TECHNOLOGY EXPLAINER
Levelized Cost of Electricity (LCOE)
— the arithmetic that says lifetime matters more than efficiency
Japan's official review of power generation costs estimates the cost of electricity from perovskite solar cells in 2040 at ¥15.3/kWh as a reference value (excluding policy costs). Its assumptions are a construction cost of ¥204,000/kW, a capacity factor of 14% and an operating life of 20 years. Starting from values published by public bodies, this article breaks LCOE into its parts and, with every assumption stated, calculates how many yen it moves when lifetime, degradation, efficiency and discount rate change.
- What LCOE is, in three points
- Japan's targets: ¥20, ¥14, and cost-competitiveness without support
- The reference value in the cost review: what is inside ¥15.3/kWh
- Our calculation: reproducing the ¥15 range from the published inputs
- Our calculation: what happens when lifetime, degradation, efficiency and discount rate change
- A materials engineer's view (1): with a 10-year life, LCOE rises about 1.6 times
- A materials engineer's view (2): efficiency pays off through cost per unit area
- A materials engineer's view (3): installation cost and weight, the costs beyond the panel
- How this sits against solar PV worldwide
- Open problems, and what this article could not confirm
- Glossary / References / Claim-to-source audit
Sourced = stated in published material (link given)
Our calculation = a value this article derived from assumptions it states
Not yet confirmed = a plan, target or outlook with no confirmed track record
Structural readings and materials or process interpretations are marked separately as Commentary.
No market-size or price estimates from research firms are used.
1. What LCOE is, in three points
- Definition: the levelized cost of electricity (LCOE) is the cost per kWh obtained by dividing the total cost of building, running and decommissioning a power plant by its lifetime generation. The power generation cost verification by Japan's Agency for Natural Resources and Energy says it calculates, for model plants, the total cost over the whole life cycle from construction to decommissioning, converted to present value at a (real) discount rateSourced
- Discount rate: for continuity with earlier results, the review uses a uniform (real) discount rate of 3%Sourced
- How perovskites are treated: because the technology is still under development and its costs are not necessarily predictable, the review presents perovskite solar cells as a reference value, estimated with certain assumptions based on manufacturers' outlooksSourced
2. Japan's targets: ¥20, ¥14, and cost-competitiveness without support
Japan's Next-Generation Solar Cell Strategy (November 2024), compiled by a public-private council convened by METI (the Ministry of Economy, Trade and Industry), sets the following targets for the cost of electricityNot yet confirmed.
| Timing | What the strategy says |
|---|---|
| By 2025 (fiscal year) | Establish technology that makes the ¥20/kWh set out under the Green Innovation Fund possible (establishing the core technologies to achieve a generation cost of ¥20/kWh or less under certain conditions). The Green Innovation Fund is a government fund, administered by NEDO, that supports R&D toward carbon neutrality |
| By 2030 (fiscal year) | Establish technology that makes ¥14/kWh possible. Aim to build GW-scale production capacity without waiting until 2030 |
| 2040 | Aim to achieve a generation cost that allows deployment without support (¥10/kWh to ¥14/kWh or less). The ¥10/kWh applies "where costs fall substantially through progress in R&D and the like" |
The statements themselves are Sourced (Next-Generation Solar Cell Strategy [Reference 2]), but they are targets whose achievement has not been confirmedNot yet confirmed. The Green Innovation Fund's "Next-Generation Solar Cell Development Project" is given as ¥64.8 billion.
