Levelized Cost of New Electricity Generating Technologies
ShareThisMay 12, 2009· 28 Comments
Updated February 2nd, 2010
The Energy Information Administration (EIA) produces forecasts of energy supply and demand for the next 20 years using the National Energy Modeling System (NEMS)[1]. These forecasts are updated annually and published in the Annual Energy Outlook (AEO).[2] All sectors of the energy system are represented in NEMS, including the electric power generation, transmission, and distribution system.
To meet electricity demand, the EIA represents the existing generating plants, retires those that have come to the end of their economic life, and builds additional plants to meet projected demand from the residential, commercial, industrial, and transportation sectors. As a result, EIA must represent a slate of technologies, their capital and operating costs, their availability and capacity factors, the financial structure and subsidies, the time to construct the plant, the utilization of the plant, and expected future cost changes, including fuel input for fossil and nuclear plants.
To determine the most economic technology for the type of demand (base, intermediate, or peaking load) for which new capacity is needed, NEMS competes the technologies based on the economics of their levelized costs. Levelized costs represent the present value of the total cost of building and operating a generating plant over its financial life, converted to equal annual payments and amortized over expected annual generation from an assumed duty cycle.
The table below provides the average national levelized costs for the generating technologies represented in the AEO2010 reference case.[3] The values shown in the table do not include financial incentives such as state or federal tax credits, which impact the cost and the competitiveness of the technology. These incentives, however, are incorporated in the evaluation of the technologies in NEMS based on current laws and regulations in effect at the time of the modeling exercise, as well as regional differences in the cost and performance of the technology, such as labor rates and availability of wind or sun resources.
In the AEO2010 reference case, a 3-percentage point increase in the cost of capital is added when evaluating investments in greenhouse gas intensive technologies such as coal-fired power plants without carbon capture and sequestration (CCS) technology and coal-to-liquids plants. The 3-percentage point adjustment is similar to a $15 per ton carbon dioxide emissions fee when investing in a new coal plant without CCS technology. This adjustment represents the implicit hurdle being added to greenhouse gas intensive projects to account for the possibility that they may need to purchase allowances or invest in other greenhouse gas emission-reducing projects that offset their emissions in the future. Thus, the levelized capital costs of coal-fired plants without CCS are likely higher than most current coal project costs.
The levelized cost for each technology is evaluated based on the capacity factor indicated, which generally corresponds to the maximum availability of each technology. However, some technologies, such as a conventional combined cycle turbine, that may look relatively expensive at its maximum capacity factor may be the most economic option when evaluated at a lower capacity factor associated with an intermediate load rather than base load facility.[4]
Simple combustion turbines (conventional or advanced technology) are typically used for peak load, and are thus evaluated at a 30 percent capacity factor. Intermittent renewable resources, e.g. wind and solar, are not operator controlled, but dependent on the weather or the sun shining. Since the availability of wind or solar is dependent of forces outside of the operator’s control, their levelized costs are not directly comparable to those for other technologies although the average annual capacity factor may be similar. Because intermittent technologies do not provide the same contribution to system reliability as technologies that are operator controlled and dispatched, they may require additional system investment as back-up power that are not included in the levelized costs shown below.
