Contacts
1207 Delaware Avenue, Suite 1228 Wilmington, DE 19806
Let's discuss your project
Business Address: 1207 Delaware Avenue, Suite 1228 Wilmington, DE 19806

Solar Power Cost 2026: What a Solar Megawatt-Hour Costs, by Segment, and the Price It Has to Beat

Solar power cost 2026 — utility-scale solar LCOE –98/MWh vs gas combined-cycle –129/MWh, Lazard data charted by Axis Intelligence Research Solar Price-to-Beat Ratio SPBR™ 2026 — utility solar 0.77 vs new gas, 1.33 vs wholesale power, by Axis Intelligence Research

Solar Power Cost 2026

By Axis Intelligence Research

Co-author: Aidan Jad | Last updated: October 9, 2026 | License: CC BY 4.0

New-build, unsubsidized utility-scale solar costs $40 to $98 per megawatt-hour in the United States in 2026, according to Lazard’s LCOE+ Version 19.0, up from $38–78/MWh a year earlier. The midpoint, $69/MWh, is still below new gas combined-cycle at $90/MWh. Built solar costs $1.06 per watt for a single-axis-tracking utility plant in Q2 2026 (SEIA/Wood Mackenzie).


Quick Answer: How Much Does Solar Power Cost in 2026?

Utility-scale solar costs $40–98/MWh unsubsidized (Lazard, July 2026), or $1.06/W to build with trackers (SEIA/Wood Mackenzie, Q2 2026). Commercial and community solar costs $88–197/MWh. Axis Intelligence Research’s Solar Price-to-Beat Ratio (SPBR™) puts utility solar at 0.77 against new gas plants, but 1.33 against 2026 wholesale power.

Key Findings

  1. According to Lazard’s 2026 LCOE+, the unsubsidized LCOE of U.S. utility-scale solar is $40–98/MWh, with an average of $69/MWh, up 18% from last year.
  2. Axis Intelligence Research calculates a Solar Price-to-Beat Ratio (SPBR™) of 0.77 for utility-scale solar versus new gas combined-cycle: a solar MWh costs 23% less than a new-gas MWh.
  3. Against EIA’s forecast 2026 wholesale price of $52/MWh, the same solar MWh scores an SPBR™ of 1.33: utility solar is cheaper to build than gas, but costs more than the wholesale market currently pays.
  4. Commercial and community solar reaches an SPBR™ of 1.01 against the 2026 commercial retail rate of 14.00¢/kWh (EIA): it is effectively at parity before any tax credit.
  5. Berkeley Lab measured utility-scale PV at $1.61/W-AC ($1.22/W-DC) for 2024 projects, with an LCOE of $60/MWh before tax credits and a market value of just $32/MWh.

Solar LCOE in 2026: Cost per MWh by Segment

The levelized cost of energy (LCOE) is the constant price per megawatt-hour at which a new plant recovers its capital, operating costs and financing over its life. It is the standard way to compare a solar farm with a gas plant that has a fraction of the capital cost but a fuel bill for thirty years.

Lazard 2026 LCOE by technology (unsubsidized, new build)

TechnologyLCOE range ($/MWh)Average ($/MWh)Source
Solar PV — Utility$40–98$69Lazard LCOE+ v19.0
Solar PV — Community & C&I$88–197$142Lazard LCOE+ v19.0
Solar PV + Storage — Utility$61–156$109Lazard LCOE+ v19.0
Wind — Onshore$37–99$68Lazard LCOE+ v19.0
Gas Combined Cycle$51–129$90Lazard LCOE+ v19.0
Gas Peaking$144–276$210Lazard LCOE+ v19.0
Coal$72–177$125Lazard LCOE+ v19.0
U.S. Nuclear$175–255$215Lazard LCOE+ v19.0

Source: Lazard, Levelized Cost of Energy+ 2026, July 2026. Lazard assumes 60% debt at 8% and 40% equity at 12%.

Lazard’s 2026 report dropped rooftop residential solar from its comparison chart. For homeowner prices, Axis Intelligence Research keeps a separate dataset built from Berkeley Lab’s project-level data on what it costs to install solar panels, where the 2025 median was $3.6/W.

