Finding the Cheapest Clean Power Options

By Thomas R. Casten and Jeffrey A. Smith

December 2009

I. Assessing the Options

Congress is debating measures to combat climate change, but most discussions focus on subsidizing technologies rather than reducing greenhouse-gas emissions. Clean energy advocates want the government to support wind turbines and solar collectors. Nuclear proponents seek the same for reactors, while the coal industry argues it could capture and sequester carbon with sufficient public-sector support. Missing from the debate is factual analysis of the cost effectiveness of which clean technologies offer the most greenhouse-gas reductions at the least cost.

Rather than respond to the subsidy requests from lobbyists, lawmakers should evaluate which clean-energy approaches provide more greenhouse-gas reductions per dollar of taxpayer support. Which approaches, in fact, are both cheaper and cleaner? With such information, Congress could modernize energy and environmental regulations, and encourage entrepreneurial investments in low cost clean energy generation.

The overall goal of climate legislation, of course, should be to reduce greenhouse-gas emissions. Given that generation of heat and power accounts for more than two-thirds of U.S. emissions, we examine the economics of fourteen cleaner generation options that already are in commercial service. To understand societal impacts, we ignored subsidies and included the full costs of delivering electricity, which varies by generation type. We included the expenses associated with grid investments, line losses, and system reserve requirements. We did not include the health costs—variously estimated to be $60 to $120 per delivered megawatt-hour—associated with electricity generation, from coal.

The 14 clean generation options include seven that use renewable energy including solar photovoltaics (Solar PC), biomass electricity only (Biomass Elec.), concentrating solar electricity generation (Conc. Solar), off-shore and on-shore wind, the same geothermal electricity only (Geothermal) and biomass-fueled combined heat and power (Biomass CHP). Four conventional central generation options include integrated gasification combined cycle (IGCC), same with 50% carbon capture and sequestration (Coal w/Seq.), nuclear and natural gas combined cycle gas turbine (NG CCGT). Finally, four approaches that use energy twice—recycling otherwise-wasted energy from electricity generation for thermal use or recycling industrial waste energy into heat and power. These energy recycling approaches include fossil-fueled combined heat and power (CHP); recycling industrial waste energy to make electricity only (RE-Elec.); recycling industrial waste energy to produce both heat and power (RE-CHP) and biomass-fired combined heat and power (Biomass CHP), which is also renewable. The analysis is limited to currently commercial technologies. Existing options, of course, will improve their value propositions over time, and entrepreneurs will develop new options, including wave power,1 ocean thermal differences,2 and off-shore geothermal.3 Nonetheless, there may be useful insights from analyzing today’s commercial clean energy options.

Figure 1 compares each option’s cost per megawatt-hour of delivered electricity. The options are color coded into the three broad categories. Delivered prices are compared to the 2008 average U.S. retail price of electricity ($98 per megawatt-hour or 9.8 cents per kilowatt-hour) and to the price of delivered electricity from existing coal-fired power plants.

Figure 1: Clean Electricity Generation Options: Cost / Delivered MWh



Not surprisingly, delivered electricity from most of these cleaner generation options costs more than 2008 average costs. Coal, despite its image and history, is no longer cheap. Although the fuel itself remains relatively inexpensive, new coal-fired power plants require expensive construction and pollution controls. New coal, nuclear, and other centralized power stations also require long-distance transmission lines, which lose 9.5 percent of their power on average and lose over 20 percent of power during system peak loads. All raise the cost of electricity.

When we look at total costs before subsidies, wind—the dominant new renewable energy, with 15,600 megawatts installed in the U.S.—is expensive. Wind power’s delivered costs are between $200 and $275 per megawatt-hour, even though a wind farm typically sells its power for roughly $90-120 per megawatt-hour (including capacity payments). The difference can be explained by taxpayer subsidies, including a ten-year $20 per megawatt-hour production tax credit or a one-time 30% investment tax credit, as well as the costs, borne by all customers, of the transmission lines to bring electricity from remote wind farms to consumers. Wind’s capital amortization per MWh is high because the turbines operate only 30-32 percent of the time since they cannot spin in low winds and must shut down in high winds. Wind also requires backup generation. Utilities with growing wind capacity operate gas turbines at part load (very low efficiency), in order to have instant capacity when the wind dies down. The Bonneville Power Administration recently experienced a swing of 890 megawatts, 17%—20% of load, over a 30-minute period because the wind suddenly stalled across a wide area. A Carnegie Mellon Electric Industry Center study found that some wind generation actually increases greenhouse gases because of the carbon emissions from these spinning reserve plants.4

Solar photovoltaic costs, not surprisingly, exceed all other clean generation options, largely because of high installed capital cost ($6,000- $9000 per kilowatt) and low load factor. While each kilowatt of fueled generation can theoretically generate nearly 8200-8500 kilowatt-hours per year, a kilowatt of solar collector capacity will only generate 1750 kWh—20% of the year because of clouds, nightfall, and shifts of the sun from east to west throughout the day. Concentrated solar projects appear to have a lower capital cost ($5,000/kW), but this somewhat speculative given the lack of commercial deployments of these plants.

