Why the energy has to be discounted too, the capital recovery shortcut and where it misses, and how capacity factor and the discount rate move the answer.
The levelised cost of electricity is the present value of every cost over the project's life divided by the present value of every megawatt-hour it produces, both at the same discount rate. For an illustrative 100 MW solar plant costing $100,000,000, with a 24 per cent capacity factor, $2,000,000 a year of operating cost and a 7 per cent discount rate over 30 years, the LCOE is $52.48 per MWh. It is the flat power price at which the project's NPV is exactly zero.
Worked in full in The Project Finance Handbook by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
A developer is comparing a solar project against offtake bids. Before looking at tax equity, debt sizing or the capture price, it needs one number: what a megawatt-hour costs to produce over the life of the plant. All figures are illustrative, nominal and pre-tax; no market price is implied.
| Input | Value |
|---|---|
| Capacity | 100 MW |
| Capital cost | $1,000,000 per MW, $100,000,000 |
| Capacity factor | 24% |
| Annual degradation | 0.5% |
| Operating cost, year 1 | $2,000,000 |
| Operating cost escalation | 2% a year |
| Life | 30 years |
| Discount rate | 7.0% |
Energy. Year one output is 100 MW × 8,760 hours × 24 per cent = 210,240 MWh, falling 0.5 per cent a year. Over 30 years the plant produces 5,870,565 MWh.
Costs. The capital cost is spent at year zero, $100,000,000. Operating costs escalate from $2,000,000, and their present value at 7 per cent is $30,481,866. Total present value of costs: $130,481,866.
Discounted energy. Each year's megawatt-hours are discounted at the same 7 per cent. The sum is 2,486,365 "present-value MWh".
LCOE = Σ (Capext + Opext) / (1 + r)t ÷ Σ Energyt / (1 + r)t
= 130,481,866 / 2,486,365 = $52.48 per MWh
Excel, with costs in C2:C32 and energy in D2:D32 from year 0: =(C2+NPV(7%,C3:C32))/NPV(7%,D3:D32)
Of the $52.48, $40.22 recovers the capital and $12.26 the operating costs. Solar is a capital-cost business, which is why the discount rate matters so much.
The check: sell every megawatt-hour at a flat $52.48 and the project's NPV at 7 per cent is exactly zero. If a model's LCOE does not pass that test, it is not an LCOE.
| Method | LCOE |
|---|---|
| Discounted costs over undiscounted energy | 22.23 |
| Nothing discounted | 30.85 |
| Capital recovery factor shortcut (no degradation, no escalation) | 47.84 |
| Costs and energy both discounted at 7% | 52.48 |
Discounting the costs but not the energy is the classic error and the most damaging: it divides money today by megawatt-hours spread over 30 years, and understates the break-even price by 57.6 per cent. Not discounting anything is less wrong only by accident. The capital recovery shortcut, capex times 0.08059 plus year-one opex, over year-one energy, is a respectable first estimate at $47.84, but it misses degradation and escalation, both of which push cost per MWh up over time.
| Discount rate | CF 20% | CF 24% | CF 28% |
|---|---|---|---|
| 5% | 54.37 | 45.31 | 38.84 |
| 7% | 62.97 | 52.48 | 44.98 |
| 9% | 72.29 | 60.24 | 51.63 |
At a 24 per cent capacity factor, each 100 basis points on the discount rate adds about $3.82. Capital cost moves it linearly: $44.44 at $800,000 per MW and $60.52 at $1,200,000. A 30 per cent investment tax credit, treated crudely as a reduction in capital cost at year zero, brings the LCOE to $40.41, a reduction of $12.07, which is why subsidy terms can decide whether a solar project works at all.
Beyond the discounting error, the frequent misuse is treating LCOE as the price the project will earn. A solar plant produces when every other solar plant does, so the price it actually captures is lower than the average price, often by a wide margin; see how to calculate a solar capture rate. And the rate in an LCOE is a single blended rate, while a financed project has senior debt sized on coverage, equity with its own hurdle, and tax. LCOE says what the energy costs. Whether the deal works is the job of the project model: a project finance model sizes the debt from CFADS and solves the equity return.
The book takes a solar and storage project from capital cost to equity return, with and without its tax credit; the free workbook for this case includes the power price each version needs to reach its hurdle.
Because LCOE is the constant price that makes NPV zero, and revenue is price times energy, so energy must be discounted at the same rate as costs. Dividing discounted costs of $130,481,866 by the undiscounted 5,870,565 MWh gives $22.23/MWh instead of $52.48, understating the break-even price by 57.6 per cent.
Annualise capex with the capital recovery factor, r / (1 minus (1 + r) to the power minus n), add year-one opex and divide by year-one energy. At 7 per cent over 30 years the factor is 0.08059, giving $47.84/MWh. It ignores degradation and opex escalation, so it understates the full calculation of $52.48 here.
Only if the price is flat for the whole life and the discount rate equals the return investors require. Project finance cares about the price that clears debt coverage and the equity hurdle, after tax, and real merchant revenue depends on capture price, not average price. LCOE is a cost benchmark; at 7 per cent and a 24 per cent capacity factor this one is $52.48/MWh.
This article is one calculation from The Project Finance Handbook. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: infrastructure, data centres and energy → · All 453 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.