Why utility solar LCOE rose 18% in a year

Lazard’s high end moved further than its low end: the range went from $38–78/MWh in 2025 to $40–98/MWh in 2026. Lazard attributes the rise to higher capital costs, sustained interest rates, tariff pass-through and supply chain repricing. Over the long run the picture is different: Lazard’s utility solar average is down 81% since its 2009 edition.

Analyst commentary — Aidan Jad: Read the spread, not the midpoint. The low end barely moved, $38 to $40. The high end jumped $20. That says the best developers, with secured modules, cheap interconnection and good sites, are still building near their 2025 costs. Everyone else is paying for the tariff and the interest rate. A 2026 solar number without its range is half a number.

How much does solar cost per MWh in the rest of the world?

IRENA’s global weighted-average LCOE for new solar PV was $44/MWh in 2025, unchanged from 2024. India came in lowest at $35/MWh, China at $36/MWh, Brazil at $37/MWh and Germany highest among major markets at $65/MWh. Excluding China, the global average was $55/MWh. IRENA attributes U.S. and German costs close to double China’s to permitting delays, interconnection bottlenecks and higher balance-of-system costs, according to IRENA’s Renewable Power Generation Costs in 2025. IRENA also reports that more than 90% of utility-scale renewable capacity commissioned in 2025 was cheaper than the lowest-cost new fossil option.

The U.S. utility midpoint of $69/MWh is a different measure (Lazard models a single illustrative project; IRENA weights actual projects by capacity), so the two figures should not be subtracted. What both show is the same ordering: solar is among the cheapest new generation in every major market.

Utility-Scale Solar Cost per Watt: What It Costs to Build a Solar Farm

Installed price, 2026 and measured 2024 projects

MetricValueAs ofSource
Utility-scale, single-axis tracking$1.06/WQ2 2026SEIA/Wood Mackenzie SMI Q3 2026
Utility-scale, fixed-tilt$0.95/WQ2 2026SEIA/Wood Mackenzie SMI Q3 2026
Utility-scale module price$0.33/WQ2 2026SEIA/Wood Mackenzie SMI Q3 2026
Measured capex, 2024 projects (capacity-weighted)$1.61/W-AC ($1.22/W-DC)2024 CODBerkeley Lab
Measured capex, median 2024 project$1.70/W-AC2024 CODBerkeley Lab
Lazard capital cost assumption, utility PV$1,250–1,850/kWJuly 2026Lazard LCOE+ v19.0

SEIA/Wood Mackenzie prices are modeled bottom-up; Berkeley Lab’s are measured from EIA and FERC filings for projects that came online. Sources: Berkeley Lab, U.S. Utility-Scale Solar 2025 Data Update, October 2025.

The two figures measure different things. Wood Mackenzie’s $1.06/W is a modeled turnkey price for a system built today, in DC watts. Berkeley Lab’s $1.22/W-DC is what projects that came online in 2024 actually reported, including interconnection and network upgrades. The gap is mostly scope, not error.

How project size moves the price

Berkeley Lab found that 2024 projects larger than 250 MW-AC cost $1.38/W-AC, against $2.19/W-AC for projects of 5–20 MW. By Axis Intelligence Research’s calculation, that is a 37% discount for scale. In DC terms, Berkeley Lab reports $1.10/W-DC for the 250–800 MW band and $1.62/W-DC for 5–20 MW.

Solar farm cost by region

Region (2024 projects)Measured capexLCOE before tax creditsLCOE after tax creditsSource
ERCOT (Texas)$1.38/W-AC ($1.06/W-DC)$52/MWh$34/MWhBerkeley Lab
CAISO (California)—$52/MWh$34/MWhBerkeley Lab
Non-ISO West—$48/MWh$31/MWhBerkeley Lab
NYISO (New York)—$108/MWh$77/MWhBerkeley Lab

Dashes indicate Berkeley Lab did not publish a regional capex figure in its summary slides. Source: Berkeley Lab, October 2025.

New York solar costs roughly twice what Texas solar costs, per MWh. Smaller projects, weaker sun and expensive interconnection all stack in the same direction.

