Every solar quote comes with a price tag. But the number that actually tells you whether it’s a good deal isn’t the sticker price sitting at the bottom of the proposal — it’s the LCOE.
If you’ve been staring at a solar quote wondering whether €35 a megawatt-hour is actually a good number, or you’re trying to figure out why two developers quoted wildly different prices for what looks like the same rooftop, you’re in the right place. We’re going to walk through the actual math — not just wave at a formula and point you toward a calculator.
What Is LCOE, Really?
LCOE, or Levelized Cost of Energy, is the average cost of producing one unit of electricity from a solar system over its entire lifetime, once you’ve accounted for every cost and every kilowatt-hour it will ever generate. In plain terms: take all the money a project will ever spend, take all the electricity it will ever produce, and divide one by the other. That’s your cost per megawatt-hour.
It sounds simple. It isn’t, quite — because money spent five years from now isn’t worth the same as money spent today, and neither is electricity produced in year 25 versus year one. That’s where the formula earns its keep.
The LCOE Formula, Broken Down
Here’s the standard formula:
LCOE = Σ [(CAPEX + OPEX) / (1 + r)^t] ÷ Σ [Energy Production / (1 + r)^t]
Translated into plain English: add up every year’s costs, discounted back to today’s money. Add up every year’s electricity output, also discounted back to today. Divide the first number by the second. That’s your LCOE.
Here’s what each piece actually means:
- CAPEX — the upfront money: panels, inverters, racking, permitting, installation. This is the big number you see on the initial quote.
- OPEX — the ongoing costs: cleaning, monitoring, insurance, and — this one catches people off guard — inverter replacement, since inverters typically don’t last as long as the panels do.
- r (discount rate) — the return a lender or investor expects, or your own cost of capital. This one variable moves the final number more than almost anything else.
- t — the year you’re in, running from year one out to the end of the system’s life, usually 25 to 30 years for solar.
- Annual energy production — how much electricity the system actually generates each year, adjusted downward for degradation (panels don’t perform quite as well at year 20 as they did at year one).
None of these numbers exist in isolation. Change the discount rate by a single point and the whole calculation shifts more than you’d expect. We’ll get to that.
Let’s Actually Run the Numbers
Forget theory for a second. Say you’re looking at a utility-scale project with these assumptions:
- CAPEX: €1,000,000
- Annual OPEX: €20,000
- System lifetime: 25 years
- Annual energy production: 1,500 MWh
- Discount rate: 5%
In year one, that €20,000 in OPEX gets discounted by dividing by (1.05)¹ — so it’s worth about €19,048 in today’s money. By year 10, the same €20,000 discounted by (1.05)¹⁰ is worth roughly €12,278. By year 25, it’s down to about €5,910. Costs further in the future simply count for less — that’s the entire logic of discounting.
The same thing happens to the energy side. Your 1,500 MWh in year one is worth 1,500 MWh in the calculation. By year 25, discounted the same way, that 1,500 MWh (already reduced by degradation, more on that below) counts for a fraction of its nominal value in the sum.
Add up every year’s discounted costs — CAPEX up front plus 25 years of discounted OPEX — and you get the total present value of costs. Add up every year’s discounted, degradation-adjusted energy output, and you get total present value of energy. Divide one by the other.
Run the full 25-year calculation with these numbers and you land somewhere around €35 per MWh. Compare that against a market electricity price of €50/MWh, and the project clears a healthy margin — that’s the entire point of running LCOE in the first place. It’s not an abstract exercise; it’s the number that tells you whether the deal actually works.
What Actually Moves This Number
Not all inputs are created equal. Two of them do most of the heavy lifting.
Degradation. Solar modules lose a small amount of output every year — typically 0.4% to 0.5% annually after an initial dip in the first year from light-induced degradation. Skip this adjustment entirely and you can overestimate lifetime energy production by 8% to 12%, which quietly makes your LCOE look better than it really is. If a quote doesn’t mention degradation at all, that’s worth asking about.
Discount rate. This is the one people underestimate the most. A higher discount rate front-loads the weight of near-term costs and shrinks the value of far-future energy production, which pushes LCOE up. A lower discount rate does the opposite. Two developers using the same CAPEX and the same panels can hand you meaningfully different LCOE figures purely because they assumed a different cost of capital. In practice, most residential installers won’t volunteer the discount rate they used — it’s worth asking directly.
Irradiance and location. Your annual energy production estimate is only as good as the solar irradiance data behind it. A project modeled with optimistic sun-hour assumptions will show a lower LCOE on paper than it will deliver in reality. This is one reason regional differences in sunlight and peak sun hours matter so much for accurate project economics — the same panels in two different locations can produce very different numbers.
Self-consumption, for what it’s worth, affects the economic value of a residential or commercial system — how much you save versus how much you export — but it doesn’t change the LCOE calculation itself. That’s a common point of confusion worth clearing up early.
