Merit order
Based on Wikipedia: Merit order
The Hidden Auction That Sets Your Electric Bill
Every second of every day, an invisible auction determines what you pay for electricity. Power plants line up like bidders at a market, each offering to sell their electrons at a certain price. The cheapest offers win first. The most expensive win last—if they win at all.
This system is called the merit order, and understanding it explains everything from why your summer afternoon electric bill spikes to why wind farms are reshaping entire energy markets.
How the Electricity Auction Works
Imagine you're running a city's power grid. Right now, at this exact moment, people are turning on air conditioners, factories are firing up machinery, and someone just plugged in their electric car. You need to generate exactly the right amount of electricity to match this demand—not too much, not too little.
You have options. A nuclear plant can sell you power for about three cents per kilowatt-hour. A natural gas plant might charge six cents. An old diesel generator sitting idle downtown wants twelve cents. A solar farm will sell at essentially zero marginal cost—the sun is free, after all.
The merit order system lines up all these generators from cheapest to most expensive. You start buying from the cheapest source first. When that's not enough, you move to the next cheapest. You keep going up the ladder until you've purchased exactly enough electricity to meet demand.
This is called economic dispatch, and it's been the backbone of electricity markets for decades.
The Crucial Detail: Everyone Gets the Same Price
Here's where it gets interesting. In most electricity markets, every generator that gets dispatched receives the same price—the price offered by the last, most expensive generator needed to meet demand.
Think about that for a moment. If demand is low and only nuclear and some gas plants are running, everyone might get paid five cents per kilowatt-hour. But on a scorching August afternoon when every air conditioner in the city is running full blast, you might need to fire up that expensive diesel generator at twelve cents. Suddenly, everyone—including the nuclear plant that was happy to sell at three cents—gets paid twelve cents.
This is why peak demand drives electricity prices so dramatically. It's not that the average cost of electricity changed. It's that you needed one very expensive source, and that source set the price for everyone.
Baseload, Peaking, and the Dance Between Them
Power plants roughly fall into two categories based on their role in this auction.
Baseload plants run constantly. Nuclear reactors, large coal plants, and some hydroelectric dams fall into this category. They're expensive to build but cheap to operate. Starting and stopping them is slow and costly, so operators prefer to run them around the clock. They sit at the bottom of the merit order, always getting dispatched, always making money.
Peaking plants are the opposite. They're relatively cheap to build but expensive to run. Gas turbines and diesel generators can start up in minutes, making them perfect for those afternoon demand spikes. They might only run a few hundred hours per year, but they charge premium prices when they do.
Between these extremes sit combined-cycle gas plants and other mid-merit generators, running perhaps twelve hours a day, filling the gap between constant baseload and occasional peaks.
How Renewables Scrambled the Merit Order
Solar panels and wind turbines have a peculiar characteristic: once you've built them, generating electricity costs almost nothing. The sun doesn't send an invoice. The wind doesn't charge by the cubic meter. The only ongoing costs are maintenance—someone to clean the panels, someone to oil the turbine bearings.
This near-zero marginal cost means renewables jump straight to the front of the merit order queue.
The effects have been dramatic. A 2007 study by the Fraunhofer Institute for Systems and Innovation Research in Karlsruhe, Germany found that solar power was reducing wholesale electricity prices by ten percent on average. During sunny early afternoons, prices dropped by as much as forty percent. By 2006, German consumers were actually saving more money through lower wholesale prices than they were paying in renewable energy surcharges.
A follow-up study covering 2008 to 2012 quantified the effect more precisely. For every additional gigawatt-hour of renewable electricity fed into the German grid, wholesale prices fell by about 0.11 to 0.13 euro cents per kilowatt-hour. By 2012, the total price-suppressing effect of wind and solar exceeded one euro cent per kilowatt-hour.
Wind power alone was saving German consumers five billion euros annually.
The Duck Curve: When Timing Creates Problems
But here's a complication that keeps grid operators up at night.
Solar energy peaks around noon. Human activity—and electricity demand—peaks in the late afternoon when people come home from work, turn on their televisions, start cooking dinner, and crank up the air conditioning after a hot day.
When you plot electricity demand over a typical California day and subtract solar generation, you get a shape that looks remarkably like a duck. Low demand in the morning (the duck's tail), plunging demand at midday as solar floods the grid (the duck's belly), then a sharp spike in the late afternoon as solar fades but demand soars (the duck's neck and head).
This duck curve represents a fundamental challenge. During the belly of the duck, wholesale prices can collapse to zero or even go negative—there's so much solar that grid operators are essentially paying people to take electricity off their hands. But just a few hours later, during the neck, prices spike as expensive peaking plants scramble to meet demand that solar can no longer serve.
The mismatch between when renewables generate and when people consume means that zero marginal cost solar doesn't automatically translate to zero cost electricity. You still need something to fill the gap—either expensive peaking plants or expensive batteries.
The Mathematics Beneath the Market
Behind this seemingly simple auction lies sophisticated mathematics that would make a calculus professor sweat.
The fundamental problem is optimization: how do you minimize the total cost of meeting electricity demand while respecting physical constraints? You can't push more electricity through a power line than it can carry. You can't ramp a nuclear plant up and down like a gas turbine. You need to maintain system stability, which requires keeping voltage and frequency within tight tolerances.
Power engineers represent the grid as a network of buses—connection points where generators, loads, and transmission lines meet. At each bus, the electricity flowing in must equal the electricity flowing out, minus whatever is lost as heat in the transmission lines.