3. The reference value in the cost review: what is inside ¥15.3/kWh
The Power Generation Cost Verification Working Group of the Agency for Natural Resources and Energy (summary report of February 2025) estimated the 2040 cost of electricity from perovskite solar cells at ¥16.4/kWh including policy costs and ¥15.3/kWh excluding them, presented as reference valuesSourced.
| Input | Value | Explanation in the review |
|---|---|---|
| Construction cost | ¥204,000/kW | Equipment (panel) cost ¥143,000/kW plus installation and other costs ¥61,000/kW. Analysed from the cost outlooks of six manufacturers taking part in the public-private council, and from the periodic reports on commercial rooftop PV projects installed in 2023 under FIT/FIP (Japan's feed-in tariff and feed-in premium schemes) |
| Disposal cost | ¥6,000/kW | As above |
| O&M cost | ¥3,700/kW/year | Median of commercial rooftop PV projects under FIT/FIP |
| Capacity factor | 14% | Analysed from the six manufacturers' efficiency outlooks and the median of rooftop projects, assuming an efficiency of 20% for PV panels installed in 2023 |
| Operating life | 20 years | Median of the six manufacturers' outlooks (25 and 30 years in the review of commercial PV) |
| Model plant | 250 kW, rooftop | The same size as the commercial PV model plant. Rooftop mounting is assumed so as to compare under the same conditions |
| Production scale assumed | about 1 GW a year per company | Based on manufacturers' outlooks assuming a certain scale of annual production is achieved by 2040 |
All Sourced (summary report on power generation cost verification [Reference 1]; the same inputs also appear in the Next-Generation Solar Cell Strategy [Reference 2]). The review asks readers to note that installation on walls or windows reduces irradiance and so lowers the capacity factor, and that the actual cost of electricity can vary greatly with scale, mounting type and type of encapsulant, among other things.
The breakdown is construction ¥11.2, property tax ¥1.0, disposal ¥0.2 and O&M ¥3.0 (¥15.3 in total, excluding policy costs); adding policy costs of ¥1.1 equivalent to the IRR and ¥0.03 of budget-related costs gives ¥16.4. Because a new FIT/FIP category for next-generation solar cells is still under consideration and no IRR has been set, the review says it used, for convenience, the 4% IRR level of commercial PVSourced. In the same review, utility-scale PV comes to ¥10.0/kWh in 2023 (excluding policy costs) and ¥6.6 to ¥8.4/kWh in 2040, and residential PV to ¥14.0/kWh in 2023 and ¥7.6 to ¥10.4/kWh in 2040Sourced.
4. Our calculation: reproducing the ¥15 range from the published inputs
Assumptions (published inputs plus this article's simplifications)Our calculation
- Construction cost ¥204,000/kW, O&M ¥3,700/kW/year, disposal ¥6,000/kW (paid at the end of the operating life), capacity factor 14%, operating life 20 years, discount rate 3% (all published values)
- Generation is constant every year (no degradation), and property tax is left out of the calculation (this article's simplifications)
Calculation
- Annual generation: 1 kW × 8,760 h × 0.14 = 1,226.4 kWh
- Capital recovery factor at 3% over 20 years = 0.0672. Construction: ¥204,000 × 0.0672 ÷ 1,226.4 kWh = about ¥11.2/kWh
- O&M: ¥3,700 ÷ 1,226.4 kWh = about ¥3.0/kWh
- Disposal: the present value of ¥6,000 paid after 20 years (about ¥3,322), annualised = about ¥0.2/kWh
- Total = about ¥14.4/kWh (excluding property tax)
This matches the review's breakdown (construction ¥11.2, O&M ¥3.0, disposal ¥0.2), and adding the published property tax of ¥1.0 gives about ¥15.4, almost the same as ¥15.3. Assumptions and limits: this article has not checked the details of how the review calculates property tax, so the published value is simply added. The sensitivity calculations below use this ¥14.4 excluding property tax as the base case.
