top of all three sits the security standard. Most grids are run so that the loss of any single component, called an N-1 contingency, causes no loss of supply. That means a corridor of two circuits is usually run at the capacity of one, because the remaining circuit must be able to take the whole flow if its neighbour trips.
Power does not follow the contract
If a trader in the north sells electricity to a buyer in the south, the power does not travel down a private path between them. It spreads across every parallel route in proportion to the electrical properties of each, in the same way water spreads through a network of pipes.
Two consequences follow. First, a transaction between two points can overload a line that neither party thought it was using. Second, trades inside one country can push large unscheduled flows through neighbouring countries, which is why Poland and Czechia installed equipment on their German borders to control flows they were not party to, and why European grid operators spend so much effort on coordinated calculation of cross border capacity.
Redispatch, curtailment and the bill
When the market’s outcome would overload the grid, the operator intervenes. The standard tool is redispatch: pay a generator on the constrained side to reduce output, and pay another on the other side to increase it. The market result stands financially, and the physical adjustment is settled separately.
Germany is the clearest example. Wind generation is concentrated in the north and offshore, heavy industrial demand sits in the south and west, and the north to south corridors have been slower to build than the wind farms. On windy days the operators routinely pay northern wind and coal plants to turn down and southern gas plants to turn up. Because Germany trades as a single bidding zone, the market price does not reflect this internal constraint at all, so the whole cost surfaces as redispatch charges rather than as a price difference between north and south. That design choice is one of the longest running arguments in European electricity policy, and is covered further in our note on bidding zones.
Where redispatch is not possible or would be more expensive, wind and solar are simply curtailed: told to stop producing energy that has no fuel cost, because the wires cannot carry it.
How markets price congestion
Systems handle the same physics in two quite different ways.
Nodal pricing. Most US markets calculate a price at every individual point on the network, typically thousands of them. When a line binds, prices separate: below the constraint they fall, sometimes to zero or negative, and above it they rise. The price difference is the congestion cost, made visible rather than socialised. Generators and buyers can hedge it with financial transmission rights. The US grid zones on this site are the balancing authorities that sit on top of these markets.
Zonal pricing. Europe divides itself into larger zones, usually whole countries, with one price inside each. Constraints between zones set how much can be traded; constraints inside a zone are handled out of market, through redispatch. The system is simpler for traders and worse at telling anyone where new generation or new demand would be useful.
Texas shows what happens when transmission catches up. Wind farms in the west of the state were heavily curtailed in the late 2000s because the lines east were full. A large programme of new transmission built in the early 2010s cut that curtailment sharply and let west Texas wind reach the cities. The generation had existed all along; the wires were what unlocked it.
Why this matters for reading the data
Congestion explains several things that otherwise look strange in electricity statistics.
A country can have surplus low carbon generation and still burn gas, because the surplus and the gas plant are in different places. Prices can be negative in one region and high in another on the same afternoon. A wind farm’s capacity factor can fall even in a windy year, because it was ordered to stop. And an interconnector can sit unused while prices differ across it, because the constraint is inland rather than at the border.
None of this is visible in an annual generation total. It is the main reason a grid can look, on paper, as though it has already solved a problem it is in fact still paying to work around.