What Does Commercial Solar Cost per Watt in 2026?

Commercial system prices rose 5.6% year over year to $1.77/W in Q2 2026, according to SEIA/Wood Mackenzie, even as distributed-market module prices fell 16% to $0.37/W. Lazard’s capital cost assumption for community and C&I solar is $1,750–2,950/kW, producing an LCOE of $88–197/MWh.

Axis Intelligence Research calculates that a commercial watt cost 1.67 times a tracking utility-scale watt in Q2 2026 ($1.77 ÷ $1.06), and residential turnkey pricing at $3.36/W cost 3.17 times as much. Modules are no longer what separates those prices. Labor, roofs, permitting, customer acquisition and smaller system size do.

Analyst commentary — Aidan Jad: The module got 16% cheaper and the commercial system got 5.6% more expensive. That is the whole story of distributed solar in one quarter. Once the panel is a third of a dollar, the price lives in the racking, the electrician, the structural letter and the sales cycle, and none of those follow Chinese polysilicon down.

The Solar Price-to-Beat Ratio (SPBR™): Is Solar Actually Cheaper?

“Solar is the cheapest power” is true and incomplete. Cheapest compared with what? A utility planning a new plant compares solar with new gas. A merchant developer compares it with the wholesale price. A warehouse owner compares rooftop solar with the commercial rate on the utility bill. Each buyer has a different price to beat.

The Solar Price-to-Beat Ratio (SPBR™) is an original Axis Intelligence Research metric. It divides the cost of a solar MWh in a segment by the price that segment’s buyer would otherwise pay for a MWh.

Formula: SPBR™ = segment solar LCOE ($/MWh) ÷ price to beat ($/MWh)

  • Below 1.0: solar is the cheaper option for that buyer.
  • Around 1.0: parity; the decision turns on financing, tax credits and site.
  • Above 1.0: solar costs more than the alternative on a pure energy basis.

Inputs: Lazard LCOE+ v19.0 averages (July 2026); EIA STEO October 2026 forecasts for the 2026 wholesale price ($52/MWh across 11 hubs) and the 2026 commercial retail price (14.00¢/kWh = $140/MWh); Lazard’s marginal cost of existing gas combined-cycle ($42/MWh average).

Worked example: Utility solar vs. new gas combined-cycle = $69 ÷ $90 = 0.77.

SPBR™ readings, October 2026

Solar segmentPrice to beatSolar LCOE ($/MWh)Price to beat ($/MWh)SPBR™Source
Utility-scaleNew gas combined-cycle69900.77Lazard; Axis
Utility solar + storageNew gas peaker1092100.52Lazard; Axis
Community & C&ICommercial retail rate, 20261421401.01Lazard; EIA; Axis
Utility solar + storageNew gas combined-cycle109901.21Lazard; Axis
Utility-scaleWholesale power, 202669521.33Lazard; EIA; Axis
Utility-scaleExisting gas combined-cycle (marginal)69421.64Lazard; Axis

SPBR™ calculated by Axis Intelligence Research, October 9, 2026. This is the baseline reading. Licensed CC BY 4.0.

The table splits cleanly in two. Against anything that has to be built, a new combined-cycle plant or a new peaker, solar wins, and solar-plus-storage beats a new peaker by almost half. Against anything that already exists, the wholesale market or a paid-off gas plant burning $3.45/MMBtu gas, solar loses on energy alone.

Berkeley Lab’s measured data point the same way. Projects that came online in 2024 had an average LCOE of $60/MWh before tax credits, and solar’s wholesale energy-plus-capacity value averaged $32/MWh nationally. That is an SPBR™ of 1.88 on measured numbers, falling to 1.28 after tax credits (Axis Intelligence Research calculation from Berkeley Lab figures).

Analyst commentary — Aidan Jad: This is the reading that explains the 2026 market better than any headline LCOE. Solar is the cheapest thing you can build. It is not the cheapest thing you can buy, because the cheapest thing you can buy is power from a plant somebody finished paying for twenty years ago. Demand growth changes that math: when load outgrows the existing fleet, the price to beat stops being the wholesale price and becomes the cost of new capacity, and at that price solar scores 0.77.