5. Our calculation: what happens when lifetime, degradation, efficiency and discount rate change
Assumption: starting from the base case in Section 4 (¥14.4/kWh), only one input is changed at a timeOur calculation
| Input changed | Value | LCOE (yen/kWh) | Difference from base |
|---|---|---|---|
| Operating life | 10 years | 22.9 | +8.6 (about 1.6 times) |
| Operating life | 15 years | 17.2 | +2.8 |
| Operating life | 25 years | 12.7 | −1.7 |
| Operating life | 30 years | 11.6 | −2.8 |
| Degradation (annual output loss) | 0.5% a year | 15.0 | +0.6 |
| Degradation | 1% a year | 15.6 | +1.3 |
| Degradation | 2% a year | 17.0 | +2.6 |
| Capacity factor | 12% | 16.8 | +2.4 |
| Capacity factor | 10% | 20.1 | +5.7 |
| Discount rate | 0% | 11.6 | −2.8 |
| Discount rate | 5% | 16.5 | +2.1 |
| Discount rate | 8% | 20.1 | +5.7 |
| Equipment cost | ¥60,000/kW | 9.8 | −4.6 |
The equipment cost of ¥60,000/kW is the value the Next-Generation Solar Cell Strategy describes as "the equipment cost needed to reach a generation cost of around ¥10/kWh (about ¥60,000/kW)" (our translation)Sourced. Our calculation gives ¥9.8 (excluding property tax), the same order as the strategy's "around ¥10/kWh".
Assumptions and limits: degradation is assumed to reduce output by a constant percentage each year, and is reflected in generation discounted at 3%. The capacity factors of 10% and 12% are this article's assumptions, with walls and similar surfaces in mind, and are not published values. These are one-at-a-time sensitivities; in reality several inputs move at once.
6. A materials engineer's view (1): with a 10-year life, LCOE rises about 1.6 times
In the calculations in Section 5, the input that moved LCOE most was operating life. Cutting the operating life from 20 years to 10 raises LCOE from ¥14.4 to ¥22.9 (about 1.6 times)Our calculation. The government review sets the operating life of perovskites at 20 years, shorter than for commercial PV (25 and 30 years), because that is the median of the six manufacturers' outlooksSourced.
What this shows is that, from an LCOE perspective, improving a material so that it does not degrade is worth as much as, or more than, extra efficiency. Degradation of 1% a year adds ¥1.3, and 2% a year adds ¥2.6. As Section 7 shows, raising efficiency from 15% to 20% lowers LCOE by about ¥2.0, so the increase from 2% annual degradation outweighs the gain from 5 points of efficiencyOur calculation. Materials development that improves the durability of encapsulants, barrier films and charge transport layers can be valued directly as a reduction in cost per kWh (our commentary). Methods for assessing durability were covered in our explainers on outdoor testing and durability assessment, and on encapsulation and barrier layers.
7. A materials engineer's view (2): efficiency pays off through cost per unit area
Equipment cost is expressed in yen per kW, but much of the cost of materials and manufacturing scales with area (m²). As an assumption for its demand estimates, the Next-Generation Solar Cell Strategy puts the generating capacity of roofs and walls at 150 W/m² (a conversion efficiency of 15%)Sourced.
Our calculation: converting the equipment cost of ¥143,000/kW to a per-area cost for a 150 W/m² panel gives about ¥21,450/m². Assuming efficiency rises while this per-area cost stays the same, equipment cost and LCOE (excluding property tax) becomeOur calculation:
- 15% efficiency: equipment cost ¥143,000/kW → LCOE about ¥14.4
- 18% efficiency: equipment cost about ¥119,000/kW → LCOE about ¥13.1
- 20% efficiency: equipment cost about ¥107,000/kW → LCOE about ¥12.4
Assumptions and limits: installation and other costs (¥61,000/kW) and O&M are held constant regardless of efficiency, but in reality some installation costs scale with area, so the benefit of higher efficiency could be somewhat larger. Going from 15% to 20% efficiency lowers LCOE by only about ¥2, far short of the increase from cutting the operating life from 20 years to 10 (about ¥8.6).
This is not to say that efficiency research is pointless; it is a difference in how the two act: efficiency works through cost per unit area, while lifetime works directly through the number of years in the denominator. When setting materials development goals, work that adds one point of efficiency and work that cuts degradation by 1% a year can be compared on the same yen-per-kWh yardstick (our commentary).
