Talk:Levelized cost of electricity

Move discussion in progress
There is a move discussion in progress on Talk:Cost of electricity by source which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 15:32, 11 December 2019 (UTC)

Wrong Definition and Need for References
The claim that LCOE is "the net present value of the generated electrical energy over the lifetime of a electricity generating plant" (which I hadn't realized also occurs in the short description) is certainly wrong in many ways. Firstly it is an attempt to define not the discounted market value of *all* the energy produced but the npv of the *costs* on a per unit basis. And it is indeed just an "attempt" as the decision to discount future power at the same rate as money is only one of many options (and probably not a good one as money depreciates with inflation but amounts of energy do not).

Furthermore, there are various places where I have identified a citation needed as I don't think that the definitions and claims made in the article are self evident. Without such attribution it seems that they are just being presented as original research - which I believe Wikipedia frowns on. alQpr (talk) 22:22, 5 January 2020 (UTC)

In fact, the LCOE definition presented above is mistaken. The simple proof is presented here: https://drive.google.com/file/d/1Ls35IrbVfiDpA98x3A6zIN5SWjOfVlgs/view?usp=sharing adi_epp@wp.pl — Preceding unsigned comment added by 5.173.130.162 (talk) 10:03, 15 March 2021 (UTC)


 * I don't think that it would be good practice to follow a google file link from an anonymous poster. Leotohill (talk) 14:10, 15 March 2021 (UTC)

Ok ! You are right. I am now registered user Adiepp (talk) 10:48, 16 March 2021 (UTC).

Recently, I noticed that Wikipedia's description of the LCOE definition is wrong. I have prepared a proposal for changes (link): https://en.wikipedia.org/wiki/Draft:Levelized_cost_of_energy. I would be pleasantly surprised if any of the users commented on this. The situation is difficult because a large number of publications are based on this incorrect definition, and unfortunately there are few of us who see the nonsense of this definition. Unfortunately, I was unable to present a proof without some integral calculus theorems. However, the example given in the prepared amendment is so simple that it can be easily checked. I also understand that Wikipedia relies on proven sources of information. Unfortunately, there are very few of them in the case under consideration. Recently, James Loewen published an article on the subject. Unfortunately, the article was based on the observation of facts and had no mathematical basis. Maybe that's why he changed his mind! --Adiepp (talk) 09:18, 29 March 2021 (UTC)


 * I don't think that your proposed changes should be accepted as-is. LCOE has a widely accepted definition and long history. An attempt to give it a different definition would be confusing. While alternate metrics may provide additional insight into the economic value of a project,  they are indeed alternate metrics.  Loewen suggests "Present Value Cost of Energy (PVCOE)".


 * What I could accept would be some brief text about the weaknesses of LCOE, or its improper use. This could include a reference to alternate metrics. A sentence that mentions PVCOE and references the Loewen paper would be Ok. If you want to provide a deeper explanation of an alternate metric, it should go into a new WP article, referenced from this one.  However, if your analysis is not Loewen's, and is purely your own, it is Original Research, and not allowed. Leotohill (talk) 18:31, 29 March 2021 (UTC)

Since Loewen issued something of a retraction of his first paper I think a better reference is the more recent paper (currently listed as reference #7) by Loewen, James; Gagnon, Peter; Mai, Trieu. "A resolution to LCOE is not the metric you think it is". Utility Dive. Retrieved 7 October 2020. The key point that was agreed by both Loewen and his critics is that LCOE is a valid measure ONLY if calculated in real terms (i.e., adjusted for inflation). "What had originally caught Mr. Loewen's attention — and what none of us fully recognized until working through the problem together — is that, if you calculate LCOE in nominal terms, and then compare projects with unequal lifetimes, the values produced by LCOE can incorrectly rank-order technologies." What this means is that the discount factor applied to the energy in the denominator should NOT include inflation and should only reflect real growth of the economy and I think it is vital that this proviso be included in the article.alQpr (talk) 19:22, 29 March 2021 (UTC)

P.S. I would also retitle the paragraph headed "Discount rate of low-carbon technologies" (to remove the unnecessary restriction) and incorporate it into the main "Calculation" section ahead of the other "Considerations"alQpr (talk) 19:29, 29 March 2021 (UTC)