The cheapest clean generation alternatives use energy twice, instead of burning fuel to generate electricity or make a product and burn more fuel to make thermal energy. Using fuel twice by recycling waste energy from factories or from electricity generation can profitably replace existing coal plants, reducing carbon-dioxide emissions, and reducing societal costs.

II. Costs per Ton of CO2 Reduction

Another way to rank cleaner generation options is to attribute all costs or savings relative to the average cost of power to carbon reduction. Figure 2 shows a wide range of societal savings/costs per ton of CO2. Recall that the costs of delivered power are compared to the average for 2008 and the avoided carbon is the difference in carbon emissions relative to an existing coal plant.

Figure 2: Cost per ton of CO2 reduction vs. Existing Coal


Several points emerge from the figure:

III. Saving One Wedge of CO2

Yet another ranking compares each option’s cost of avoiding a “wedge” of carbon reductions as outlined by Robert Socolow and Stephen Pacala, Co-Directors of The Carbon Mitigation Initiative at Princeton University. In order to keep atmospheric concentrations below 500 parts per million of CO2 equivalent, they divided the problem into stabilization wedges of 25 billion tons of CO2 equivalent over 50 years, and suggested that the world must reduce emissions by seven wedges versus business as usual. Although the evidence since 2004 suggests a need for even lower greenhouse gas concentrations, the wedge remains a useful climate mitigation concept.

Figure 3 depicts the net present value5 of deploying enough of any option to avoid one wedge. Avoiding one wedge with the nine electricity-only options, whether conventional or renewable (except geothermal) has a net present cost to society that ranges from $400 billion to $2.7 trillion per wedge. By contrast, the four options that use energy twice can eliminate one wedge and create net present savings to society of $200 to $700 billion per wedge. Avoiding one wedge with solar will cost $3.4 trillion with recycled energy CHP.

Figure 3: Net present value / Net present cost per wedge of CO2 avoidance (assuming $100/ton CO2 Cost)


We calculate that building 47,000 megawatts of new energy recycling capacity (using waste heat to make heat and power) at factories would eliminate one entire wedge and save $650 billion in present value. Similarly, recycling the byproduct heat from 63,000 megawatts of new CHP generation would displace one wedge and save $500 billion.

How much new capacity is needed to provide one wedge of carbon savings? For starters: replacing 100% of present U.S. coal-fired generation with new coal gasification plants that do not sequester CO2 would avoid less than one wedge—22.5 billion tons over 50 years. The other thirteen clean options would avoid one wedge if they replaced between 8% and 62% of U.S. generation. These alternatives are summed up in Figure 4.

Figure 4: Percent of total U.S. generation to avoid one wedge of CO2


Recycled Energy: A study for the U.S. Environmental Protection Agency6 identified 64,000 megawatts of recycled energy potential from 16 industries. Installing three-fourths of this potential—47,400 megawatts—would avoid one wedge.

Recycled Energy Electricity Only: Installing all 64,000 megawatts of the EPA identified recycled energy potential, but throwing away the byproduct heat, would avoid one wedge. If we assume the recycling potential is met with 32 gigawatts of recycled energy with CHP and 32 gigawatts of recycled energy electricity only, the U.S. could avoid 1.1 wedges of CO2.

Fossil Fueled CHP: A Department of Energy study7 identified 135 to 150 gigawatts of fossil fueled CHP potential. Installing all of this potential as a mix of biomass and fossil fueled would avoid 1.65 wedges with new CHP.

Biomass Electric Only: The U.S. annual wood supply, without thinning protected forests or building new roads, has been estimated at 368 million dry tons per year,8 sufficient to power 1.47 wedges of biomass CHP.

Geothermal: It would take 76 gigawatts of new geothermal capacity to avoid one wedge, compared to 3.2 gigawatts of installed geothermal capacity in 200, an unlikely 23 fold increase. Enormous geothermal resources have been identified off shore where there are rifts in the earth’s plate, but there are no demonstrations yet built and no solid estimates of costs. We assume geothermal could supply 20% of one wedge.

IV. Policy Options

By deploying only the identified potential for the profitable clean generation options, the U.S. could avoid three of the seven wedges the world needs to avoid. The required investment of $490 billion would produce a net present value of $1.7 trillion after full capital recovery, about $200 million societal savings per year. The resulting generation would provide savings of $5 to $70 per ton of avoided CO2, before accounting for any cost or value of reducing carbon emissions. Unfortunately, this is not the path of current conventional wisdom.