8. A materials engineer's view (3): installation cost and weight, the costs beyond the panel
The Next-Generation Solar Cell Strategy says that, according to FIT/FIP periodic report data, installation accounts for about 30% of the total capital cost of rooftop PV, and that it is necessary to look at the scope for cutting installation costs, not just panel costs. Citing an IRENA report, it also says that Japan's share of installation in capital cost is shown to be high compared with other countriesSourced. In the inputs to the cost review, too, installation and other costs are ¥61,000/kW (about 30%) of the ¥204,000/kW construction costSourced.
For film-type (flexible) cells, the strategy aims for competitiveness by capturing added value across the whole life cycle, including manufacturing, transport, installation, construction, collection, replacement, disposal and recyclingSourced.
Light, bendable materials could change the very cost structure of installation: mounting structures, reinforcement, heavy machinery and labour. Adhesive, fixing and waterproofing components (adhesives, tapes, sealants) could become "hidden protagonists" that affect LCOE as much as the absorber materials (our commentary). However, this article has not been able to confirm any official figures for the cost difference between installation methods.
9. How this sits against solar PV worldwide
The International Renewable Energy Agency's (IRENA) "Renewable Power Generation Costs in 2024" gives a global weighted-average LCOE of US$0.043/kWh for new utility-scale solar PV commissioned in 2024, down 90% from US$0.417/kWh in 2010, with installed costs down 87% from US$5,283/kW to US$691/kW. In China it was US$0.033/kWh and in India about US$0.038/kWhSourced.
The report notes that renewables are capital-intensive and LCOE is highly sensitive to financing conditions, and that IRENA's weighted average cost of capital (WACC) assumptions for 2024 ranged from 3.8% in Europe to 12% in AfricaSourced. In our calculation in Section 5, too, raising the discount rate from 3% to 8% lifted LCOE from ¥14.4 to ¥20.1Our calculation.
IRENA's values are actual results for large ground-mounted projects, while the Japanese review's values are estimates for a 250 kW rooftop model plant; currency, discount rate, scale and mounting all differ. This article does not convert one into the other to compare them. Whether perovskites compete on price with silicon in large ground-mounted plants, or go into places where silicon cannot be installed, is a difference in positioning that changes what LCOE they need to aim for (our commentary).
10. Open problems, and what this article could not confirm
(1) It is a reference value
The cost review treats its perovskite estimate as "an estimate at the current stage" and as a reference value. It also notes that the mounting types and scales in manufacturers' cost outlooks vary and do not necessarily assume a 250 kW rooftop installationSourced.
(2) The 20-year operating life is a median of outlooks
Twenty years is the median of six manufacturers' outlooks, not a 20-year outdoor track recordNot yet confirmed. The longest outdoor trial reported is around three years (see our explainer on outdoor testing and durability assessment).
(3) What this article could not confirm
Perovskite-related values in NREL's Annual Technology Baseline (ATB) were not consulted, because the NREL site could not be accessed from this article's working environment. The formula for property tax, official capacity factors for wall and window installations, and costs by installation method were also not confirmed within the scope of this article.
- Japan's targets are ¥20 by 2025, ¥14 by 2030 and ¥10 to ¥14/kWh or less by 2040 (cost-competitive without support)Not yet confirmed
- The cost review's 2040 reference value is ¥15.3/kWh excluding policy costs and ¥16.4/kWh including them. The assumptions are a construction cost of ¥204,000/kW, a 14% capacity factor, a 20-year operating life and a 3% discount rateSourced
- Calculating from the published inputs gives about ¥14.4/kWh excluding property tax, matching the published breakdownOur calculation
- With a 10-year life, it is about ¥22.9 (about 1.6 times). Lifetime has a bigger effect than improving efficiency from 15% to 20% (about ¥2)Our calculation
- About 30% of the construction cost is installation and other costs. Lightness and ease of installation feed directly into LCOESourced
- The global weighted average for utility-scale PV was US$0.043/kWh (2024). The assumptions differ, so no direct comparison is madeSourced
11. Glossary
- LCOE (levelized cost of electricity)
- Present value of total lifetime costs ÷ present value of lifetime generation, expressed in yen/kWh.
