Costs of electricity generation by technology

I've put together life-cycle cost analyses of different power sources. There are several different studies, so this is a meta-survey.

These are all-in costs: they include (almost) everything - construction, operation, maintenance, fuel, decommissioning - amortized over the lifespans of the plant. There are some inconsistencies: for instance, the EIA figures also add transmission costs (< 1 c/kWh), and different assumptions are used everywhere - different discount rates, etc. Also, there are regional cost differences beyond those which can be corrected by using PPP conversions. So comparing absolute values across studies is probably not meaningful.

The studies come from across time and space. Right now it has what I think is the most sensible conversions: real 2009 US dollars, inflation- and PPP- adjusted. Any economists' input would be very welcome.

One more caveat: these are generating costs - they do not include external costs, including environmental ones (pollutants, CO2 emissions) or electric grid ones (load-balancing). For instance, natural gas backup would add ~3 c/kWh to wind farms, according to the RAENG study (I do not include this in the table). It would be nice to have cost analyses of the major proposals - things like continent-wide HVDC networks, molten-salt heat storage, chemical batteries, hydrogen fuel cells. These may even be in the studies I've linked - they are huge documents, I could easily have missed them.

(I do not consider hydroelectricity or pumped-water storage to qualify, as they have too limited geographic potential. Actually several of the energy sources I list don't have much theoretical potential either (wave power), but I include them anyway.)

This is a living document. By which I mean, it's horrible and will need lots of revisions. In many cases my interpretation of which figure is the relevant one for the table may be wrong (FOAK or NOAK? 5% or 10% discount? Supercritical or subcritical steam? Leave the outlier in or throw it out? 2006 figure or 2010 projection (from a 2006 study)? etc., etc.) Continuing - I haven't included the coal and gas figures from the CEC analysis, because I couldn't figure out which was which. And I may have misclassified some of the "wind" resources as onshore or "coal" resources as conventional pulverized-fuel (PF), which were my defaults when there was no further specification. The EIA reference isn't acceptable - I linked to a secondhand compilation by Next Big Future (a prolific blogger), but I need to find the same figures in the original (gigantic) EIA report. And this table is terribly incomplete - I left out some very good cost studies from not having heard of them.

Any advice, corrections, suggestions are VERY welcome! Please, help me!

Color key (all figures in 2009 US dollar cents (PPP) per kWh)

< 4 c/kWh 4 - 5 5 - 6 6 - 7 7 - 8 8 - 9 9 - 10 10 - 11 11 - 12 12 - 13 13+

IEA

EIA

CSIRO

DTI

CEC

RAENG

MIT-N

MIT-C

ANSTO

MIT-G

UChicago

Biomass

11.0 8.0 14.6

Coal (IGCC)

3.5 - 3.7 7.7 - 13.5 5.7 5.3

Coal (PF)

1.8 - 6.6 9.7 2.6 - 4.4 4.5 6.2 4.8 - 5.0 3.6

Coal+CCS

12.6 5.4 - 5.5 6.7

Geothermal

3.1 11.4 8.1 6.9 - 9.4 3.7 - 9.4

Hydro (large)

5.2 2.7 10.2

Hydro (river)

4.6 - 17.2

Hydro (small)

4.7 - 9.6 11.7 11.5 8.9 - 17.5

Natural gas

4.5 - 7.0 8.2 3.5 - 4.7 3.9 6.5 3.9

Nuclear

2.7 - 5.5 11.0 6.4 6.2 6.9 - 10.2 4.1 8.4 3.3 3.5 - 5.2

Solar conc. PV

15.9 - 30.2

Solar PV

13.9 - 56.0 40.6 103.2 26.3 - 62.4

Solar thermal

19.1 27.0 17.8 15.9 - 30.2

Tidal

29.4

Wave

32.3 63.1 - 123.2 11.8

Wind (offshore)

5.8 - 10.9 23.5 12.0 9.8

Wind (onshore)

3.6 - 10.7 14.5 5.7 10.1 6.2 - 10.2 6.6

The conversions I used:

Consumer Price Index (inflation)

2001 2002 2003 2004 2005 2006 2007 2008 2009
177.1 179.9 184.0 188.9 195.3 201.6 207.3 215.3 212.6

PPP factors

Countrycurrency conversionPPP factor
USAnone100
UK1 GBP = 1.6332 USD108
Australia1 AUD = 0.7976 USD97

