The Merit Order Effect That Determines Which Power Plants Run First

When a city's air conditioners kick on during a summer heatwave, the electrical grid does not simply ask every available power plant to generate more current. Instead, grid operators rely on a strict economic hierarchy to decide exactly which turbine spins up next to meet the surging demand. This hierarchy—known as the merit order—is the invisible engine of deregulated wholesale electricity markets around the world. It dictates not only the physical dispatch of power across thousands of miles of high-voltage transmission lines, but also the wholesale price that every single generator receives for their electricity. By understanding this stack, observers can decode why power prices fluctuate and how the energy transition is fundamentally rewiring the economics of the grid.[7]

The merit order functions as a vast supply stack, rebuilt for every delivery period. It ranks every available generation unit by its short-run marginal cost, which is the expense required to produce one additional megawatt-hour (MWh) of electricity. Construction costs, debt service, and grid connection fees do not factor into this ranking at all. Because those capital expenses are already sunk, a power plant deciding whether to generate in a particular hour is only comparing its variable operating costs—primarily fuel and maintenance—against the price it can capture in the market.[1][2]

At the very bottom of this stack sit renewable energy sources like wind and solar. Because they require no fuel to operate, their marginal cost of production is practically zero. Nuclear power plants also sit near the bottom of the merit order for related but distinct reasons. While they do consume uranium, fuel is a small fraction of their overall cost base, and the technical penalties for cycling a reactor's output up and down are significant. As a result, nuclear plants generally run continuously and bid low into the market to ensure they are dispatched.[1]

As electricity demand rises throughout the day, the grid operator moves up the stack, dispatching progressively more expensive plants to maintain the delicate balance between supply and load. This sequence typically means calling on coal-fired plants first, followed by highly efficient combined-cycle natural gas plants, and finally, during extreme peak demand, less efficient gas peaking plants that sit idle for most of the year. The crucial mechanism of this market design is that the last, most expensive plant needed to meet total demand sets the clearing price for the entire market for that specific time interval.[1][2]

If a wind farm offers its output at €0 per MWh and a natural gas plant offers at €80 per MWh to meet the final increment of demand, both facilities receive €80 per MWh for their power. This "pay-as-clearing" model is highly intentional. "Since a generator's bid influences whether it operates rather than its earnings, the rational approach is to bid truthfully at marginal cost," notes energy analytics firm Montel. The difference between a plant's marginal cost and the final clearing price—called infra-marginal rent—is what enables generators to recover the massive capital expenses that the merit order itself ignores.[1]

The rapid deployment of wind and solar power over the last decade has introduced a profound disruption to this traditional system, a phenomenon widely known as the merit order effect. Each megawatt-hour of zero-marginal-cost renewable energy effectively pushes the entire supply curve to the right. By displacing the most expensive thermal plants at the top of the stack, renewables lower the cost of the marginal plant needed to meet demand. This dynamic systematically suppresses the wholesale clearing price for every generator on the grid, fundamentally altering the revenue models that legacy power plants have relied upon since the inception of deregulated markets.[3][5]

This dynamic explains why wholesale electricity prices in regions like California, Texas, and Germany can plummet significantly on sunny or windy days, even when global fossil fuel prices remain static. However, it also creates a structural challenge known as price cannibalization. As researchers in a 2025 MDPI study observed, "the increment of PV lowers the average market price during hours when PV is most productive, thereby reducing the market value and revenue that PV and other generators can capture." In Alberta's energy-only market, where offers can range from 0 to 999.99 Canadian dollars per MWh, the influx of renewables heavily influences these clearing prices.[3]

Furthermore, the merit order is not absolute; physics sometimes overrides economics. Grid operators like PJM in the United States use Security Constrained Economic Dispatch (SCED) to manage the physical realities of the transmission network. If a cheap generator is located behind a congested transmission line, the operator must skip it and dispatch a more expensive plant closer to the load—an action known as out-of-merit dispatch. From 2004 to 2007, PJM reported that implementing SCED reduced transaction curtailment requests in excess of 1,000,000 megawatt-hours, optimizing the grid's physical constraints while minimizing excess costs.[6]

As the electrical grid transitions, the merit order is being stretched at both ends. The bottom of the stack is increasingly crowded with zero-cost renewables, while the top is becoming more volatile, driven by extreme weather events and surging demand from data centers. A 2026 analysis by the Dallas Fed found that existing data centers have already increased wholesale prices by 2 to 6 percent nationwide, adding an estimated $1 to $5 per MWh to the nationwide wholesale price during the summer of 2022. If proposed construction proceeds under high-utilization scenarios, wholesale prices could rise by up to 50 percent by 2028, testing the limits of the merit order framework.[4]