- Discount rate
- The rate used to convert future costs and generation into present value. 3% (real) in the cost review.
- Capital recovery factor
- A factor that converts an initial investment into equal annual amounts, given a discount rate and number of years. About 0.0672 at 3% over 20 years.
- Capacity factor
- Actual annual generation divided by the generation if the plant ran at rated output all year.
- Operating life
- The number of years a plant is assumed to run. 20 years for perovskites in the review.
- Construction cost
- Equipment (panel) cost plus installation and other costs, in yen/kW.
- O&M cost
- Costs incurred every year during operation, such as labour and repairs.
- Policy costs
- Costs related to support such as FIT/FIP and to government budgets. Usually not included under international methods.
- IRR
- Internal rate of return: the project return assumed when setting FIT/FIP prices.
- Model plant
- A standard power plant assumed for cost estimation. For perovskites, 250 kW on a rooftop.
- WACC
- Weighted average cost of capital: the weighted average of the costs of debt and equity.
- Cost-competitive without support
- Reaching a cost of electricity at which deployment proceeds without policy support.
12. References (primary sources)
- Power Generation Cost Verification Working Group, Advisory Committee for Natural Resources and Energy (Agency for Natural Resources and Energy) “Summary report on power generation cost verification”, 6 February 2025 (PDF, in Japanese) https://www.enecho.meti.go.jp/committee/council/basic_policy_subcommittee/mitoshi/cost_wg/pdf/cost_wg_20250206_01.pdf
- Public-Private Council for Expanding the Deployment of Next-Generation Solar Cells and Strengthening Industrial Competitiveness (METI) “Next-Generation Solar Cell Strategy”, November 2024 (PDF, in Japanese) https://www.meti.go.jp/shingikai/energy_environment/perovskite_solar_cell/pdf/20241128_1.pdf
- International Renewable Energy Agency (IRENA) “Renewable Power Generation Costs in 2024”, Executive summary (2025, PDF) https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2025/Jul/IRENA_TEC_RPGC_in_2024_Summary_2025.pdf
13. Claim-to-source audit
| Claim in the text | Basis | Label |
|---|---|---|
| That the total cost of a model plant is converted to present value at a (real) discount rate; the categories of capital cost (construction, property tax, disposal) and O&M; a uniform 3% discount rate. That perovskites are presented as a reference value because the technology is under development and costs are not highly predictable. The 2040 reference values of ¥16.4 (with policy costs) and ¥15.3 (without) and their breakdown (construction 11.2, property tax 1.0, disposal 0.2, O&M 3.0, IRR-equivalent policy cost 1.1, budget-related 0.03). The IRR of 4% used for convenience. The inputs (construction ¥204,000/kW = equipment 143,000 + installation and other 61,000; disposal ¥6,000/kW; O&M ¥3,700/kW/year; capacity factor 14%; operating life 20 years = median of six manufacturers; 250 kW rooftop; about 1 GW a year per company) and their basis; that wall or window installation lowers the capacity factor; that costs can vary with scale, mounting and encapsulant; that manufacturers' assumptions vary and this is an estimate at the current stage. Utility-scale PV ¥10.0 in 2023 and ¥6.6 to ¥8.4 in 2040, residential ¥14.0 in 2023 and ¥7.6 to ¥10.4 in 2040 (excluding policy costs), and a 25-year operating life for utility and residential PV (2040 table) | Reference 1 https://www.enecho.meti.go.jp/committee/council/basic_policy_subcommittee/mitoshi/cost_wg/pdf/cost_wg_20250206_01.pdf | Sourced |
