$69/MWh: What the cheapest-new-power headline leaves out of the bill
October 30, 2026 · Frisian News
Photo: Wikimedia Commons
Lazard's 2026 LCOE report puts utility-scale solar at $69/MWh, below natural gas at $83. But Lazard's own report states the metric excludes reliability, transmission and flexibility costs. A peer-reviewed study finds solar's system cost reaches approximately €565/MWh in a grid without integration infrastructure.
Utility-scale solar carries a levelized cost of $69 per megawatt-hour in the United States, according to Lazard's 2026 Levelized Cost of Energy+ report. Onshore wind sits at $68. Combined-cycle natural gas lands at $83. Those three numbers dominate coverage of the report and produce a reliable headline: renewables are the cheapest way to build new electricity generation.
Lazard's report describes LCOE as the average lifetime cost of building and operating a power plant, expressed as the cost of generating one megawatt-hour of electricity. The same document states that LCOE "doesn't capture every aspect of operating a power system, such as reliability, transmission, or flexibility." That sentence appears in the same report that produced the headline. It rarely appears in the headline.
A power plant's cost and a power system's cost are different things. Solar and wind generate only when the sun shines or the wind blows. Covering the hours when they do not requires backup capacity, storage, or grid-balancing. None of those costs enter the LCOE calculation. Lazard acknowledges this directly, noting that meeting growing demand will require "expanded transmission infrastructure," "shorter interconnection queues," and "continued investment in energy storage," costs that sit outside the metric the industry uses most.
A peer-reviewed study published in the journal Energy, led by Henrik Lund of Aalborg University and co-authored by 10 other researchers, attempts to measure those system costs. It introduces a metric called SLCOE, system-based levelized cost of energy, which adds grid-balancing and storage costs to the basic figure. In Denmark's current electricity-only grid, solar's SLCOE reaches approximately €565 per megawatt-hour, the study finds. The authors argue that figure reflects the cost of any single technology supplying a grid without the flexibility of a fully integrated energy system, not an inherent property of solar panels.
In a modeled future climate-neutral system with full sector coupling (thermal storage, hydrogen via electrolysis, electric-vehicle smart charging and flexible heat-pump operation), the picture changes. A least-cost mix of offshore wind and solar reaches approximately €46 per megawatt-hour in that scenario, against approximately €100 per megawatt-hour for nuclear. The authors note those results are partly Denmark-specific, given its wind-dominant resource base and existing flexibility infrastructure.
The catch is the coupling infrastructure itself. That €46 figure assumes the full integration system is already in place. Christian Breyer, a professor of solar economy at LUT University in Finland and a co-author of the study, told pv magazine that the metric addresses a fundamental gap: "If one only optimizes within the electricity sector, one will not be able to identify these much better solutions."
The 2026 LCOE+ report notes that "continuous upward revisions to demand projections have driven a sharp increase in announced new-build gas generation despite a 15-year high LCOE and historically long development lead times." Grid operators, in other words, are not simply following the LCOE ranking when they make capacity decisions. Lazard also observes that renewable energy is "predominantly capital cost-driven while conventional technologies carry higher fuel and variable cost components," reinforcing "the need for a diverse generation fleet."
The LCOE figures from Lazard are real, and the long-term cost declines they record are real: solar has fallen from $359 per megawatt-hour in 2009 to $69 in 2026, an 81 percent drop. The metric answers one question precisely: what does it cost to build and operate this type of plant? It does not answer the question governments and grid operators actually face: what does it cost to keep the lights on?
That question is harder, and no single chart answers it yet.
Published October 30, 2026 · Frisian News