Does the tax credit change the SPBR™?

Yes, materially. Using Lazard’s subsidized ranges, utility solar with the ITC has a midpoint of $55/MWh and with the PTC $48.5/MWh. Against the 2026 wholesale price, that moves the utility SPBR™ from 1.33 to 1.06 (ITC) or 0.93 (PTC). Commercial solar with the ITC has a midpoint of $113.5/MWh, an SPBR™ of 0.81 against the commercial rate.

Axis Intelligence Research calculates the federal credit is worth $14–20.5/MWh to a utility-scale project at the midpoint, about 20–30% of its unsubsidized cost. That figure is why the 2026 credit deadline matters.

What Happens to Solar Costs After the Tax Credit Deadline?

Under the One Big Beautiful Bill Act, the §45Y production credit and §48E investment credit end for wind and solar facilities placed in service after December 31, 2027, if construction began after July 4, 2026. IRS Notice 2025-42 sets the physical work test as the method for establishing construction start, with a continuity safe harbor of four calendar years.

The practical effect for 2026–2030 pricing:

  • Projects that began physical work before July 5, 2026 keep credit eligibility and can be placed in service through the safe-harbor window. Their PPA prices can still reflect subsidized LCOE.
  • Projects that started later must be online by the end of 2027 or price their power at unsubsidized LCOE: on Lazard’s numbers, roughly $14–20.5/MWh higher at the midpoint.
  • Storage keeps its credit. Lazard notes that the OBBBA preserved the storage ITC through 2033, which favors solar-plus-storage over standalone solar for post-deadline projects.

Residential buyers already crossed their cliff: the §25D credit for homeowner-owned systems ended for systems completed after December 31, 2025, as covered in our residential solar installation cost data.

What Moves the Cost of Solar Power?

Cost of capital

Solar is a capital-cost technology: Lazard’s breakdown attributes $36 of the $40/MWh low-end utility LCOE to capital and only $3 to fixed O&M. That makes the interest rate a cost input. In Lazard’s sensitivity, utility solar’s average LCOE runs from $53/MWh at a 4.2% WACC to $83/MWh at 10.0%, a 57% swing from financing alone (Axis Intelligence Research calculation).

Capacity factor and the sun

Berkeley Lab reports average capacity factors from 17% in the least sunny regions to 31% in the sunniest, and single-axis tracking adds about five percentage points in high-irradiance regions. Because the capital cost is fixed, every point of capacity factor lowers the cost per MWh.

Interconnection

Berkeley Lab counted 956 GW of solar in U.S. interconnection queues at the end of 2024, 47% of it paired with storage. If historical patterns hold, Berkeley Lab estimates only about 9% of requested solar capacity is ultimately built. Interconnection and network upgrade costs sit inside the measured capex, which is one reason measured costs exceed modeled ones.

Operations and maintenance

Median O&M for utility PV fell from about $40/kW-AC-year in 2012 to about $11/kW-AC-year in 2024, according to Berkeley Lab. Lazard’s 2026 fixed O&M range for utility solar is $8.25–26.25/kW-year.

Degradation

Berkeley Lab measured an average annual output decline of 1.6% at the project level, 1.3% for projects of 25 MW-AC or more and 1.7% for smaller ones. That figure includes inverter failures, stuck trackers and soiling, not only module degradation.

Fully Loaded Cost of a 1 MW Solar Plant: Years 1, 3 and 5

Price per watt is what the developer pays on day one. The owner’s cumulative outlay also includes operations. Axis Intelligence Research computes an undiscounted fully loaded cost per megawatt from Lazard’s 2026 midpoint inputs:

Formula: Cumulative cost (year n) = capital cost per MW + (fixed O&M per MW-year × n)

Configuration (per 1 MW)Capital costFixed O&M / yrYear 1Year 3Year 5Source
Utility PV, Lazard midpoint$1,550,000$17,250$1,567,250$1,601,750$1,636,250Lazard; Axis
Utility PV, Berkeley Lab measured$1,610,000$11,000$1,621,000$1,643,000$1,665,000Berkeley Lab; Axis
Community & C&I, Lazard midpoint$2,350,000$16,500$2,366,500$2,399,500$2,432,500Lazard; Axis

Midpoints: utility capex $1,250–1,850/kW, O&M $8.25–26.25/kW-yr; C&I capex $1,750–2,950/kW, O&M $13–20/kW-yr (Lazard). Berkeley Lab row uses $1.61/W-AC capex and $11/kW-AC-yr median O&M. Excludes financing, land, insurance, property tax and inverter replacement. Estimates by Axis Intelligence Research.