Some acronyms

CCScarbon capture and storage
conc. PVconcentrating photovoltaic
IGCCintegrated gasification combined cycle
kWhkilowatt-hour
PFpulverized fuel
PVphotovoltaic
Sources

US Department of Labor | Consumer Price Index (CPI)

OECD | Purchasing Power Parities Data (PPP)

Google Finance | Currency Converter

[IEA] (5% only) OECD International Energy Agency | Projected Costs of Generating Electricity (2005 update) (summaries pp. 53, 63, 74)

[EIA] US Energy Information Administration | Annual Energy Outlook 2009 (compiled by Next Big Future blog)

[CSIRO] Australian Commonwealth Scientific and Research Organization | The heat is on: the future of energy in Australia (summary p. 44)

[CEC] California Energy Commission | Comparative Costs of California Central Station Electricity Generation Technologies (summary p. 14)

[RAENG] Royal Academy of Engineering & PB Power | Costs of Generating Electricity Report (summary pp. 9-10)

[DTI] UK Department of Trade & Industry (?) | Energy white paper: meeting the energy challenge
in particular the subsections
(PDF) Impact of banding the renewables obligation: costs of electricity production (summary table p. 6)
(PDF) Nuclear power generation cost benefit analysis (p. 4)

[MIT-N] MIT | The Future of Nuclear Power: an interdisciplinary MIT study (2009 update) (p. 7)

[MIT-C] MIT | The Future of Coal: an interdisciplinary MIT study (pp. 35, 46)

[ANSTO] Australian Nuclear Science and Technology Organisation | Introducing nuclear power to Australia - an economic comparison (summary table p. 58)

[MIT-G] MIT | The Future of Geothermal Energy - The Future of Impact of Enhanced Geothermal Systems (EGS) (p. 308)

[UChicago] University of Chicago | The Economic Future of Nuclear Power (p. 14)

10 comments:

  1. Nice chart.

    One thing though. You split hydro into three classes. You should provide an output range as run-of-the-river instalations come in quite a range.

    ReplyDelete
  2. Very interesting chart bringing a wide variety of information sources together in one location. Dramatically illustrates the differences in results depending on input assumptions.

    Since natural gas only has one category, am I correct in assuming that it does not include any capture of the emitted CO2? Does it include any allowance for purchasing CO2 emissions allowances?

    ReplyDelete
  3. Wow, tare better at this than I am. Would you consider cross posting to Nuclear Green?

    ReplyDelete
  4. Good work! A strong indication of the distance solar has to travel to become competitive in the grid supply market (and ocean power too).

    Two suggestions:

    Format: use white letters on the dark backgrounds (mostly the blue ones).

    References: put a date (year) next to each.

    ReplyDelete
  5. Have you considered the cost of fuel at any given time? Particularly natural gas. It's viability as an alternative (and it's influence on coal particularly) varies depending on the current gas prices/contracts. Power companies will often own coal and gas units, and vary the load based on fuel prices / demand.

    ReplyDelete
  6. You calculated per kWh. What uptimes did you use? Were those uptimes normalized across all the studies? It'd be more useful to calculate the figures per kW (e.g. the magic number to make renewables mass viable is around $2000/kW) instead.

    Similarly, were all of these based on the relevant capacity factors for each technology? Were those really the same across all studies (I doubt it because they are calculated by sampling actual plants and targets differ and the sample size is in most studies 1 or 2. This is also why the numbers differ so much among studies). The same holds for technology maturity adjustments. The EIA uses these. Others don't.

    I can help. I am working on a similar study right now (using only EIA data), but would need to know more info on how this was calculated. Without normalizing all the technical inputs across all studies, the results are unfortunately not very meaningful...

    ReplyDelete
  7. No, cost per kW is virtually meaningless. Energy is sold in units of kWh (i.e., in units of energy) not in kW.

    For example, I can purchase a 6 kW diesel generator for about $1000 -- or a price of about $170/kW -- but I'm not going to run the grid off of it, nor is the cost of the electricity that it produces competitive with the cost of electricity from the grid in most places.

    ReplyDelete
  8. Are you planning on updating this? The CEC has an updated Comparative costs of CA Central Station Electricity Generation. http://www.energy.ca.gov/2009publications/CEC-200-2009-017/CEC-200-2009-017-SF.PDF

    ReplyDelete
  9. It would also be valuable to make sure a date is included for each source, maybe sort by date. Otherwise the comparison of studies is almost meaningless. Thank you.

    ReplyDelete