LCOE Is a Misleading Metric: Why Levelized Avoided Cost of Energy Is the Only Way to Value Dispatchable Power

When a logistics company purchases a fleet of delivery vehicles, evaluating them solely on their upfront sticker price guarantees a catastrophic misallocation of capital if the cheaper trucks cannot operate in the rain. The vehicles might cost less per unit, but the systemic failure to deliver goods on time destroys the business. The global energy transition is currently managing its multi-trillion-dollar capital allocation using an identical mathematical blind spot.

The Levelized Cost of Energy (LCOE) has served as the dominant metric for comparing power generation technologies across the industry. It measures the lifetime cost to produce a single megawatt-hour of electricity, dividing capital and operating expenses by total expected output. However, it entirely ignores whether the grid actually needs that power at the exact moment it is generated.[6]

This blind spot did not matter when grids were built entirely around dispatchable power. When comparing a coal plant to a combined-cycle gas turbine, both assets can be turned on and off at will by grid operators. Because their product is identical in its availability, comparing them purely on their production cost yields a mathematically sound decision.

The rapid deployment of intermittent renewables broke this assumption. Wind and solar generation have zero fuel costs, which structurally drives their LCOE down over time. By 2020, 62% of all new renewable power generation added globally boasted a lower LCOE than the cheapest new fossil fuel option.[6]

But wind and solar are non-dispatchable; they produce electricity when the weather dictates, not when demand peaks. The World Resources Institute notes that treating all electricity as equal fundamentally misuses the LCOE metric, because a megawatt-hour generated at noon on a sunny Sunday has a vastly different economic value than a megawatt-hour generated at 8:00 PM during a winter freeze.[1]

The strongest counter-argument from renewable developers is that LCOE remains the most transparent, standardized way to track manufacturing and deployment efficiency. It provides a pure baseline that strips away regional market distortions, allowing policymakers to see the undeniable cost curve improvements in solar panels and wind turbines.

Yet a baseline cost is useless if the product's value fluctuates wildly. The Electric Power Supply Association (EPSA) argues that relying on LCOE actively misleads policymakers by hiding the massive integration costs required to keep the lights on when intermittent sources drop offline.[2]

To fix this, the US Energy Information Administration (EIA) formally recommended in 2014 that non-dispatchable sources be evaluated using a different metric: the Levelized Avoided Cost of Energy (LACE). Rather than measuring what a plant costs to build, LACE measures what a plant is actually worth to the system.[6]

LACE calculates the financial cost that the grid would incur to provide the exact same electricity if the new project were never built. As Heaven Designs summarizes the distinction: "LCOE answers 'what does this energy cost to produce?' LACE answers the other half of the question: 'what is that energy worth, because of what it lets the grid avoid?'"[4]

These two metrics are designed to be used together to generate a value-cost ratio. If a utility evaluates a new solar project with an LCOE of $30 per megawatt-hour, that number alone means nothing. But if the energy and capacity that project displaces works out to a LACE of $42 per megawatt-hour, the project is economically viable because it delivers more value than it costs.[4]

The inverse is equally true, and it explains why dispatchable power sources like nuclear, geothermal, and natural gas often look uncompetitive on paper but remain vital in practice. A geothermal plant might have a high LCOE due to immense upfront drilling costs, but because it provides 24/7 firm capacity, it avoids the need to build expensive backup peaker plants or massive battery arrays.[5]

Consequently, the geothermal plant's LACE is exceptionally high. When LACE exceeds LCOE, the asset is a net benefit to the grid, regardless of how expensive its raw production cost appears in isolation. The ratio of LACE to LCOE must be greater than 1 for a project to be genuinely feasible.[6]

The Institute for Energy Research (IER) has actively campaigned against the standalone use of LCOE, arguing that it systematically undervalues the reliability of traditional baseload generation. When grid planners optimize solely for the lowest LCOE, they inadvertently force the system to absorb massive, uncounted costs elsewhere.[3]

This macro perspective is captured by a third metric: Total System Cost (TSC). TSC accounts for the lifetime cost of a technology alongside the financial implications of its impact on grid reliability, including the billions of dollars required for new transmission lines and utility-scale battery storage.[5]

The transition from asset-level accounting to system-level accounting is no longer optional. As grids push toward deep decarbonization, continuing to allocate capital based on the cheapest megawatt-hour rather than the most valuable one will mathematically guarantee a more fragile, expensive power network.[5]