 * I think it would suffice to summarize the point made in the 7 October article, something like: If projects of unequal lifetime are compared, it is important that costs be adjusted for inflation, that is, use real costs instead of nominal costs.  Otherwise, the calculated levelized costs are not comparable.
 * From that article, I see the the NREL has been using real costs all along. It would be interesting to know how common is the use of real vs nominal costs in LCOE calculations for projects in general. Leotohill (talk) 23:04, 29 March 2021 (UTC)

It would be nice if the first 4 references (to Lai et al in which the formula is quoted but never derived) could be replaced, or at least supplemented, with an authoritative reference to the *derivation* of the formula. Such a reference would probably include a quotable description of what is intended by the discount rate "r", of the effects of changing it, and of when it should or should not be taken as the same in both numerator and denominator. To my mind an article which does not address these issues right at the outset is of little value.alQpr (talk) 00:14, 30 March 2021 (UTC)

Answer for Leotohill: The suggestion of "Original Research" does not fit the situation. I used the well-known theorem from integral calculus, so this is only a practical application of mathematics and has nothing to do with the original research. I assume that no one is questioning the correctness of long-established and known integral calculus theorems. I also gave an example with fixed cost and constant production, showing what the definitions of LCOE1 and LCOE2 lead to. Unfortunately, none of the debaters paid attention to it, and this is just an example showing what the definition of LCOE1 leads to. This example also confirms Loewen's conjecture. Unfortunately, in order to prove it in a general case, one must use theorems from integral calculus. As I wrote above, I haven't found a simpler method yet. --Adiepp (talk) 10:06, 30 March 2021 (UTC)

Answer for alQpr: Your remark on supplementing the main article with a correct definition makes sense from a mathematical point of view. Unfortunately, the only derivation that meets the basic criteria, such as the existence of the LCOE constant, which additionally meets the costs-revenues equilibrium equation is the definition presented in my study LCOE2 (position 11 in the draft). The LCOE2 is commonly used in the oil industry [12]. It is likely that mathematicians working in this field had noticed this problem before. Loewen also proposed this definition. The whole problem is production discounting, as you aptly wrote in 2020. As I showed in my study, production discounting leads directly to Loewen's article. This violates the concept of economic discounting of financial flows. Therefore, I wrote that the definition of LCOE1 is an economic paradox. This definition cannot be corrected to make it economically meaningful without ignoring the costs-revenues equilibrium equation.--Adiepp (talk) 10:43, 30 March 2021 (UTC)


 * This is the literal definition of original research. Present your paper for review to a reputable peer-reviewed journal, and when it is published, we can include it in this article. --Ita140188 (talk) 12:10, 30 March 2021 (UTC)


 * Science is the belief in the ignorance of reputable reviewers. Unfortunately, an example is a current page of wiki about LCOE.--Adiepp (talk) 12:59, 31 March 2021 (UTC)

Regarding the issue of incorrect definition, I too faced a similar question on the reson for levelizing Energy. NREL has a publication titled A Manual for the Economic Evaluation of Energy Efficiency and Renewable Energy Technologies by W. Short, D. Packey, and T. Holt, published in March 1995. It gives a good explanation for the definition of LCOE. It first derives the Total Life Cycle Cost (TLCC) as the present value of all costs incurred in the project's lifetime [pp 42]. There is an example given to distinguish between current dollar and constant [pp 43-44]. Further, it defines LCOE in terms of TLCC [pp 47-48]. From this it should be clear why the current definition of LCOE (on this wiki page) is correct. It would be helpful to cite this reference as the primary reference to avoid any further confusion. --Kcdsouza (talk) 21:59, 9 July 2022 (UTC).