By year five, operations account for about 5% of a utility plant’s cumulative outlay. Solar is front-loaded: almost everything is paid before the first MWh, which is exactly why the interest rate moves its LCOE so hard.

How to Use These Numbers When Buying or Contracting Solar Power

  1. Match the benchmark to your role. A utility planner should compare solar LCOE with new-build alternatives; a corporate buyer with a PPA offer should compare it with forward wholesale prices; a commercial site owner should compare it with the delivered retail rate. The SPBR™ table shows each.
  2. Convert every bid to the same unit. Capex in $/W-DC or $/W-AC (Berkeley Lab uses both), energy in $/MWh, and state whether a tax credit is in the price.
  3. Ask when construction started. For utility projects, a pre-July 5, 2026 construction start is worth roughly $14–20.5/MWh on Lazard’s midpoints.
  4. Price storage separately. Berkeley Lab found batteries added $36/MWh before tax credits and $25/MWh after to a PV project’s LCOE in 2024.
  5. Check the region. Berkeley Lab’s post-credit LCOE ranged from $31/MWh in the non-ISO West to $77/MWh in NYISO for 2024 projects.

Solar Power Cost FAQ

What is the cheapest new electricity source in the U.S. in 2026?

On Lazard’s unsubsidized 2026 numbers, onshore wind ($37–99/MWh) and utility-scale solar ($40–98/MWh) have the lowest new-build costs. New gas combined-cycle averages $90/MWh. Existing, paid-off plants can run for less: Lazard puts existing gas combined-cycle at $32–51/MWh.

Why are solar PPA prices lower than solar LCOE?

Because PPAs are priced after tax credits. Berkeley Lab found that since 2016, levelized PPA prices have closely tracked LCOE with tax credits included, and PPAs for 2024 projects averaged $29/MWh, against a pre-credit LCOE of $60/MWh.

How much does a 100 MW solar farm cost to build?

At Berkeley Lab’s measured 2024 average of $1.61/W-AC, a 100 MW-AC plant cost about $161 million, by Axis Intelligence Research’s calculation. Projects above 250 MW-AC averaged $1.38/W-AC, and those of 5–20 MW averaged $2.19/W-AC.

Is solar plus battery storage cheaper than a gas peaker?

Yes. Lazard’s 2026 utility solar-plus-storage LCOE is $61–156/MWh, against $144–276/MWh for a new gas peaker. Axis Intelligence Research’s SPBR™ for that comparison is 0.52.

Does solar power still need subsidies to compete?

It depends on what it competes with. Against new gas plants, unsubsidized utility solar already wins (SPBR™ 0.77). Against the 2026 wholesale price, it needs the credit: the SPBR™ moves from 1.33 unsubsidized to 0.93 with the PTC.

Why did solar costs rise in 2026 when panel prices fell?

Modules are a small share of system cost. Utility modules fell 2% to $0.33/W while tracking system prices rose 2% to $1.06/W (SEIA/Wood Mackenzie, Q2 2026). Lazard cites capital costs, interest rates, tariffs and supply chain repricing for the LCOE increase.

How much has solar LCOE fallen since 2010?

IRENA reports the global cost of solar PV power has fallen 89% since 2010. Lazard’s U.S. utility-scale average is down 81% since its 2009 edition, even after the 2026 rise.

What is solar’s wholesale market value?

Berkeley Lab estimated solar’s average energy-plus-capacity value at $32/MWh nationally in 2024, down from $48/MWh in 2023. CAISO was lowest at $18/MWh and SPP highest at $60/MWh, reflecting how much solar each grid already has.