The optimization problem then becomes: find the generation level at each power plant that minimizes total cost while ensuring power flows balance at every bus and no transmission line exceeds its capacity.
This is hard. The equations are nonlinear. The variables interact in complex ways. A change in generation at one plant affects power flows throughout the entire network.
Grid operators use specialized software that solves these equations every few minutes, adjusting dispatch orders as demand rises and falls, as generators trip offline unexpectedly, as wind speeds change and clouds pass over solar farms.
Locational Marginal Prices: Geography Matters
Here's something that might surprise you: electricity in New York City doesn't cost the same as electricity in upstate New York, even at the same moment.
Transmission lines have limited capacity. If you're in a congested part of the grid—a city with high demand but limited transmission connecting it to distant power plants—local generators become more valuable. You can't import cheap power from elsewhere; those wires are already full. You need local generation, even if it's more expensive.
This creates locational marginal prices. The price of electricity varies from bus to bus depending on how congested the local transmission network is. In some markets, there are thousands of different prices at any given moment, one for each node in the grid.
These price differences create incentives. If electricity consistently costs more in a certain location, that's a signal to build generation there, or to build new transmission lines to bring in cheaper power from elsewhere.
Adding Pollution to the Equation
The traditional merit order optimizes for one thing: cost. But burning fossil fuels produces pollution, and society increasingly wants to factor that into dispatch decisions.
Environmental dispatch adds another dimension to the optimization problem. Now you're not just minimizing cost—you're minimizing some combination of cost and pollution. This gets complicated fast. A coal plant might be cheaper than a gas plant, but it produces more sulfur dioxide, nitrogen oxides, and carbon dioxide per kilowatt-hour. How do you trade off a dollar saved against a ton of carbon emitted?
Researchers have thrown every optimization technique they have at this problem. Particle swarm optimization, which mimics how flocks of birds find food. Neural networks trained on historical dispatch data. A modified version of the bees algorithm—yes, inspired by how bees search for nectar—that incorporates chaotic mathematical modeling.
One particularly clever tool is the Locational Emissions Estimation Methodology, developed at Wayne State University in Detroit. It links real-time electricity consumption to the resulting pollution by combining price signals from grid operators with emissions data from the Environmental Protection Agency. Originally built to optimize water pumping systems—pumping water takes enormous amounts of electricity—it's now used more broadly to help large electricity consumers shift their usage to times when the grid is cleaner.
The German Paradox
Germany offers a fascinating case study in merit order effects.
On one hand, renewable energy has dramatically reduced wholesale electricity prices. Wind and solar pushing expensive generators out of the merit order saves consumers billions of euros annually.
On the other hand, German retail electricity prices are among the highest in Europe. German households pay a surcharge—currently over fifty euros per megawatt-hour—specifically to fund renewable energy development through what's called the Renewable Energy Sources Act.
Both facts are true simultaneously. Renewables lower the wholesale price while raising the retail price through support payments. Whether consumers come out ahead depends on the balance between these effects, which has shifted over time as renewable capacity has grown and support payments have evolved.
There are also costs the merit order doesn't capture. Building new transmission lines to connect offshore wind farms to southern industrial centers. Developing storage to smooth out the variability of wind and solar. Trading electricity with neighboring countries to balance supply and demand. These costs show up elsewhere in consumers' bills, not in the marginal cost of generation.
Why This Matters for Your Electric Bill
Understanding the merit order helps explain phenomena you've probably noticed without understanding.
Why does electricity cost more in the afternoon than at night? Because afternoon demand is higher, pushing the market further up the merit order into more expensive generators.
Why do some utilities offer cheaper rates for running your washing machine at midnight? They're trying to shift demand to times when cheap baseload plants are underutilized and expensive peakers aren't needed.
Why do wholesale electricity prices occasionally go negative? When renewable generation exceeds demand and baseload plants that can't easily shut down keep running, there's literally more electricity than the grid can use. Prices go negative as generators pay consumers to take power off their hands.
Why are battery systems becoming so valuable? They can charge during the belly of the duck curve when prices are low or negative, then discharge during the neck when prices spike. They arbitrage the difference.
The Future of Merit Order
The merit order system was designed for a world of predictable fossil fuel plants. A world where generators had clear marginal costs—this much coal produces this much electricity—and could be ramped up or down on command.
The renewable energy transition is stress-testing this paradigm. When the sun shines and wind blows, marginal costs approach zero. But the sun doesn't always shine and the wind doesn't always blow. Intermittency requires either expensive storage, expensive backup generation, or expensive transmission to import power from wherever it happens to be sunny or windy at the moment.
Some economists argue we need entirely new market designs. Instead of just paying generators for energy produced, we might need to pay them for capacity—the ability to generate when needed—or for flexibility—the ability to ramp quickly up or down. We might need to value not just the electrons themselves but the reliability and predictability of their delivery.
Others argue the merit order system is adapting just fine. Storage is becoming cheap enough to smooth renewable variability. Smart grids are enabling demand to shift in response to price signals. The fundamental logic—dispatch the cheapest available resource first—remains sound even as the mix of resources changes.
What's certain is that the invisible auction continues every second of every day, generators lining up from cheapest to most expensive, electrons flowing from wherever they're produced to wherever they're needed. Understanding this auction is understanding the hidden machinery that keeps the lights on.