Adding current estimates
I'd like to add some of the current estimates as listed on "Cost of electricity by source." Sampenrose (talk) 17:54, 16 February 2020 (UTC)
 * I agree. --PJ Geest (talk) 09:07, 4 June 2021 (UTC)
 * Also in other articles examples are given, even if the article is about a general concept, so there is no reason to remove the graph with examples.--PJ Geest (talk) 09:14, 4 June 2021 (UTC)
 * I disagree. An appropriate chart for this article might illustrate lcoe sensitivity to discount rate, or facility lifetime, or even dispatchability.  The chart in question is overly complex,  plotting four dimensions:  source of energy, installed capacity, price, and time.  Clearly, that  chart is meant to illustrative the relationship between these dimensions, and does nothing to clarify the meaning of lcoe. Leotohill (talk) 02:35, 5 June 2021 (UTC)
 * I don't think the chart is overly complex, there is only one dimension more than a normal graph. It would be just a normal graph with three dimensions (two axes and a legend) if you would remove the years. The graph just shows the LCOE evolving over time / installed capacity. Also in for example the article Life expectancy, the first graph in the article shows the life expectancy over time. Does it clarify the meaning of Life Expectancy - no, is it a useful graph for the article - yes. It is useful to include in an article the order of magnitude of a metric. There is nowhere in the guidelines of Wikipedia you cannot do that. On the contrary, I think it is good practice to include it. --PJ Geest (talk) 15:53, 5 June 2021 (UTC)
 * I'm still not very fond of the chart, but if you put it back, I won't remove it again. Leotohill (talk) 22:32, 6 June 2021 (UTC)
 * Are you more fond of the second graph? --PJ Geest (talk) 13:12, 7 June 2021 (UTC)
 * Yes! Simpler, but also includes gas non-peaker, which is good. Leotohill (talk) 15:49, 7 June 2021 (UTC)

What about the LCOE estimates by IEA. https://www.iea.org/reports/projected-costs-of-generating-electricity-2020 In this publication, nuclear has the lowest LCOE and there is also an emphasis on life-cycle extension of power plants which seems even cheaper. --Mikeschaerer (talk) 19:42, 3 August 2023 (UTC)

Another update needed
Sampenrose (talk) 14:03, 8 July 2022 (UTC)
 * 1) Lazard has a more recent release,, which is itself due to be updated in a few months.
 * 2) Lazard is just one source.
 * 3) The graphic isn't great: small, hard to read, and focused on the recent past rather than the present or a longer stretch of history.
 * 4) I think we should have two data summaries: one for current estimates from a range of sources, one for historical trends.
 * 5) For current estimates, a table and / or bar chart makes sense, and we should show its freshness (or staleness).
 * 6) A time series graphic makes sense for historical trends, but it should be editable by Wikipedians (maybe using SVGChart)?


 * I've just inserted a reference to an even more recent LCOE chart at Lazard (published April 2023). I plan to update the chart within a few days. — RCraig09 (talk) 20:12, 31 August 2023 (UTC)

Cleaning up introduction
There was a "citation needed" summary sentence in the introduction; I replaced it with a quotation.

This sentence: "The LCOE is a first-order economic assessment of the cost competitiveness of an electricity-generating system that incorporates all costs over its lifetime: initial investment, operations and maintenance, cost of fuel, and cost of capital.[2]" has errors:

1. LCOE is not a measure of competitiveness; "levelized avoided cost" is per https://www.eia.gov/outlooks/aeo/pdf/electricity_generation.pdf. 2. LCOE does not incorporate all costs, notably, it omits the costs of transmission and pollution. 3. The citation is old and 404s in both the original site and Wayback.

I'm going to replace it with another sentence from that EIA reference for now; it would be great to track down a second reference and swap it in. Sampenrose (talk) 16:30, 17 February 2020 (UTC)

The link in reference #2 (Levelized Cost and Levelized Avoided Cost of New Generation Resources in the Annual Energy Outlook 2019 [1]) currently goes to a different document dated Feb2021 alQpr (talk) 21:52, 29 March 2021 (UTC)
 * I have corrected the title of the reference. Leotohill (talk) 23:16, 29 March 2021 (UTC)

Claim needs support or removal.
"One of the most important limitations of LCOE is that it ignores time effects associated with matching electricity production to demand."

This is a strong claim with no evidence. "A limitation" is supported by, for example, Joskow 2011. We should support or remove the claim. — Preceding unsigned comment added by Sampenrose (talk • contribs) 04:22, 18 February 2020 (UTC)
 * I don't think this claim is very controversial, especially in energy systems with high renewable energy penetration. Anyway, I will find a reference in the next few days. --Ita140188 (talk) 06:19, 18 February 2020 (UTC)
 * I added a reference. I agree it's not controversial though. Clayoquot (talk &#124; contribs) 15:04, 5 June 2022 (UTC)

Content cut from Discount rate section
I removed the following:


 * The differences in outcomes for different assumed discount rates are dramatic — for example, NEA LCOE calculation for residential PV at 3% discount rate produces $150/MWh, while at 10% it produces $250/MWh. LCOE estimate prepared by Lazard (2020) for nuclear power based on 12% discount rate gives $163/MWh. Large differences in absolute values of LCOE estimates at different discount rates do not, however, cripple its policy guiding power.


