A utility-scale solar farm typically brings in $90,000 to $140,000 per megawatt each year from selling electricity. That works out to roughly $12,000 to $25,000 per acre. Operating margins are high, usually 65% to 78%. But once you subtract depreciation and loan interest, the net margin can land anywhere from slightly negative to around 25%. Where a project ends up depends on three things: the price it sells power at, how much sun it gets, and how it was financed.
The quick math behind solar farm revenue
Every solar farm’s revenue comes from the same formula:
Megawatts × 8,760 hours × capacity factor × price per MWh
There are 8,760 hours in a year. Capacity factor is the share of that time the farm effectively runs at full output. Think 20% to 24% in much of the Midwest and Northeast, and 27% to 32% in the sunniest parts of Texas and the Southwest, especially with tracking panels.
Take 1 MW at a 27% capacity factor. That’s about 2,365 MWh a year. Here’s what different power prices do to it:
| Power price | Revenue per MW | Revenue, 100 MW farm |
|---|---|---|
| $35/MWh | $82,800 | $8.3M |
| $45/MWh | $106,400 | $10.6M |
| $55/MWh | $130,100 | $13.0M |
| $65/MWh | $153,700 | $15.4M |
Most utility-scale projects need 5 to 8 acres per megawatt, so a 100 MW farm sits on 500 to 800 acres. At $106,000 per MW and 6 acres per MW, you’re looking at about $17,700 in electricity sales per acre per year. Compare that to cropland, and you can see why so many landowners take the call.
One thing trips up almost everyone: “MW” can mean two different things. Panels are rated in DC, inverters in AC. A “100 MW” project might be 100 MWac, or 130 MWdc built to deliver 100 MWac. When you see a revenue-per-megawatt number anywhere, check which one it uses. Mixing them up can throw a comparison off by 25% or more.
Who actually gets paid
“How much does a solar farm make” has three different answers, depending on who you are.
The landowner gets a lease, usually $500 to $2,000 per acre per year, with 1% to 2% annual increases and terms of 25 to 40 years. Lease 200 acres at $1,000 and that’s $200,000 a year with no construction risk and no electricity to sell. Cash rent for cropland in much of the Midwest often runs $150 to $350 an acre, which is the whole story of why farmers sign. The catch is that your land is locked up for decades, and the decommissioning clause matters a lot more than the rent number. Also worth knowing: that $1,000 is only about 5% to 6% of what the farm earns per acre. The landowner gets paid first and gets paid the least.
The developer finds the site, lines up the lease, gets permits and fights for a spot in the interconnection queue. Most developers make their money by selling the project once it’s ready to build or just built, often for something like 5 to 20 cents per watt. On 100 MW, that’s $5M to $20M. Plenty of projects die along the way and earn nothing, so the winners have to cover the losers.
The owner-operator holds the farm and collects electricity revenue for 30 years or so. That’s the number the rest of this article is about.
A 100 MW farm, start to finish
Let’s build a realistic one. The farm is 100 MWac in Texas, with a 27% capacity factor, a $48/MWh power contract, and a total build cost of $1.40 per watt-AC ($140M, including interconnection and tariffs on equipment).
| Year-one line | Amount |
|---|---|
| Revenue (236,500 MWh × $48) | $11.4M |
| O&M and monitoring | –$1.5M |
| Insurance | –$0.45M |
| Land lease (about 600 acres) | –$0.4M |
| Property tax and local payments | –$0.5M |
| Asset management and admin | –$0.3M |
| EBITDA | $8.2M (72% margin) |
Divide that $8.2M by the $140M build cost and you get a 5.9% yield on cost. Not thrilling. Now add the federal tax credit. If the project qualifies for the 30% credit and sells it at about 93 cents on the dollar, that’s roughly $39M back, which drops the net cost to about $101M. The yield jumps to 8.1%, and simple payback shrinks from about 17 years to about 12.
That gap is the whole reason the credit mattered so much. It’s also why what happened to it in 2026 matters (more on that below).
Why “net profit margin” misleads on solar
If you’ve seen solar margin numbers online, you’ve probably seen them in net terms. Here’s the problem with that. Net profit margin is what’s left after depreciation and interest, and for solar those two lines are huge and depend on how the owner set things up, not on how well the farm runs.
Run our example. Book depreciation on $140M over 30 years is about $4.7M a year. That leaves roughly $3.5M of operating profit. Now say the owner financed half the project, $70M at 6.5%. Interest alone is about $4.5M. The farm shows a small loss on paper while cash flow still covers the loan payment, just without a lot of room to spare. Two owners with the same farm can report net margins 20 points apart.
