How has the cost of generating electricity changed over time?

New-generation costs changed unevenly: solar and wind plunged during the 2010s, then partly rebounded after 2021. Higher financing, equipment, tariff, and supply-chain costs drove that reversal. They remain far below 2009 levels; other technologies changed less uniformly. LCOE measures plant costs, not retail prices or full system needs. Evidence is strongest from 2009 onward and weighted toward U.S. projects. At higher variable-renewable shares, systems need more storage, transmission, flexible demand, or dispatchable backup.

Electricity generation became much cheaper—then partly rebounded

Main finding: The cost of new electricity generation has fallen sharply for solar and wind, but not for every technology or in a straight line. Most of the decline occurred during the 2010s. Since 2021–2022, higher financing, equipment, tariff, and supply-chain costs have reversed part of it.

The clearest comparison uses levelized cost of electricity, or LCOE. It spreads construction, financing, operating, and fuel costs across lifetime generation. LCOE is a plant-cost measure, not a retail electricity price. It excludes distribution and often omits transmission, balancing, reliability, and environmental costs.

The 2010s brought an exceptional decline

Global evidence shows that solar’s unit cost fell 85% and wind’s fell 55% between 2010 and 2019. Lithium-ion battery costs also fell 85%. Deployment grew more than tenfold for solar. The IPCC attributes these changes to research, demonstration funding, deployment support, learning, and scale (IPCC, 2022).

Lazard’s longer U.S.-focused series shows the same transformation. From 2009 to 2026, utility-scale solar LCOE declined 81%, while onshore wind declined 52%. In 2009, Lazard estimated solar at $323–$394/MWh and wind at $101–$169/MWh. By 2026, their ranges were $40–$98/MWh and $37–$99/MWh, respectively (Lazard, 2026).

Measured installation costs confirm that this was not merely a modeling result. U.S. utility-scale solar capital cost reached $1.61/W in 2024, down 73% from 2010 in real 2024 dollars. Larger projects, improved equipment, and near-universal single-axis tracking helped lower cost and increase output (Berkeley Lab, 2025).

Other technologies changed less uniformly. The IEA found offshore wind’s median expected LCOE fell from above $150/MWh in its 2015 edition to well below $100/MWh in 2020. It found lower expected new-nuclear costs than in 2015, but large regional differences. Extending existing nuclear plants was cheaper than building new low-carbon capacity. Gas costs fell mainly because fuel prices fell, illustrating their commodity exposure (IEA, 2020).

Costs have risen from their recent lows

The long-run decline masks a recent reversal. Lazard reports that wind LCOE rose 70% and solar rose 85% from their 2021 lows through 2026. It attributes the increase to higher capital costs, sustained interest rates, tariffs, and supply-chain repricing. Even so, both remained far below 2009 levels and below conventional new-build alternatives in Lazard’s analysis.

Berkeley Lab finds the same turn in completed projects. U.S. utility-scale solar LCOE increased 25% from 2022 to $60/MWh before tax credits in 2024. With available credits, it was $41/MWh. Higher financing costs, slightly higher capital cost, and weaker expected performance caused the increase.

This distinction matters. Technology learning can keep reducing equipment needs, while interest rates or trade restrictions raise the cost of turning equipment into a financed power plant. Tax policy can then shift the reported cost again.

Today’s ranking depends on assumptions and system value

EIA’s 2026 outlook projects 2031 U.S. LCOEs, including available credits, of $56.75/MWh for onshore wind, $58.33 for solar, $77.46 for combined-cycle gas, and $87.81 for advanced nuclear. Offshore wind is $118.79, while four-hour battery storage is $152.61. These are modeled future plants, not observed 2026 costs (EIA, 2026).

Low LCOE also does not mean identical service. Wind and solar output depends on weather and time. Berkeley Lab estimates that adding batteries raised a 2024 solar project’s LCOE by $36/MWh before credits, or $25 after credits. The IEA finds that variable renewables’ system value generally declines as their market share rises, creating greater need for storage, flexible demand, transmission, or dispatchable backup.

Therefore, the strongest conclusion concerns new plant-level generation. Solar and wind experienced historic cost declines, followed by a meaningful but partial rebound. Conventional costs were less transformed and remain sensitive to fuel, utilization, construction risk, and carbon policy. Future costs will depend on whether learning and scale outpace financing, supply-chain, tariff, and grid-integration pressures.

The evidence is strongest from 2009 onward and is weighted toward U.S. projects. It does not provide one consistent global series across all technologies. Methodologies, currencies, tax treatment, and capacity factors differ, so estimates should be compared within each study rather than treated as universal prices.