 * A choice of 10% discount rate results in the energy production in 20 years being assigned accounting value of just 15%, which nearly triples the LCOE price. This approach, which is considered prudent from today's private financial investor's perspective, is being criticised as inappropriate for assessment of public infrastructure that mitigates climate change as it ignores social cost of the CO2 emissions for future generations and focuses on short-term investment perspective only. The approach has been criticised equally by proponents of nuclear and renewable technologies, which require high initial investment but then have low operational cost and, most importantly, are low-carbon. According to Social Cost of Carbon methodology, the discount rate for low-carbon technologies should be 1-3%.

I removed these passages as I think they are more confusing than helpful and they conflate three issues: 1) choosing a discount rate affects LCOE, 2) discount rates should be set differently for high-carbon and low-carbon technologies, and 3) what the discount rate for low-carbon technologies should be. Clayoquot (talk &#124; contribs) 16:44, 5 June 2022 (UTC)

"Levelized" listed at Redirects for discussion
An editor has identified a potential problem with the redirect Levelized and has thus listed it for discussion. This discussion will occur at Redirects for discussion/Log/2022 June 6 until a consensus is reached, and readers of this page are welcome to contribute to the discussion. Clayoquot (talk &#124; contribs) 05:05, 6 June 2022 (UTC)

Requested move 6 June 2022

 * The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this discussion. 

The result of the move request was: Moved (non-admin closure) >>> Extorc . talk  17:45, 12 June 2022 (UTC)

Levelized cost of energy → Levelized cost of electricity – I cannot move this page myself. This article is only about electricity not other forms of energy - for example it does not include heat from combined heat and power plants. The current title is likely to make some readers confused between "energy" and "electricity" Chidgk1 (talk) 15:15, 6 June 2022 (UTC)


 * Support. The terms are roughly equally used in Google Scholar, but "electricity" leaves less room for misinterpretation in this context. Clayoquot (talk &#124; contribs) 16:11, 6 June 2022 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
 * Support Energy has many forms but this article is about 'electricity'. 99.165.88.9 (talk) 00:36, 12 June 2022 (UTC)
 * Support good idea, I tried to do the move but found a redirect at the destination which points here, and it should be the other way around. If someday other parts of the energy sector get a "levelized" page, the proposed redirect page "levelized cost of energy" can be turned into a disambig page at that time. NewsAndEventsGuy (talk) 00:44, 12 June 2022 (UTC)

Long quote in long sentence
The LCOE "represents the average revenue per unit of electricity generated that would be required to recover the costs of building and operating a generating plant during an assumed financial life and duty cycle", and is calculated as the ratio between all the discounted costs over the lifetime of an electricity generating plant divided by a discounted sum of the actual energy amounts delivered.

Over the years I've found that deferring central definitions to long quotes interferes with editorial improvements to the ordering and clear statement of the article's primary material.

In this hybrid wallop, I've bolded many of the words that contribute to the rising shriek of the Great Primordial Yawn.

It makes me begin to contemplate what page of what textbook has dented the most foreheads, all world record. &mdash; MaxEnt 14:12, 4 October 2022 (UTC)


 * proposal

I mainly did this for my own notes, but something like this might work here, too.

LCOE represents the amortized life-cycle revenue per MW-hour [added hour] needed to recover the costs of construction and operation. LCOE is calculated as the ratio between all discounted electrical generation costs and all discounted electrical revenues. In multimodal facilities, such as those that also generate and sell process heat, it can be a somewhat subjective matter to isolate costs and revenues pertaining solely to electrical generation.

&mdash; MaxEnt 14:54, 4 October 2022 (UTC)

Isn't Lazard a biased reference?
For nuclear power, Lazard base its LCOE on the cost of the Vogtle Plant, whose overnight cost is catastrophically higher than planned for all sorts of reasons - including the bankruptcy of Westinghouse in 2017 - and very much higher than any other nuclear plant built in the last decade. Shouldn't another reference be used? Jacques de Selliers (talk) 00:44, 15 February 2024 (UTC)