Levelized Cost of New Generating Technologies, 2016
($2008 per megawatt hour)
| Plant Type | Capacity Factor (%) | Levelized Capital Cost | Fixed O&M | Variable O&M (including fuel) | Transmission Investment | Total System Levelized Cost |
| Conventional Coal | 85 | 69.2 | 3.8 | 23.9 | 3.6 | 100.4 |
| Advanced Coal | 85 | 81.2 | 5.3 | 20.4 | 3.6 | 110.5 |
| Advanced Coal with CCS | 85 | 92.6 | 6.3 | 26.4 | 3.9 | 129.3 |
| Natural Gas-fired | ||||||
| - Conventional Combined Cycle | 87 | 22.9 | 1.7 | 54.9 | 3.6 | 83.1 |
| - Advanced Combined Cycle | 87 | 22.4 | 1.6 | 51.7 | 3.6 | 79.3 |
| - Advanced CC with CCS | 87 | 43.8 | 2.7 | 63.0 | 3.8 | 113.3 |
| - Conventional Combustion Turbine | 30 | 41.1 | 4.7 | 82.9 | 10.8 | 139.5 |
| - Advanced Combustion Turbine | 30 | 38.5 | 4.1 | 70.0 | 10.8 | 123.5 |
| Advanced Nuclear | 90 | 94.9 | 11.7 | 9.4 | 3.0 | 119.0 |
| Wind | 34.4 | 130.5 | 10.4 | 0.0 | 8.4 | 149.3 |
| Wind-Offshore | 39.3 | 159.9 | 23.8 | 0.0 | 7.4 | 191.1 |
| Solar PV | 21.7 | 376.8 | 6.4 | 0.0 | 13.0 | 396.1 |
| Solar Thermal | 31.2 | 224.4 | 21.8 | 0.0 | 10.4 | 256.6 |
| Geothermal | 90 | 88.0 | 22.9 | 0.0 | 4.8 | 115.7 |
| Biomass | 83 | 73.3 | 9.1 | 24.9 | 3.8 | 111.0 |
| Hydro | 51.4 | 103.7 | 3.5 | 7.1 | 5.7 | 119.9 |
Source: Energy Information Administration, Annual Energy Outlook 2010, http://www.eia.doe.gov/oiaf/aeo/electricity_generation.html
[1] Energy Information Administration, NEMS documentation, http://www.eia.doe.gov/oiaf/aeo/overview/index.html
[2] Energy Information Administration, Annual Energy Outlook 2010, http://www.eia.doe.gov/oiaf/aeo/index.html
[3] Energy Information Administration, Annual Energy Outlook 2010, http://www.eia.doe.gov/oiaf/aeo/electricity_generation.html
[4] Base load plants are facilities that operate almost continuously, generally at annual utilization rates of 70 percent or higher. Intermediate load plants are facilities that operate less frequently than base load plants, generally at annual utilization rates between 25 and 70 percent. Peaking plants are facilities that only run when the demand for electricity is very high, generally at annual utilization rates less than 25 percent.
May 20th, 2009 at 10:58 am
Very relevant data; our thanks and appreciation!
Two questions:
1) what was the number of years (lifetime) used for the various generating systems in the cost analysis?
2) what is the “energy pay back ratio” for the various systems (i.e. years after the total potential energy generated – since day one – is equal to the total energy used to manufacture, install and operate up to that year)
May 22nd, 2009 at 9:07 am
The financial life of the units are assumed to be 20 years, i.e. planning decisions to build a new unit are based on a life cycle cost analysis over a 20-year period. The actual life of a generating unit, however, far exceeds the 20-year financial life. The analysis retires units based on planned retirements by utility companies and past operating experience.
May 26th, 2009 at 2:08 am
Dear Administrator,
Is that possible to find the LEC cost for the systems which dont generate electricty,but produce only steam.If so,which are the factors considered?
Thanking you in advance
K.Muniasamy
May 27th, 2009 at 8:10 am
Those costs are embedded in the National Energy Modeling System that the Energy Information Administration (EIA) uses. You can see some of the cost components that go into calculating them at: http://www.eia.doe.gov/oiaf/aeo/assumption/index.html.
Select the residential, commercial, transportation, and/or industrial model components, depending on what technologies you are interested in. For combined heat and power systems, for example, see Table 6.7 in the industrial sector.
June 25th, 2009 at 1:55 am
Someone needs to compute the cost of CO2 replacing anexisting coal unit (~1 tonCO2/MWH) with generation at the above prices. For those technologies that emit zero CO2 then the above cost is the cost of removing a ton of CO2:
Nuclear $107 per ton CO2
Wind $141-229 per ton CO2
Solar $263-396 per ton CO2
Hydro $114 per ton of CO2
For technologies that emit CO2 a half ton of CO2/MWH you need to double the above costs:
Gas turbine combined cycle $160/ton CO2
These are much higher than the $15/ton on which many of the cost numbers are based. The power pools are calculating $50-100 per ton to fuel switch however the pool needs to have both sufficient coal and GTCC capacity to make it happen( and most don’t in a large scale). Consistent with this the EU had a price of about $50/ton (before overallocation of CO2 credits was discovered) which was based on utility fuel switching.
All of these are cost numbers and do not reflect market dynamics which could multiply them several fold as we’ve seen in other energy commodity markets in time of shortage or “speculation”.
THe Waxman studies base $15/ton CO2 on buying international offsets for 50-80% of the reduction. The idea that someone would sell to us at $15 when our marginal cost is $50-400 violates basic common sense. We’ll be at the mercy of a “CO2 OPEC”.
August 6th, 2009 at 8:35 am
Dear IER:
Thanks for your “Levelized Costs of New Generating Technologies” analysis. Some questions:
1. I estimate that you did the analysis from data at EIA, rather than pulling the numbers directly from EIA. Correct?
2. What accounts for the differences in your table and the EIA Figure 57 in the AEO Outlook 2009: “Least Expensive Technology Options Are Likely Choices for New Capacity”? For example, your numbers for wind power are 142 and 230 per Mw-hour and the AEO Outlook puts this number at about 100.