Gross profit margin isn’t much better. Sunlight is free, so there’s no real cost of goods sold. Those 55% to 75% “gross margins” you see quoted are mostly operating margins with a different label.
What the industry actually looks at:
- EBITDA margin: how much of each revenue dollar survives operating costs. Healthy farms land in the high 60s to high 70s.
- Yield on cost: year-one EBITDA divided by total project cost. Quick and brutally honest.
- Unlevered IRR: the annual return if you paid cash for everything. After-tax, most utility-scale deals aim for roughly 7% to 10%.
- DSCR: how many times cash flow covers the loan payment. Lenders usually want to see around 1.3 or higher.
Margins by project type
| Project type | Typical size | Revenue per year | EBITDA margin |
|---|---|---|---|
| Utility-scale, solar only | 100 MWac | $9M – $14M | 65% – 78% |
| Utility-scale with 4-hour battery | 100 MW + 400 MWh | $12M – $18M | 60% – 75% |
| Community solar | 5 MWac | $0.7M – $1.0M | 45% – 60% |
| Commercial rooftop | 1 MW | $110K – $180K | 55% – 70% |
Community solar and rooftop projects earn more per megawatt because they sell at or near retail-style rates instead of wholesale prices. They also cost more to build per watt, and community projects spend real money finding and managing subscribers. Their margins are lower, and their returns depend heavily on state programs.
What changed in 2026
Any article that treats the solar tax credit as a sure thing is out of date. A federal law passed in July 2025 cut the credits short for solar. Here’s how it works now:
- Projects that began construction by July 4, 2026 keep the credits.
- Projects that started after that date have to be placed in service by the end of 2027 to qualify.
- Treasury’s updated rules on what counts as “beginning construction” got stricter. The old shortcut of spending 5% of costs to lock in eligibility no longer works for most large projects. You need real physical work.
- New restrictions limit equipment tied to certain foreign entities, which mostly means China-linked suppliers.
- The low-income community bonus is awarded competitively and goes mostly to small projects. Don’t assume a 100 MW farm gets it.
So what does that mean for profit? Farms that locked in credits are worth more now. New ones have to make the numbers work without the credit, and that’s much harder. Without it, our example needs a power price in the mid-$50s to reach a similar 7% yield. Rules shift, so any real project needs a tax advisor to confirm eligibility before the model gets finalized.
Storage, merchant risk and the midday problem
Solar farms all produce at the same time. In sunny grids like California and Texas, that means wholesale prices sag around midday, and in spring they can fall to zero or below. A farm with a fixed-price contract doesn’t care. A farm selling at market prices (“merchant”) absolutely does.
That’s where batteries come in. A 4-hour battery lets a farm hold energy and sell it in the evening, when prices tend to be higher. That can lift revenue by a third or more. However, a battery matched to a 100 MW farm can cost well over half as much as the solar side, and it degrades on its own schedule. It’s a good deal in some markets and a bad one in others.
Where profit leaks
Interconnection. Waiting years for grid approval is the classic killer. Land options, studies, deposits and legal fees keep piling up while nothing earns a dime. Queues in some regions run several years. This is also a cash game: development burns money long before revenue shows up, so your cash runway matters as much as your margin.
Degradation. Panels lose around 0.5% of output a year. By year 25, that farm produces about 89% of what it did in year one. Revenue slides while costs creep up.
Inverter replacement. Inverters often need replacing around year 12 to 15. On 100 MW, that’s a few million dollars most quick models skip.
Weather and insurance. Hail has wrecked large Texas projects in recent years. Premiums and deductibles reflect it.
Curtailment. Grid operators sometimes tell farms to cut output when there’s too much power. Depending on the contract, you may not get paid for what you didn’t send.
What moves the number most
Using our 100 MW example, here’s what a single change does to results:
| Change | Effect |
|---|---|
| Power price +$5/MWh | +$1.18M revenue, EBITDA up about 14% |
| Capacity factor 27% → 29% | +$0.84M revenue, EBITDA up about 10% |
| Operating costs +10% | –$0.3M, EBITDA down about 4% |
| Build cost +10% ($14M) | Yield on cost drops from 5.9% to about 5.3% |
| One-year interconnection delay | About $8M of EBITDA pushed back a year, with interest still running |
Notice how little operating costs matter compared to price and build cost. A farm can run perfectly and still disappoint if it was built too expensively or sold power too cheaply. That’s why experienced investors look at three things before anything else: whether a figure is AC or DC, whether the power price includes tax credits, and whether a margin is operating or net. Get those right and most solar numbers online start making sense.