Where 2026 Numbers Actually Stand
Worth grounding this in current reality rather than an outdated “solar is basically free” narrative. According to Lazard’s 2026 Levelized Cost of Energy report, <cite index=”3-1″>unsubsidized utility-scale solar in the U.S. now ranges from roughly $40 to $98 per megawatt-hour, up from $38 to $78/MWh in 2025</cite> — a jump driven mainly by higher capital costs, interest rates, and tariff pressure rather than anything about the technology itself. Elsewhere in the world, the picture looks different: sun-rich regions with lower financing costs and mature supply chains are still landing utility-scale projects well below $40/MWh.
So “what’s a good LCOE” doesn’t have one universal answer — it depends heavily on your region, financing terms, and the year you’re pricing against. But the €35/MWh worked example above sits comfortably within a competitive range for a well-financed project in a strong-irradiance location, even in a year where costs have generally risen.
LCOE vs. Payback Period vs. IRR — They’re Not the Same Thing
This trips people up constantly, so it’s worth being direct about it.
| Metric | What it measures | Best used for |
| LCOE | Average cost per unit of energy over the project’s life | Comparing the true cost of different energy sources or project designs |
| Payback Period | How long until savings/revenue cover the initial investment | Quick gut-check on cash flow timing |
| IRR / ROI | Overall rate of return on the investment | Comparing this project against other investment opportunities |
A project can have a great LCOE and a mediocre IRR if the financing structure is unfavorable, or vice versa. LCOE tells you about the cost of the electricity itself; IRR and payback tell you about the investment. They answer different questions, and a lower LCOE isn’t automatically the “better” choice if the financing terms or site constraints tilt things another way.
What LCOE Doesn’t Tell You
This is the part most articles skip, and it’s the part that actually matters if you’re making a real decision.
LCOE gives you an average cost across the entire project life — it doesn’t capture when that electricity gets produced relative to when it’s actually needed. A megawatt-hour generated at noon on a sunny day isn’t worth the same, economically, as one generated at 7pm during peak demand. LCOE also leaves out grid integration costs, curtailment (when a grid operator tells a solar plant to reduce output because there’s too much supply), and the value of firm, dispatchable power. None of that shows up in the formula.
That doesn’t make LCOE useless — far from it. It’s still the best single number for comparing the raw cost-efficiency of different generation technologies or project designs. Just don’t treat it as the entire economic picture. It’s one very useful piece of a larger puzzle that also includes financing, grid value, and how the electricity will actually be used or sold.
Tools That Do the Math For You
Once you understand what’s happening under the hood, using a calculator stops feeling like a black box and starts feeling like a shortcut. A few worth knowing:
- NREL’s Comparative PV LCOE Calculator — good for trade-off analysis between different PV technologies and configurations.
- RatedPower’s LCOE calculator — pulls location-specific irradiance data, which matters more than most people realize for utility-scale accuracy.
- PV-Maps’ LCOE guide — a solid step-by-step methodology if you want to work through assumptions manually before automating.
- Energy Solutions’ LCOE Calculator — lets you stress-test assumptions like CAPEX, OPEX, capacity factor, and lifetime one at a time, which is genuinely useful for understanding sensitivity.
Plug your own CAPEX, OPEX, and irradiance data into one of these, and you’ll get a number specific to your project rather than a generic industry average.
Related Reading
If you’re building out a fuller picture of solar economics and how sunlight actually turns into a number on a spreadsheet, these go deeper on the pieces that feed into LCOE:
- What Is Solar Energy?
- The Photovoltaic Effect Explained
- Solar Radiation and Insolation
- Peak Sun Hours Explained
- Solar vs. Wind Energy
- On-Grid vs. Off-Grid Solar Systems
FAQs
What’s a good LCOE for solar in 2026?
It depends heavily on region and financing. Per Lazard’s 2026 report, unsubsidized utility-scale solar in the U.S. runs roughly $40–$98/MWh, while strong-irradiance markets with lower financing costs can still land well under $40/MWh. There’s no single global benchmark — compare against your local grid price and regional peers instead.
Does LCOE include self-consumption?
No. Self-consumption — how much of the energy you use on-site versus export to the grid — affects the economic value of a system, but it doesn’t change the LCOE calculation itself. LCOE is purely a cost-per-unit-of-energy figure.
How is LCOE different from ROI or payback period?
LCOE measures the average cost of the electricity produced. Payback period measures how long it takes for savings to cover the initial cost. IRR/ROI measures overall investment return. A project can score well on one and poorly on another, depending on financing structure.
Why does the discount rate matter so much?
Because it determines how much weight future costs and future energy production carry in the calculation. A higher discount rate shrinks the value of far-future electricity output, which pushes LCOE up — even if nothing about the physical system changed.
Do solar panels really lose efficiency every year?
Yes. Expect roughly 0.4% to 0.5% annual degradation after an initial first-year dip from light-induced degradation. Ignoring this can overstate lifetime energy production by 8–12%, which quietly understates LCOE.
Can I calculate LCOE without a fancy tool?
Yes, with the formula above and a spreadsheet — you just need CAPEX, OPEX, discount rate, expected energy production, and system lifetime. Calculators like NREL’s or RatedPower’s just automate the discounting math and add location-specific irradiance data.