Thanks much,
Steve Goreham
August 10th, 2009 at 11:21 am
Steve,
The numbers come directly from EIA. They are available upon request from the Coal and Electric Power Division. The difference between the numbers in the table on the IER website, which are for 2016, and Figure 57 of the Annual Energy Outlook (AEO) 2009 is that the numbers in the table are from the revised AEO 2009 that includes the stimulus, i.e. the American Recovery and Reinvestment Act (ARRA) of February 2009. (See http://www.eia.doe.gov/oiaf/servicerpt/stimulus/index.html for EIA’s service report.) In the case of the revised AEO 2009, more wind capacity is built earlier in the forecast than in the AEO 2009 without the stimulus (10 gigawatts more by 2010 and 33 gigawatts more by 2020). As more wind units are constructed, the better wind sites are used up earlier, and wind becomes more expensive due to access and resource availability issues.
- Mary Hutzler
August 24th, 2009 at 1:46 pm
Dear IER,
Is the hydro 114.1 number, conventional hydro, hydrokinetic, or combined? I ask because non-conventional hydro might be higher capacity factor.
thx
ed
August 26th, 2009 at 3:42 pm
Ed,
The number is conventional hydro.
Thanks,
Mary Hutzler
September 16th, 2009 at 11:50 pm
Dear IER:
Thank you for your Cost analysis.
I have some questions:
1) What discount rate did you use?
2) If there is summery of assumptions, please let me know.
Best regards
September 17th, 2009 at 11:24 am
Koji,
The assumptions are contained in the report: Assumptions to the Annual Energy Outlook 2009 in the Electricity chapter. The link is: http://www.eia.doe.gov/oiaf/aeo/assumption/electricity.html.
Thanks,
Mary Hutzler
September 22nd, 2009 at 9:09 am
What are the prices of natural gas and coal (fuel factors) modeled in the analysis? I went to the link with the assumptions but it only explained the methodology, not the actual prices used. Thanks.
September 24th, 2009 at 6:55 am
Andy,
The fuel prices are determined by the model at equilibruim. They can be found in Table A1 of the following link: http://www.eia.doe.gov/oiaf/servicerpt/stimulus/pdf/sroiaf(2009)03.pdf.
Thanks,
Mary Hutzler
September 25th, 2009 at 1:22 pm
Just to be clear re natural gas prices. Are the prices used based on the Henry Hub prices in Table A1?
Thanks.
December 9th, 2009 at 5:32 pm
I like your graph comparing the cost od 20 different sources of energy. It would be even more useful if you provided also info how much CO2 is generated per kw of power, by each. Thank you.
Wane Leposavic, investor
Las Vegas
December 13th, 2009 at 4:15 pm
dear IER,
thank you for this analysis. I was wondering, with this data in mind, it is possible that for example wind and solar can be profitable compared to other technologies? Even with carbon taxes, nuclear beats them all?
Thanks!
December 13th, 2009 at 5:13 pm
dear IER,
about my previous question, let me formulate it differently. The average retail sales price for electricity in the US is around 10 cents/kwh ($100 per Mwh). If you draw that $100 line inside the above graph, only 3 technologies go under that $100 line.
I don’t get how technologies that are more expensive than the average retail price can be profitable.
Thanks a lot for your answer!
Ray
April 15th, 2010 at 2:55 am
On wikipedia the Levelized costs’ denominator ( kwh or other unit of energy production) is discounted as if it were cash. Can anyone explain why that is done? thanks,
Mike
April 15th, 2010 at 12:07 pm
@mike honey,
Basically, you start with the notion that LCOE is the electricity price at the point where the present-value revenue for the electricity (price * quantity) equals the present-value cost (using the notation from wikipedia, where P is the price of electricity in year t):
sum (Et * P * (1+r)^-t)= sum (It + Mt + Ft) * (1+r)^-t)
So when you solve for P, the discount factor on the left hand side ofthe equation ends up in the denominator, along with Et.
So LCOE=P=[ sum (It + Mt + Ft) * (1+r)^-t)]/[sum(Et*(1+r)^-t]
April 16th, 2010 at 1:39 pm
EIA makes a critical judgment call when estimating wind energy costs by assuming a capacity factor of 40%. Barring a major future technological breakthrough, actual U.S. wind energy capacity factors bump along at or below 25% (as low as 22%) in recent years, even after the advent of the latest turbine designs. Replace the 40% (fiction) with 24% (reality), and wind levelized costs jump to nearly $250/MW-hr, more than 2x nuclear. Also, biomass looks cheap, but there is not enough of it to matter in the big picture.