Most current clean energy policy focuses on renewables only, especially wind turbines. To induce deployment of enough wind to avoid two wedges will require about $1.3 trillion of net present cost, the equivalent of $96 per ton of avoided CO2.

The nuclear option, assuming no further controversy, would require an investment of $1 trillion to avoid two wedges, and it carries an equivalent cost of $81 per ton of avoided carbon.

Building 153 gigawatts (half of today’s coal fleet) of coal-fired central generation that sequesters half of its CO2 emissions would avoid one wedge of carbon. This would require $1.1 trillion of capital investment and cost society $170 per ton of avoided carbon. The net present cost of this option is $1.3 trillion.

These findings will undoubtedly generate skepticism, in part because the options that use energy twice—that recycle waste energy—are not yet a part of the public awareness, even though 10,000 megawatts of such capacity have been in operation for years. No credible comparable study incorporates the costs of delivery and all subsidies.

Speculation about why policies favor high-cost low-carbon generation options could fill a book. Vested interests? Lack of knowledge? Industry lobbying for conventional approaches? Backlash from removing old subsidies? Cost-plus regulatory mentality? Regardless of reasons, the data show is that efficient generation that uses energy twice is largely ignored. While all other generation, both clean and dirty, receives large subsidies, energy recycling is ignored.

To compare subsidies, we calculated the expected production of electricity over 20 years of each option from installing one megawatt of capacity. We then determined the total societal subsidies—tax credits, production tax credits, and in the case of coal, payment for health and environmental damages to that one megawatt plant over the same 20 years of production. Our conclusions are shown in Figure 5.

Figure 5: U.S. subsidy per lifetime megawatt-hour of production for each option.


Although all renewable and all recycled energy generation is pristine—using no incremental fossil fuel and emitting no incremental carbon, the range of subsidies is huge.

Recall that the average price for one MWh in 2008 was $98. The U.S. currently provides a $54/MWh subsidy to all power from a solar collector. Wind and geothermal receive about $10 per lifetime MWh. Biomass receives about $5 per lifetime MWh of subsidy.

By contrast, CHP and recycled energy—the options that use energy twice—received no subsidy prior to 2009, and are now eligible for $1.34/lifetime MWh.

Why the difference? One argument is that renewables are infant technologies and will improve the value proposition over time. However, using energy twice would also improve its value proposition with wide deployment. Are the renewable lobbyists more effective? Or are these disparities a reflection of cost-plus mentality and a conviction that we need all options to mitigate climate change?

The even more surprising finding is the lifetime subsidies to conventional coal that dwarf subsidies to any clean generation. A Harvard study9 found that each MWh of conventional coal-fired generation causes $60 of health and environmental costs, none of which are charged to the power. A comparable study by the Ontario Medical Association10 put the health and environmental costs at $120 per MWh.

Consider these disparities between subsidies of various clean energy options in light of the fact that 40% of U.S. CO2 comes from electricity generation. How are these policies consistent with mitigating climate change? One is tempted to agree with the great American philosopher Pogo who said, “We have met the energy and he is us.”

V. Why Does Policy Ignore Options That Use Energy Twice?

Actual policies give top encouragement to high-cost, high carbon coal, strongly encourage high-cost, low-carbon options, and virtually ignore low-cost, low-carbon options.

The matrix shown in Figure 6 places the generation options into four quadrants—high and low cost and high and low carbon.

Figure 6: Generation Options—The Four Quadrants

One would think policy should aim towards low-cost, low-carbon options first and only then focus on higher cost low-carbon options. Sensible policy would never encourage high-cost, high-carbon options. But look at actual policy, which is the remnant of one-hundred years of regulation of electricity. Many believe existing coal is low cost, based on the prices charged to consumers. However, these prices exclude the $60 to $120 per MWh that society pays for the health and environmental impacts of coal-based generation. Coal is a high coat option to society. Policy perversely favors coal, the high-cost, high carbon option.

This study’s directional conclusions prompt several questions. Can society afford policies that favor costly over profitable options to reduce CO2? Is it sensible to make political choices among generation options instead of setting environmental goals and letting markets choose technologies? Should we not first install clean generation that also reduces the costs of delivered energy?

We hope this study will encourage modernized policies that foster level competition among all cleaner generation options. No one can predict how each present option will improve or what cleaner-generation alternatives will emerge. Still, these findings show the potential societal value of modernizing regulations and eliminating subsidies that distort market. By focusing on the generation technologies that offer the most greenhouse-gas reductions at the lowest cost, the U.S. can provide a model for other countries to profitably reduce their own CO2 emissions

Appendix A: Notes on Data

Relevant data for each option were taken from current Department of Energy/Energy Information Administration information where available and from authors experience in recycled energy plants. Data include capital costs, generation efficiency, likely hours of annual operation, line losses, cost of transmission and distribution capital, need for system reserves and spinning reserves and cost of capital.