Methodology

Axis Intelligence Research assembled this page on October 9, 2026 from five primary sources, each fetched and read that day.

LCOE and capital cost assumptions: Lazard LCOE+ 2026 PDF, Version 19.0 (July 2026). Ranges and averages as published by Lazard; subsidized midpoints computed by Axis as the simple average of Lazard’s low and high subsidized values.

Measured utility-scale data: Berkeley Lab’s Utility-Scale Solar 2025 Data Update summary slides (October 2025), covering projects that came online through 2024. Berkeley Lab labels 2024 capex estimates as preliminary. This is the latest edition published at retrieval date; it is flagged as older data because the 2026 edition was not yet available.

2026 installed prices: SEIA/Wood Mackenzie US Solar Market Insight Q3 2026 (September 2026), as published in its public summary; Wood Mackenzie models prices bottom-up.

Global benchmarks: IRENA, Renewable Power Generation Costs in 2025 (July 2026).

Price-to-beat inputs: EIA Short-Term Energy Outlook, October 6, 2026, and its Table 2. 2026 values are partly forecast.

Axis calculations:

  • SPBR™ = segment solar LCOE ÷ price to beat. Inputs listed in the SPBR™ table.
  • Credit value = unsubsidized midpoint − subsidized midpoint (utility: $69 − $55 = $14; $69 − $48.5 = $20.5).
  • Fully loaded cost = capex + fixed O&M × years (midpoints shown in the table).
  • Cost ratios: $1.77 ÷ $1.06 = 1.67; $3.36 ÷ $1.06 = 3.17; 1 − $1.38 ÷ $2.19 = 37%.
  • 100 MW plant = $1.61/W-AC × 100,000,000 W = $161 million.

All arithmetic was run in Python before entering the text.

Scope of SPBR™: LCOE is a project-level cost; it excludes transmission, firming and integration costs, and the wholesale benchmark is a 2026 forecast average across 11 hubs. Regional readings would differ.

About This Dataset

Every figure on this page, with its source document, URL, retrieval date and calculation note, is in solar-power-cost.csv, one row per observation. Licensed CC BY 4.0. Cite as: Axis Intelligence Research, Solar Power Cost 2026, 2026.

Cite This Page

APA: Axis Intelligence Research, & Jad, A. (2026, October 9). Solar power cost 2026: What a solar megawatt-hour costs, by segment, and the price it has to beat. Axis Intelligence. https://axis-intelligence.com/solar-power-cost/

MLA: Axis Intelligence Research, and Aidan Jad. “Solar Power Cost 2026: What a Solar Megawatt-Hour Costs, by Segment, and the Price It Has to Beat.” Axis Intelligence, 9 Oct. 2026, axis-intelligence.com/solar-power-cost/.

Chicago: Axis Intelligence Research, and Aidan Jad. “Solar Power Cost 2026: What a Solar Megawatt-Hour Costs, by Segment, and the Price It Has to Beat.” Axis Intelligence, October 9, 2026. https://axis-intelligence.com/solar-power-cost/.

Related Axis Intelligence Research

Recent Posts

Tesla Solar Roof Cost 2026: The Final Price per Square Foot, Why Tesla Killed It, and What Replaces It

Tesla Solar Roof Cost 2026 By Axis Intelligence Research Co-author: Aidan Jad | Last updated: October 9, 2026 | License:

Credit Card Processing Fees 2026: What Stripe, Square and PayPal Keep Above Interchange

Credit Card Processing Fees 2026 By Axis Intelligence Research Co-author: Sarah Davis | Last updated: October 9, 2026 |

App Development Cost 2026: $155K to $3.7M by App Type, and What a Team in the US, Germany or Poland Really Costs

App Development Cost 2026 By Axis Intelligence Research Co-author: Elena Rodriguez | Last updated: October 9, 2026 | Lic

Axis Intelligence Research

Stay ahead on tech & data

Get notified when we publish or update datasets, trackers, research, and reports across technology, business, AI, cybersecurity, finance, infrastructure, energy, and more.

Research updates only. No spam. Unsubscribe anytime.