| Establishing technology for ¥20/kWh by 2025 and ¥14/kWh by 2030; GW-scale capacity before 2030; ¥10 to ¥14/kWh or less, allowing deployment without support, by 2040 (¥10 where costs fall substantially). The Green Innovation Fund project of ¥64.8 billion. An equipment cost of about ¥60,000/kW needed for around ¥10/kWh. The assumption of 150 W/m² (15% efficiency). Installation at about 30% of capital cost for rooftop PV, the need to look at cutting installation costs, and Japan's high installation share citing IRENA. Life-cycle added value for film-type cells | Reference 2 https://www.meti.go.jp/shingikai/energy_environment/perovskite_solar_cell/pdf/20241128_1.pdf | Sourced |
| A global weighted-average LCOE of US$0.043/kWh for utility-scale PV in 2024, down 90% from US$0.417 in 2010; installed costs from US$5,283 to US$691/kW (down 87%); China US$0.033 and India about US$0.038. That LCOE is sensitive to financing conditions, with WACC assumptions from 3.8% in Europe to 12% in Africa | Reference 3 https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2025/Jul/IRENA_TEC_RPGC_in_2024_Summary_2025.pdf | Sourced |
| Achieving ¥20 by 2025, ¥14 by 2030 and ¥10 to ¥14/kWh or less by 2040, and a 20-year operating record | Targets and outlooks; actual achievement has not been confirmed (commentary) | Not yet confirmed |
| Annual generation of 1,226.4 kWh, a capital recovery factor of 0.0672, construction about 11.2, O&M about 3.0 and disposal about 0.2 for about ¥14.4, and about ¥15.4 with property tax of ¥1.0 added. Sensitivities to operating life, degradation, capacity factor, discount rate and equipment cost (10 years 22.9, 15 years 17.2, 25 years 12.7, 30 years 11.6; degradation 0.5% 15.0, 1% 15.6, 2% 17.0; capacity factor 12% 16.8, 10% 20.1; discount rate 0% 11.6, 5% 16.5, 8% 20.1; equipment ¥60,000/kW 9.8). About ¥21,450/m² per unit area; equipment costs of about ¥119,000 and ¥107,000/kW at 18% and 20% efficiency, and LCOE of about ¥13.1 and ¥12.4 | Our calculation. No degradation (base case), excluding property tax and policy costs, disposal paid in the final year, one input changed at a time, capacity factors of 10% and 12%, constant cost per unit area, and installation and O&M independent of efficiency are all this article's assumptions | Our calculation |
| Values from NREL's ATB, the property tax formula, official capacity factors for wall and window installations, and costs by installation method | Not stated because they could not be accessed from this article's working environment or were not confirmed within its scope (commentary) | Commentary |
| The framing of where materials matter. The reading that lifetime outweighs efficiency, the difference in how efficiency and lifetime act, and the point that development goals can be compared on the same yen-per-kWh yardstick. The reading that lightness, ease of installation and fixing components could be "hidden protagonists". The reading that differences in positioning change the target LCOE | This article's own framing and commentary based on published content and its own calculations. Not views expressed by the institutions | Commentary |
| That Figs. 1, 2 and 3 are explanatory drawings (the values in Fig. 2 are published values; those in Fig. 3 are our calculation), and that the hero image and Fig. 4 are AI-generated images | Our note (commentary) | Commentary |
Last updated 25 September 2026. Sources are limited to primary material from public bodies (the Agency for Natural Resources and Energy's power generation cost verification, METI's Next-Generation Solar Cell Strategy and IRENA's report), and no estimates from research firms are used. All of this article's calculations state their assumptions and are not published values. Values from NREL's ATB, the property tax formula and official capacity factors for wall and window installations are not stated because this article could not confirm them. The perovskite values in the cost review are those the government presented as "reference values". All figures are explanatory concept graphics. Figs. 1, 2 and 3 are vector drawings; the hero image and Fig. 4 are AI-generated images, and none of them shows a real product or installation.