April 18th, 2010 at 9:47 am
Hey Guys, thanks for a great report. I’m in the process of finishing up a review of potential gas supply from the various shale zones in North America and wanted to include a section on the possible efects of this gas supply on electric generation. I anticipated 2-5 days effort to get the information and do the calcs. Got on line this morning to start gathering the info, went to IER first and found your report – basically job done. The report I’m working on is technical and for a for profit company – it will never be seen in public, but IER will be given credit.
Note, the US energy scene is technically, potentially in very good shape and we could drive energy supplies up and costs done, cleanly, if the politicans, bureaucrats and business that seek advantage through the political and regulatory process got out of the way. Don’t mind arb’ing governmental stupidity but refuse to pimp and whore to use the political process to reach into tax payer’s pockets and transfer money into mine! Examples are available, such as a rich oil/gas guy advertising for wind generation of electric.
May 1st, 2010 at 11:23 am
Dear IER,
I am actually working on a project on wind power turbine. Is it possible to have a link for a more detailed report on wind levelized energy cost ?
The estimation you did was only for the US ?
May 8th, 2010 at 2:40 pm
$0.01 per MWh could be possible with aneutronic nuclear fusion reactor fueled with relatively inexpensive fuels such as hydrogen-boron and hydrogen-lithium. It can be a cheap, clean and safe source of electricity without any type of radioactive waste.
June 20th, 2010 at 11:11 am
Hey guys,
I am currently working on my final thesis for my bachelor-degree in Vienna, Austria. I am comparing several studies which calculate LCOE and try to explain how various studies have vast differences in their results concerning the LCOE per MWh.
One of the studies examined is the EIA-Outlook. I know this question was already posted, but I couldn’t find an answer in the link provided (see post of Koji on September 16th, 2009): For calculating the LCOE, which discount rate/cost of capital was used? I would be very grateful for an answer!
June 21st, 2010 at 2:36 pm
Nikolaus,
The average cost of capital between 2010 and 2035 is 0.115. In 2016, it is 0.114. However, for new coal plants without sequestration, 3 percentage points are added to cost of debt and cost of equity to represent a carbon risk premium. And, for new renewables online by 2015, a 2 percentage point reduction in cost of debt and equity is made to represent loan guarantees in the stimulus.
June 24th, 2010 at 9:37 am
Hey,
thanks for the information!
Is there any justification for assuming capital costs of 11,5%? The IEA for example use 5 and 10 percent in two their scenarios…
August 3rd, 2010 at 4:21 am
The numbers for solar PV and thermal are outdated. The Capital Cost for PV is about half of what is assumed here.The capacity factor for thermal is also too high especially when considering also outages and should be more like 0.26
August 3rd, 2010 at 12:23 pm
@tzvika18
The levelized cost numbers are updated each year, with the next update coming early in 2011. A recent review of the capital costs for electric power plant technologies shows that PV has the largest range in costs. There are recent utility-scale projects that are being or have been installed in the $3,000-4,000/kW range (see http://www.powermag.com/issues/cover_stories/Top-Plants-El-Dorado-Energys-Solar-Facility-Boulder-City-Nevada_2302_p2.html ), but there are also recent utility-scale projects that have been installed in the$5,000-6000/kW range, such as FPL’s DeSoto plant. The PV market is in a state of flux right now. The recession has resulted in surplus products, countries are changing their lucrative pricing policies, and Chinese manufacturers are interested in entering the global market. It is unclear what the long-term costs will be.
The costs are in units of “net-AC”, which are about 23% higher than the cost fora comparable system quoted in terms of “peak DC” watts (that is,$6,000/kW[AC]=$4,620/kW[DC]). Typically, PV module manufacturers quote prices in terms of “rated peak output”, which is the tested DC output of the module under standard test conditions.There are inverter losses, cell aging, shading, and other factors that reduce the actual output by 15-25%.
The solar thermal capacity factor includes 6 hours of energy storage.Unlike PV, where the capacity factor is essentially fixed by theresource availability, solar thermal capacity factors can be”engineered” by varying the amount of energy storage and the “solarmultiple” of the field (that is, by changing the ratio of the size ofthe collector area to the size of the generator). 26% would beconsistent with a non-storage system in a good resource area.