The Levelized Cost of Energy, Explained

How economists compare very different power plants on a single number, and what that number leaves out.

A coal plant, a solar farm and a nuclear reactor have very different costs. One pays mostly for fuel over time, another pays almost everything up front, and a third takes a decade to build. To compare them fairly, economists use the levelized cost of energy, often shortened to LCOE. It is the average cost of producing each unit of electricity over a plant’s entire life, counting everything from construction to fuel to maintenance.

Building the number

Think of LCOE as a lifetime cost divided by lifetime output. On the cost side, you add the capital cost, which is the money spent to build the plant, plus the running costs each year, such as fuel, staff and repairs. On the output side, you add up all the electricity the plant is expected to produce over its life. Divide the first by the second, and you get a cost per megawatt hour, or per unit, that can be compared across technologies.

Two ingredients change the answer a great deal. The first is the capacity factor: the share of the time a plant actually produces at full output, averaged over a year. A plant that runs almost all the time spreads its building costs across much more electricity than one that runs only when the sun shines. The second is the discount rate, which reflects the fact that money today is worth more than money in the future. Future costs and future electricity are both given less weight than those in the present, and the higher the discount rate, the more a plant with large up-front costs is penalised.

Why capacity factor matters

Picture two plants that each cost 100 million dollars to build, ignoring fuel and interest to keep things simple. Plant A has a capacity of 100 megawatts and runs at a 25 percent capacity factor for 25 years, producing about 5.5 million megawatt hours in total. Plant B has the same capacity but runs at a 50 percent capacity factor, producing about 11 million megawatt hours. Plant A's building cost works out to roughly 18 dollars per megawatt hour, while Plant B's is roughly 9 dollars. Same price tag, but twice the output means half the cost per unit.

What LCOE tells us

LCOE is useful for tracking how technologies change over time and for rough comparisons. It shows, for example, that in many countries new solar and onshore wind now have a lower LCOE than new coal or gas plants. It also shows how sensitive each technology is to different risks: gas plants are sensitive to fuel prices, while solar, wind and nuclear are sensitive to interest rates and construction costs.

What LCOE leaves out

LCOE treats every unit of electricity as equally valuable, but it is not. A unit delivered on a hot evening peak is worth far more than one delivered on a mild, sunny afternoon when power is already plentiful. LCOE also ignores system costs, the extra costs a plant creates for the rest of the grid. Wind and solar may need storage, backup plants and new transmission lines. Large conventional plants need backup for when they break down unexpectedly. Finally, LCOE usually leaves out costs that fall on society rather than the owner, such as air pollution and carbon emissions, unless they are deliberately priced in.

Choosing a power system by the lowest LCOE alone

It is tempting to rank technologies by LCOE and simply build the cheapest. But a grid made only of the lowest-LCOE option may not keep the lights on at every hour. Planners have to consider when each plant produces, what backup it needs and what it costs the wider system. LCOE is a starting point for comparison, not the final answer.

Key takeaways

LCOE is the average lifetime cost of producing each unit of electricity from a plant.
It combines capital costs, running costs and expected lifetime output.
Capacity factor and the discount rate strongly affect the result.
LCOE does not capture when electricity is produced, or the system costs a plant creates.
Pollution and carbon costs are usually excluded unless they are priced in.