We then calculated a variety of metrics for each cleaner generation option, all compared to producing the same power with an existing coal plant that complies with 2009 environmental rules for criteria pollutants (NOx, SO2, particulates, and mercury).

Notes on Assumptions Associated with the Delivered Costs of Power

Many factors go into the delivered cost of a megawatt-hour of power and the costs vary widely among the options. Here are the main assumptions used:

Required return on assets or ROA was assumed to be 10% for generation options that are typically included in monopoly utility rate base and thus enjoy guaranteed financial returns performance. We assumed that all local or distributed generation, which is not typically in rate base, would require an extra 200 to 500 basis points—to 12% to 15% ROA. This differential return on capital is a distortion of the current regulatory system.

New transmission and distribution was assumed to cost $1500 per kW of capacity. This estimate was extrapolated from 1999 studies.11

Line losses due to transmission, distribution, and congestion averaged 9.5% in the U.S.12 We assumed this average for central generation that operates 24/7 as base load—(coal and nuclear) or operates randomly (wind and concentrated solar). For natural gas-fired generation, which operated only 24% of the time in 2008, typically during high system loads, we assumed 150% of system average line losses. For local generation, where electricity flows to the hosts, regardless of financial arrangements, we assumed 1.5% line losses.

System reserve requirements were assumed to be 18% for all central generation options, which is the typical target of the National Electric Reliability Council. Local generation, because of its much smaller size and large volume of generation plants, benefits from the law of large numbers and was assumed to need only 5% system reserve requirement, per studies conducted at Carnegie Mellon Electric Industry Center.

Capital costs, operations and maintenance costs and generator efficiency levels (heat rates) were taken from the current DOE/EIA analysis and from the authors’ experience in combined heat and power and industrial recycled energy.

Delivered fuel costs were based on DOE/EIA data for 2008 including natural gas at $5.00 per MMBtu, coal at$1.29/ MMBtu, plus $.40 per ton for ash handling and disposal

Endnotes

1. www.ucsusa.org/publications/catalyst/how-it-works-wave-power.html

2. P. McKenna, “Plumbing the Oceans Could Bring Limitless Energy.” New Scientist, Nov 19, 2008.

3. T. Schuenemal, “Geothermal Sources Could Add Significant Power Generation Capacity.” CleanTechnica, in alternative energy, technology. Oct 8, 2008

4. Zerriffi, Hisham, “Electric power systems under stress; an evaluation of centralized vs distributed systems architectures”, PhD, Carnegie-Mellon University, 2004.

5. To calculate the Net Present Value, we projected savings or costs over 50 years and then discounted each year by 12%. One trillion of Net Present Costs results from a cost of $120 million per year for fifth years of $6 trillion of total costs to society.

6. Owen Bailey and Ernst Worrell, 2005, “Clean Energy Technologies: A Preliminary Inventory of the Potential for Electricity Generation.” LBNL-57451, Lawrence Berkeley National Laboratory, Berkeley, CA and M. Lowe and G. Gereffi, “Manufacturing Climate Solutions, Carbon Reducing Technologies and U.S. Jobs,” Chapter 7, Recycling Industrial Waste Energy, Feb 26, 2009, Center on Globalization, Governance & Competitiveness, Duke University, http://www.cggc.duke.edu/environment/climatesolutions

7. B. Hedman, Combined Heat & Power and Heat Recovery as Energy Efficiency Options, Briefing to Senate Renewable Energy Caucus, Sept. 10 2007; Energy and Environmental Analysis/USCHPA, Washington DC. This work was done under contract to the US DOE, in support of their goal to deploy 90 GW of CHP by 2010.

8. R .D. Perlack et al., Biomass as a Feedstock for Bioenergy and Bioproducts Industry [U.S. Department of Agriculture/Department of Energy (DOE), ORNL TM-2005/66, Oak Ridge, TN 2005].

9. Levy and Spengler of Harvard School of Public Health & Hlinka and Sullivan of Sullivan Environmental Consulting, Inc., “Estimated Public Health Impacts of Criteria Pollutant Air Emissions from the Salem Harbor and Brayton Point Power Plants”, May, 2000. This study was commissioned by the Clean Air Task Force and prepared with support from the Pew Charitable Trusts.

10. “Cost Benefit Analysis: Replacing Ontario’s Coal-Fired Electricity Generation” by DSS Management Consultants Inc & RWDI Air Inc for Ontario Ministry of Energy, April 2005, http://www.energy.gov.on.ca/pdf/electricity/coal_cost_benefit_analysis_ap

11. Arthur D. Little and Company for DOE.

12. US Department of Energy- Office of Electric Transmission & Distribution, Page 5, July 2003, http://www.oe.energy.gov/DocumentsandMedia/Electric_Vision_Document.pdf

in the newsroom

follow RED on: