Power Plants: Characteristics and Costs

Contents

• Introduction and Organization
• Types of Generating Technologies
• Electricity Demand and Power Plant Choice and Operation
• Generation and Load
• Economic Dispatch and Heat Rate
• Capacity Factor
• Utility Scale Generating Technologies
• Supercritical Pulverized Coal
• Integrated Gasification Combined Cycle (IGCC)
• Natural Gas Combined Cycle
• Nuclear Power
• Geothermal Power
• Wind Power
• Solar Thermal and Solar Photovoltaic (PV) Power
• Factors that Drive Power Plant Costs
• Government Incentives
• Renewable Energy Production Tax Credit
• Nuclear energy production tax credit
• Loan Guarantees for Nuclear and Other Carbon-Control Technologies
• Energy Investment Tax Credit
• Clean Coal Technologies Investment Tax Credit
• State and Local Incentives
• Capital and Financing Costs
• Construction Cost Components and Trends
• Financing Power Plant Projects
• Fuel Costs
• Air Emissions Controls for Coal and Gas Plants
• Conventional Emissions
• Carbon Dioxide
• Financial Analysis Methodology and Key Assumptions
• Analysis of Power Project Costs
• Case 1: Base Case
• Key Observations
• Discussion
• Case 2: Influence of Federal and State Incentives
• Key Observations
• Discussion
• Case 3: Higher Natural Gas Prices
• Key Observations
• Discussion
• Case 4: Uncertainty in Capital Costs
• Key Observations
• Discussion
• Case 5: Carbon Controls and Costs
• Key Observations
• Discussion

Summary

This report analyzes the factors that determine the cost of electricity from new power plants. These factors—including construction costs, fuel expense, environmental regulations, and financing costs—can all be affected by government energy, environmental, and economic policies. Government decisions to influence, or not influence, these factors can largely determine the kind of power plants that are built in the future. For example, government policies aimed at reducing the cost of constructing power plants could especially benefit nuclear plants, which are costly to build. Policies that reduce the cost of fossil fuels could benefit natural gas plants, which are inexpensive to build but rely on an expensive fuel.

The report provides projections of the possible cost of power from new fossil, nuclear, and renewable plants built in 2015, illustrating how different assumptions, such as for the availability of federal incentives, change the cost rankings of the technologies.

None of the projections is intended to be a "most likely" case. Future uncertainties preclude firm forecasts. The rankings of the technologies by cost are therefore also an approximation and should not be viewed as definitive estimates of the relative cost-competitiveness of each option. The value of the discussion is not as a source of point estimates of future power costs, but as a source of insight into the factors that can determine future outcomes, including factors that can be influenced by the Congress.

Key observations include the following:

• Government incentives can change the relative costs of the generating technologies. For example, federal loan guarantees can turn nuclear power from a high cost technology to a relatively low cost option.
• The natural gas-fired combined cycle power plant, the most commonly built type of large natural gas plant, is a competitive generating technology under a wide variety of assumptions for fuel price, construction cost, government incentives, and carbon controls. This raises the possibility that power plant developers will continue to follow the pattern of the 1990s and rely heavily on natural gas plants to meet the need for new generating capacity.
• With current technology, coal-fired power plants using carbon capture equipment are an expensive source of electricity in a carbon control case. Other power sources, such as wind, nuclear, geothermal, and the natural gas combined cycle without capture technology currently appear to be more economical.

---

Power Plants: Characteristics and Costs

Introduction and Organization

The United States may have to build many new power plants to meet growing demand for electric power. For example, the Energy Information Administration (EIA) estimates that the nation will have to construct 226,000 megawatts of new electric power generating capacity by 2030.¹ This is the equivalent of about 450 large power plants. Whatever the number of plants actually built, different combinations of fossil, nuclear, or renewable plants could be built to meet the demand for new generating capacity. Congress can largely determine which kinds of plants are actually built through energy, environmental, and economic policies that influence power plant costs.

This report analyzes the factors that determine the cost of electricity from new power plants. These factors—including construction costs, fuel expense, environmental regulations, and financing costs—can all be affected by government energy and economic policies. Government decisions to influence, or not influence, these factors can largely determine the kind of power plants that are built in the future. For example, government policies aimed at reducing the cost of constructing power plants could especially benefit nuclear plants, which are costly to build. Policies that reduce the cost of fossil fuels could benefit natural gas plants, which are inexpensive to build but rely on an expensive fuel.

The report provides projections of the possible cost of power for new fossil, nuclear, and renewable plants built in 2015. The projections illustrate how different assumptions, such as for the availability of federal incentives, change the cost rankings of the technologies. Key observations include the following:

• Government incentives can change the relative costs of the generating technologies. For example, federal loan guarantees can turn nuclear power from a high cost technology to a relatively low cost option.
• The natural gas-fired combined cycle power plant, the most commonly built type of large natural gas plant, is a competitive generating technology under a wide variety of assumptions for fuel price, construction cost, government incentives, and carbon controls. This raises the possibility that power plant developers will continue to follow the pattern of the 1990s and rely heavily on natural gas plants to meet the need for new power generation.
• With current technology, coal-fired power plants using carbon capture equipment are an expensive source of electricity in a carbon control case. Other power sources, such as wind, nuclear, geothermal, and the natural gas combined cycle plant without capture technology, currently appear to be more economical.

None of the projections is intended to be a "most likely" case. Future uncertainties preclude firm forecasts. The value of this discussion is not as a source of point estimates of future power costs, but as a source of insight into the factors that can determine future outcomes, including factors that can be influenced by the Congress.

The main body of report is divided into the following sections:

• Types of generating technologies;
• Factors that drive power plant costs;
• Financial analysis methodology;
• Analysis of power project costs.

The report also includes the following appendixes:

• Appendix A presents power generation technology process diagrams and images.
• Appendix B and Appendix C provide the data supporting the capital cost estimates used in the economic analysis. Appendix C also shows how operating costs and plant efficiencies were estimated for certain carbon control technologies.
• Appendix D presents the financial and operating assumptions used in the power cost estimates.
• Appendix E is a list of acronyms used in the report.

Types of Generating Technologies

The first part of this section describes how the characteristics of electricity demand influence power plant choice and operation. The next part describes the generating technologies analyzed in the report.

Electricity Demand and Power Plant Choice and Operation

Generation and Load

The demand for electricity ("load") faced by an electric power system varies moment to moment with changes in business and residential activity and the weather. Load begins growing in the morning as people waken, peaks in the early afternoon, and bottoms-out in the late evening and early morning. Figure 1 is an illustrative daily load curve.

The daily load shape dictates how electric power systems are operated. As shown in Figure 1, there is a minimum demand for electricity that occurs throughout the day. This base level of demand is met with "baseload" generating units which have low variable operating costs.² Baseload units can also meet some of the demand above the base, and can reduce output when demand is unusually low. The units do this by "ramping" generation up and down to meet fluctuations in demand.

The greater part of the daily up and down swings in demand are met with "intermediate" units (also referred to as load-following or cycling units). These units can quickly change their output to match the change in demand (that is, they have a fast "ramp rate"). Load-following plants can also serve as "spinning reserve" units that are running but not putting power on the grid, and are immediately available to meet unanticipated increases in load or to back up other units that go off-line due to breakdowns.

Figure 1. Illustrative Load Curve

The highest daily loads are met with peaking units. These units are typically the most expensive to operate, but can quickly startup and shutdown to meet brief peaks in demand. Peaking units also serve as spinning reserve, and as "quick start" units able to go from shutdown to full load in minutes. A peaking unit typically operates for only a few hundred hours a year.

Economic Dispatch and Heat Rate

The generating units available to meet system load are "dispatched" (put on-line) in order of lowest variable cost. This is referred to as the "economic dispatch" of a power system's plants.

For a plant that uses combustible fuels (such as coal or natural gas) a key driver of variable costs is the efficiency with which the plant converts fuel to electricity, as measured by the plant's "heat rate." This is the fuel input in British Thermal Units (btus) needed to produce one kilowatt-hour of electricity output. A lower heat rate equates with greater efficiency and lower variable costs. Other things (most importantly, fuel and environmental compliance costs) being equal, the lower a plant's heat rate, the higher it will stand in the economic dispatch priority order. Heat rates are inapplicable to plants that do not use combustible fuels, such as nuclear and non-biomass renewable plants.

As an illustration of economic dispatch, consider a utility system with coal, nuclear, geothermal, natural gas combined cycle, and natural gas peaking units in its system:

• Nuclear, coal, and geothermal baseload units, which are expensive to build but have low fuel costs and therefore low variable costs, will be the first units to be put on line. Other than for planned and forced maintenance, these baseload generators will run throughout the year.
• Combined cycle units, which are very efficient but use expensive natural gas as a fuel, will meet intermediate load. These cycling plants will ramp up and down during the day, and will be turned on and off dozens of times a year.
• Peaking plants, using combustion turbines,³ are relatively inefficient and burn expensive natural gas. They run only as needed to meet the highest loads.⁴

An exception to this straightforward economic dispatch are "variable renewable" power plants—wind and solar—that do not fall neatly into the categories of baseload, intermediate, and peaking plants. Variable renewable generation is used as available to meet demand. Because these resources have very low variable costs they are ideally used to displace generation from gas-fired combined cycle plants and peaking units with higher variable costs. However, if wind or solar generation is available when demand is low (such as a weekend or, in the case of wind, in the evening), the renewable output could displace coal generation.

Power systems must meet all firm loads at all times, but variable renewable plants do not have firm levels of output because they are dependent on the weather. They are not firm resources because there is no guarantee that the plant can generate at a specific load level at a given point in time.⁵ Variable renewable generation can be made firm by linking wind and solar plants to electricity storage, but with current technology, storage options are limited and expensive.⁶

Capacity Factor

As discussed above, baseload units run more often than cycling units, and peaking units operate the least often. The utilization of a generating unit is measured by its "capacity factor." This is the ratio of the amount of power generated by a unit for a period of time (typically a year) to the maximum amount of power the unit could have generated if it operated at full output, non-stop. For example, the maximum amount of power a 1,000 megawatt (MW) unit can generate in a year is 8.76 million megawatt-hours (Mwh), calculated as:

1,000 MW x 8,760 hours in a year = 8.76 million Mwh.

If this unit actually produced only 4.0 million Mwh its capacity factor would be 46% (calculated as 4.0 million Mwh divided by 8.76 million Mwh).

Note in this calculation the distinction between capacity and energy. Capacity is the potential instantaneous output of a generating unit, measured in watts.⁷ Energy is the actual amount of electricity generated by a power plant during a time period, measured in watt-hours. The units are usually expressed in thousands (kilowatts and kilowatt-hours) or millions (megawatts and megawatt-hours).

The difference between actual and theoretical maximum output is caused by planned maintenance, mechanical breakdowns (forced outages), and any instances in which the plant is backed-down from maximum output due to lack of load or because the plant's power is more expensive than that from other plants. It is rare for a plant to have a capacity factor of 100%. Baseload plants typically have capacity factors of about 70% or greater, peaking plants about 25% or less, and cycling plants fall in the middle.

Utility Scale Generating Technologies

The types of generating technologies discussed in this report are often referred to as "utility scale" plants for baseload or intermediate service. These technologies generate large amounts of electricity at a single site for transmission to customers. In 2006, large baseload and intermediate service power plants accounted for about 86% of total power generation in the United States.⁸ Utility scale plants typically have generating capacities ranging from dozens to over a thousand megawatts.

The one smaller scale generating technology covered in this report is solar photovoltaic power. The capacity of the largest U.S. central station solar photovoltaic plant, at Nellis Air Force Base in Nevada, is only 14 MW. Because of their small size, high capital costs, and low utilization rates, solar photovoltaic plants built with current technology have very high electricity production costs. Central station solar photovoltaic power is nonetheless included in the cost analysis because of public interest.

The report excludes peaking plants, which play an important but small role in the power system. The report also excludes oil-fired generation, which has all but disappeared from the nation's generating mix because of the high cost of the fuel. In 1978, oil-fired plants produced 22% of the nation's electricity. By 2007 the oil-fired share was less than 2%.⁹ Significant construction of new oil-fired plants is not expected.

The report also does not cover combined heat and power (CHP) plants. These are typically industrial plants that co-produce electricity and steam for internal use and for sale. Unlike plants that generate power exclusively to put electricity on the grid, CHP facilities have unique, plant-specific operating modes and cost structures, and economics fundamentally different from utility scale generation. CHP generation is a small part of the electric power industry, accounting for about 3.7% of total electricity output in 2007.¹⁰ Hydropower is excluded because no significant construction of new, large hydroelectric plants is expected (due to environmental concerns and the small number of available sites).¹¹

The cost analysis is for plants entering service on January 1, 2015, which means construction would start soon (between 2009 and 2013 depending on the technology). The plants therefore incorporate only small projected changes from 2008 cost and performance for mature technologies, and reflect current estimates of cost and performance for new or evolving technologies (such as advanced nuclear power and coal gasification).

The technologies covered in the report are described briefly below. Process diagrams and images of each technology are in Appendix A.

Supercritical Pulverized Coal

Pulverized coal plants account for the great majority of existing and planned coal-fired generating capacity. In this system coal is ground to fine power and injected with air into a boiler where it ignites. Combustion heat is absorbed by water-carrying tubes embedded in the boiler walls and downstream of the boiler. The heat turns the water to steam, which is used to rotate a turbine and produce electricity. Since about 2000 most plans for new pulverized coal plants have been for "supercritical" designs that gain efficiency by operating at very high steam temperatures and pressures.

In 2007, coal generation of all types¹² accounted for 49% of total power generation in the United States (see Figure 2).

Figure 2. Total U.S. Electric Power Generation by Energy Source, 2007

Sources: EIA, Electric Power Monthly March 2008, Table ES1.B, and the EIA906/923 preliminary data file for 2007.

Integrated Gasification Combined Cycle (IGCC)

In this process coal is converted to a "synthesis gas" (syngas) before combustion. IGCC plants are more expensive to build than pulverized coal generation, but proponents believe they have compensating advantages, including:

• Lower emissions of air pollutants, such as sulfur dioxide (SO₂), nitrogen oxides (NOx), and mercury. However, modern pulverized coal plants also have low emissions of air pollutants, so the advantage of IGCC plants over conventional technology is limited.
• Greater efficiency (i.e., a lower heat rate), although with current technology IGCC has only a small efficiency advantage over conventional coal plants.¹³
• The syngas that results from the gasification process can be processed to convert the carbon in the gas into a concentrated stream of carbon dioxide (CO₂). The syngas can then be processed, before it is burned, to remove the CO₂.

In principle this pre-combustion capture of CO₂ can be accomplished more easily and cheaply than post-combustion removal of CO₂ from the exhaust gases ("flue gas") emitted by a conventional coal plant. The promise of more efficient carbon capture is one of the primary rationales for IGCC technology.

Coal-fired IGCC experience in the United States is limited to a handful of research and prototype plants, none of which is designed for carbon capture. A commercial IGCC plant is being constructed by Duke Energy at its Edwardsport site in Indiana, and other projects have been proposed. However, some other power plant developers will not build IGCC plants because of concerns over cost and the reliability of the technology.¹⁴ In general, the cost and operational advantages of IGCC over conventional coal technology and the commercial readiness of IGCC technology are disputed.¹⁵

Natural Gas Combined Cycle

Combined cycle plants are built around one or more combustion turbines, essentially the same technology used in jet engines. The combustion turbine is fired by natural gas to rotate a turbine and produce electricity. The hot exhaust gases from the combustion turbine are captured and used to produce steam, which drives another generator to produce more electricity. By converting the waste heat from the combustion turbine into useful electricity the combined cycle achieves very high efficiencies, with heat rates below 7,000 btus per kWh (compared to around 9,000 btus per kWh for new pulverized coal plants). This high efficiency partly compensates for the high cost of the natural gas used in these plants.

Modern combined cycle plants, which evolved in the 1990s, have a relatively low construction cost and modest environmental impacts; can be used to meet baseload, intermediate, and peaking demand; can be built quickly; and are very efficient. Because of these advantages, since 1995 natural gas combined cycle plants have accounted for 88% of the all the new generating capacity built in the United States capable of baseload and intermediate service.¹⁶

Natural gas combined cycle plants and other types of gas-fired power plants are expected to continue to dominate capacity additions into the next decade.¹⁷ According to EIA, combined cycle plants will account for 29% of all capacity additions between 2008 and 2015.¹⁸ However, this forecast may understate actual combined cycle plant additions. The EIA estimates that coal plants will account for almost a quarter of new capacity built through 2015, the equivalent of about 170 new coal-fired generating units.¹⁹ It is questionable whether this much coal capacity will actually be built because of public opposition to new coal plants and the cost of the plants. Utilities reportedly canceled 16,577 MW of planned generating capacity in 2007, of which 84% was coal-fired.²⁰ According to a Department of Energy (DOE) report, only 12% (4,500 MW) of the coal capacity planned in 2002 to be built by 2007 was actually constructed. The report notes that "delays and cancellations have been attributed to regulatory uncertainty (regarding climate change) or strained project economics due to escalating costs in the industry."²¹

If less coal capacity is built than planned, the main replacement is likely to be combined cycle plants, the type of gas-fired unit capable of replacing a baseload coal plant. For example, in 2007, power generators in Florida planned to install 4,627 MW of new coal fired capacity through 2016. By 2008 the plans for new coal-fired capacity had dropped to 738 MW, primarily "due to environmental concerns at the State level. The majority of this decrease in planned coal-fired generation was replaced with gas-fired units."²²

Natural gas combined cycle plants accounted for 17% of total generation in 2007,²³ and natural gas plants of all types accounted for 21% of total power generation in the United States (Figure 2).

Nuclear Power

Nuclear power plants use the heat produced by nuclear fission to produce steam. The steam drives a turbine to generate electricity. Nuclear plants are characterized by high investment costs but low variable operating costs, including low fuel expense. Because of the low variable costs and design factors, nuclear plants in the United States operate exclusively as baseload plants and are typically the first plants in a power system's dispatch order. Nuclear power supplied 19% of the nation's electricity in 2007 (Figure 2).

This report discusses projected costs for Generation III/III+ technology nuclear plants. These plants are more advanced versions of the 104 reactors currently operating in the United States, and all reactors currently proposed for construction in the United States are Generation III/III+ designs. Compared to existing reactors, the Gen III/III+ plants are designed to reduce costs and enhance safety through, for example, reduced complexity, standardized designs, and improved construction techniques. Some designs also incorporate passive safety systems that are supposed to be capable of preventing a catastrophic accident even without operator action.

There are several competing Gen III/III+ designs,²⁴ but only one design has been built (General Electric's Advanced Boiling Water Reactor, of which four units have been constructed in Japan). Plants based on other Gen III/III+ designs are under construction in France, Finland, and China. As discussed later in the report, the costs of building a new nuclear plant in the United States will apparently be very high.

Geothermal Power

Geothermal plants have operated for many years in the western United States, mainly in California. In a typical binary cycle geothermal facility, wells draw hot water and steam from underground into a heat exchanger. In the heat exchanger a working fluid is vaporized and used to drive a turbine generator (the underground steam is not used directly because it contains corrosive impurities and can release air pollutants). In geothermal fields that have been depleted by years of use, such as the Geysers field in California, operators can inject water into the layers of hot rock to supplement the naturally available water and boost steam production. Unlike solar and wind power, which are weather-dependent, geothermal plants operate as dispatchable baseload plants. However, with current technology, geothermal plants are limited to small facilities (typically under 50 MW) at sites in the western United States.²⁵ In 2007, geothermal plants produced 0.4% of the nation's power supply (Figure 2).²⁶

Wind Power

Wind power plants (sometimes referred to as wind farms) use wind-driven turbines to generate electricity. An individual turbine typically has a capacity in the range of 1.5 to 2.5 MW, and a wind plant installs dozens or hundreds of these turbines. As noted above, wind is a variable renewable resource because its availability depends on the vagaries of the weather. Wind supplied 1% of total U.S. power supply in 2007 (Figure 2); EIA estimates that assuming no changes to current law and regulation, this will increase to 2.4% by 2030.²⁷

Solar Thermal and Solar Photovoltaic (PV) Power

Solar thermal and PV power are alternative means of harnessing sunlight to produce electricity. PV power uses solar cells to directly convert sunlight to electricity. To date most of the solar PV installations in the United States have been small (about one MW or less). Two exceptions are the installations at Nellis Air Force Base in Nevada (14 MW) and the Alamosa Photovoltaic Power Plant in Colorado (8 MW).

Solar thermal plants, also referred to as concentrated solar power (CSP), concentrate sunlight to heat a working liquid to produce steam that drives a power-generating turbine. Two major types of solar thermal systems are parabolic trough and power tower technologies. Parabolic trough plants use an array of mirrors to focus sunlight on liquid-carrying tubes integrated with the mirrors. Several parabolic trough installations have operated successfully in California since the 1980s, and the 64 MW Nevada Solar One plant began operating in 2007.

The power tower technology uses a mirror field to focus sunlight on a central tower, where the heat is used to produce steam for power generation. A research power tower, the Solar One/Two plant, operated for several years in the 1980s and 1990s in California. A power tower plant has recently been constructed in Spain and a 400 MW project has been proposed for California.

Several new solar thermal projects, primarily of the parabolic trough and related types, are in development. The capacity of these projects range up to 554 MW. A potential advantage of solar thermal systems is the ability to produce electricity when sunlight is weak or unavailable by storing solar heat in the form of molten salt. If storage proves economical for large-scale plants, then solar thermal facilities in regions with strong, near continuous daytime sunlight, such as the Mojave desert, could be operated as dispatchable plants with firm capacity.

In 2007, solar thermal generation accounted for 0.01% of total generation, and solar PV power for less (Figure 2).

Factors that Drive Power Plant Costs

This section of the report discusses the major factors that determine the costs of building and operating power plants. These factors include:

• Government incentives.
• Capital (investment) cost, including construction costs and financing.
• Fuel costs.
• Air emissions controls for coal and natural gas plants.

Government Incentives

Many government incentives influence the cost of generating electricity. In some cases the incentives have a direct and clear influence on the cost of building or operating a power plant, such as the renewable investment tax credit. Other programs have less direct affects that are difficult to measure, such as parts of the tax code that influence the cost of producing fossil fuel.²⁸

The economic analysis in this report incorporates the following incentives that directly affect the cost of building or operating power plants.²⁹

Renewable Energy Production Tax Credit³⁰

The credit has a 2008 value of 2.0 cents per kWh, with the value indexed to inflation. The credit applies to the first 10 years of a plant's operation. As of October 2008 the credit is available to plants that enter service before the end of 2009. The credit is currently available to new wind, geothermal, and several other renewable energy sources. New solar energy projects do not qualify, and geothermal projects can take the production tax credit only if they do not use the renewable investment tax credit (discussed below).

Nuclear energy production tax credit³¹

The credit, which is for new advanced nuclear plants, has a nominal value of 1.8 cents per kWh. The credit applies to the first eight years of plant operation. Unlike the renewable production tax credit the nuclear credit is not indexed to inflation and therefore drops in real value over time. This credit is subject to several limitations:

• It is available to advanced (i.e., Gen III/III+) nuclear plants that begin construction before January 1, 2014, and enter service before January 1, 2021.
• For each project the annual credit is limited to $125 million per thousand megawatts of generating capacity.
• The full amount of the credit will be available to qualifying facilities only if the total capacity of the qualifying facilities is 6,000 megawatts or less. If the total qualifying capacity exceeds 6,000 megawatts the amount of the credit available to each plant will be prorated. EIA estimates in its 2008 Annual Energy Outlook that 8,000 megawatts of new nuclear capacity will qualify;³² in this case the credit amount would drop to 1.35 cents per kWh once all the qualifying plants are on-line. This pro-rated value is used in the report's economic analysis of generating costs.

Loan Guarantees for Nuclear and Other Carbon-Control Technologies³³

Under final Department of Energy (DOE) rules the loan guarantees can cover up to 80% of the cost of a project, and are awarded based on a detailed evaluation of each applicant project. Entities receiving loan guarantees must make a "credit subsidy cost" payment to the federal treasury that reflects the anticipated cost of the guarantee to the government, including a probability weighted cost of default. Because the debt is backed by the federal government, it is expected to carry the highest credit rating and therefore a low interest rate.³⁴ The guarantees are unavailable to publicly owned utilities, such as municipal systems.³⁵

Congress periodically determines the total value of the guarantees that the DOE is authorized to grant. In April 2008, the Department of Energy announced plans to solicit up to $18.5 billion in loan guarantee applications for nuclear projects.³⁶ As of November 2008, DOE was considering several applications for loan guarantees.

Developers and investors have stated that the loan guarantees are critical to constructing at least the first wave of new nuclear plants. This is because of the multi-billion dollar cost of a nuclear project, which can exceed the total market value of the company building a plant. For example, in 2008 the president of Exelon Generation, which operates a large fleet of existing nuclear plants and plans to build new units, stated that constructing new nuclear plants would be "impossible" without loan guarantees.³⁷

Energy Investment Tax Credit³⁸

Tax credits under this program are available to solar and geothermal electricity generation, and some other innovative energy technologies. Wind energy systems do not qualify. The credit is 10% for geothermal systems, and is 30% for solar electric systems installed before January 1, 2017 (after which it reverts to 10%). Geothermal projects that take the investment tax credit cannot claim the renewable production tax credit.³⁹ The depreciable basis of the project for tax purposes is reduced by 50% of the credit value. The investment tax credit is available to independent power producers and investor owned utilities, but is inapplicable to tax-exempt publicly owned utilities.⁴⁰

Clean Coal Technologies Investment Tax Credit⁴¹

This tax credit can be used by investor owned utilities or independent power producers (it is inapplicable to tax-exempt publicly owned utilities). It is limited to a total of $2.55 billion in tax credits, of which (1) $0.8 billion is specifically for IGCC plants; (2) $0.5 billion is for non-IGCC advanced coal technologies, and (3) $1.25 billion is for advanced coal projects generally. The tax credits in the third category will not be awarded until after the program that encompasses the first two categories of tax credits is completed or until such other date designated by the Secretary of Energy.⁴² The depreciable basis of a project for tax purposes is reduced by 50% of the credit value.

State and Local Incentives

State and local governments can offer additional incentives, such as property tax deferrals. The combined value of the government tax breaks can run into the hundreds of millions of dollars per project. For example, Duke Energy's Edwardsport IGCC project in Indiana is expected to receive almost half-a-billion dollars in federal, state, and local tax incentives.⁴³

State utility commissions can use rate treatment of new plants as a financial incentive for the investor owned utilities they regulate. Under traditional rate making a utility is not permitted to earn a return on its construction investment until a plant is in service. This approach to ratemaking is used to motivate the utility to prudently manage construction, and to ensure that customers do not have to pay for a power plant until it is operating. However, if a project is very expensive, the time lag between when costs are incurred and when return on the investment is allowed in rates can put a financial strain on the company. If the plant is expensive, adding the return into rates as a single big adjustment can inflict "rate shock" on customers.

For these reasons, utilities sometimes argue for an alternative rate making method called "construction work in progress (CWIP) in rates." In this approach, a utility is allowed to recover in rates the return on its investment as the plant is being built. CWIP in rates relieves the utility of the financial strain of carrying an expensive investment that is yielding no income, phases-in the rate increase to customers, and decreases the utility's financial exposure if the project is delayed. On the other hand, the pressures for prudent construction management inherent in traditional ratemaking are dampened.

Some states, such as South Carolina and Mississippi, have passed legislation allowing utility projects that meet certain criteria to receive CWIP in rates.⁴⁴ In other cases utilities have received CWIP in rates under existing rules. CWIP in rates has expanded beyond its historic application to very expensive coal and nuclear projects. For example, the Kansas and Wisconsin commissions have allowed CWIP in rates for relatively small wind projects.⁴⁵

Capital and Financing Costs

Construction Cost Components and Trends

Most of the generating technologies discussed in this report are capital intensive; that is, they require a large initial construction investment relative to the amount of generating capacity built. Power plant capital costs are often discussed in terms of dollars per kilowatt (kW) of generating capacity. All of the technologies considered in this report have estimated 2008 costs of $2,100 per kW or greater, with the exception of the natural gas combined cycle plant ($1,200 see Appendix B). Nuclear, geothermal, and IGCC plants have estimated costs in excess of $3,000 per kW.

Power plant capital costs have several components. Published information on plant costs often do not clearly distinguish which components are included in an estimate, or different analysts may use different definitions. The capital cost components are:

• Engineering, Procurement, and Construction (EPC) cost: this is the cost of the primary contract for building the plant. It includes the cost of designing the facility, buying the equipment and materials, and construction.⁴⁶
• Owner's costs: these are any construction costs that the owner handles outside the EPC contract. This could include arranging for the construction of transmission and fuel delivery facilities (such as a natural gas pipeline) to a power plant.
• Capitalized financing charges: a plant developer incurs financing charges while a power plant is being built. This includes interest on debt and an imputed cost of equity capital. Until the plant is operating these costs are capitalized; that is, become part of the investment cost of the project for tax, regulatory, and financial analysis purposes (see further discussion of financing costs, below).

Construction costs for power plants have escalated at an extraordinary rate since the beginning of this decade. According to one analysis, the cost of building a power plant increased by 131% between 2000 and 2008 (or by 82% if nuclear plants are excluded from the estimate). Costs reportedly increased by 69% just since 2005. The cost increases affected all types of generation. For example, between 2000 and 2008, the cost of wind capacity reportedly increased by 108%, coal increased by 78%, and gas-fired plants by 92%.⁴⁷ The cost increases have been attributed to many factors, including:

• High prices for raw and semi-finished materials, such as iron ore, steel, and cement.
• Strong worldwide demand for generating equipment. China, for example, is reportedly building an average of about one coal-fired generating station a week.⁴⁸
• Low value of the dollar.
• Rising construction labor costs, and a shortage of skilled and experienced engineering staff.⁴⁹
• An atrophied domestic and international industrial and specialized labor base for nuclear plant construction and components.
• In the case of wind, competition for the best plant sites and a tight market for wind turbines; in the case of nuclear plants, limited global capacity to produce large and ultra-large forgings for reactor pressure vessels.⁵⁰
• Coincident worldwide demand for similar resources from other business sectors, including general construction and the construction of process plants such as refineries. Much of the demand is driven by the rapidly growing economies of Asia.⁵¹

The future trend in construction costs is a critical question for the power industry. Continued increases in capital costs would favor building natural gas plants, which have lower capital costs than most alternatives. Stable or declining construction costs would improve the economics of capital-intensive generating technologies, such as nuclear power and wind.⁵² At least some long-term moderation in cost escalation is likely, as demand growth slackens and new supply capacity is added.⁵³ But when and to what degree cost increases will moderate is as unpredictable as the recent cost escalation was unforeseen.

Financing Power Plant Projects

Even relatively small power plants cost millions of dollars. For example, the capital cost for a 50 MW wind plant would be about $105 million at $2,100 per kW of capacity. The investment cost is typically financed by a combination of debt and equity.⁵⁴ The financing structure and the cost of money depends on the type of developer and project-specific risk.

Three types of entities typically develop power plants:

• Investor-owned utilities (IOUs): IOUs are owned by private investors and are subject to government regulation of rates and conditions of service. They have guaranteed service territories and face limited competition. State utility commissions set electric rates designed to maintain the financial health of the utility, assuming it operates prudently. The commission also must approve proposals by the utility to build new power plants.⁵⁵
• Publicly-owned utilities (POUs): A POU is a utility that is an agency of a municipality, a state, or the federal government. Electric cooperatives are also considered to be POUs. Like IOUs, POUs have guaranteed service territories and face limited competition. Most POUs are small, provide only distribution service, and have limited financial and management resources.⁵⁶ But larger and some smaller POUs also own and operate power plants, sometimes as co-owners of projects where an IOU or independent power producer is the lead developer. Examples of POUs with large amounts of generation include the Tennessee Valley Authority and the municipal utilities serving the cities of Los Angeles and San Antonio. POUs set their own rates and make their own decisions to build power plants.
• Independent Power Producers (IPPs): IPPs are merchant developers and operators of power plants that sell wholesale power to utility and industrial buyers. Within limits they can sell power at whatever price the market will bear.⁵⁷ IPPs face more financial risk than regulated utilities—they do not have guaranteed service territories and can face intense competition for power sales—but can also earn larger profits. IPPs make their own decisions to build power plants.

All three types of entities play a major role in the electric power industry (Table 1). The lines between the entities can blur. Holding companies that own IOUs can also own IPPs. POUs sometimes own large shares of power projects developed by IOU or IPPs.

The cost of the money used to finance power projects varies significantly between IOU, POUs, and IPPs. A POU will normally finance a project with 100% debt at a low interest rate. The rate is low because interest paid on public debt is exempt from federal or state income taxes,⁵⁸ and because public entities have a very low risk of default (failure to make debt payments), much lower than for private businesses.⁵⁹ Typical municipal bonds have ratings in the middle or upper tiers of investment grade debt.⁶⁰

Privately owned IOUs and IPPs finance power projects with a mix of debt and equity. Debt is more costly to these companies than to POUs because it is not tax exempt and because they usually have lower credit ratings. The electric utility industry as a whole has a credit rating in the lower tier of the investment grade category (BBB).⁶¹ IPP debt often falls in the speculative category and has a higher interest rate than IOU or POU issues.⁶²

Investors expect private developers to make a significant equity contribution to a project.⁶³ Reliance on equity versus debt varies by company and project. The cost analysis used in this study assumes that IPPs and IOUs rely on, respectively, 40% and 50% equity (see Table D-1), except in the case where federal loan guarantees are available (see discussion of "Government Incentives", above). Equity is more expensive than debt,⁶⁴ and is more expensive for IPPs than IOUs because IPPs typically face more competition and financial risk.

In summary:

• Because POUs can finance a power project with 100% low-cost debt they can build power plants more cheaply than IOUs or IPPs. However, because of the small size of most POUs they do not have the financial or management resources to take on large and complex projects by themselves, so POUs often partner on projects where an IOU or IPP is the lead developer.
• IOU's typically have lower financing costs than IPP's because they have lower costs of debt and equity.⁶⁵
• Financing costs are highest for IPPs, which makes them somewhat less prone to take on the highest cost projects (such as coal and nuclear plants) unless POUs or IOUs are co-owners.

Fuel Costs

Fuel costs are important to the economics of coal, nuclear, and natural gas plants, and irrelevant to solar, geothermal, and wind power. Recent trends in the delivered cost of coal and natural gas to power plants are illustrated below in Figure 3. The constant dollar prices of both fuels have increased since the beginning of the decade, but the price escalation has been especially severe for natural gas.⁶⁶ Natural gas has also been consistently more expensive than coal. The comparatively low cost of coal partly compensates for the high cost of building coal plants, while the high cost of natural gas negates part of the capital cost and efficiency advantages of combined cycle technology.

Because it takes years to build a power plant, and plants are designed to operate for decades, generation plans largely pivot on fuel price forecasts. However, fuel prices have been notoriously difficult to predict. For example, EIA forecasts of delivered coal prices and natural gas wellhead prices have been off target by an average of, respectively, 47% and 64%.⁶⁷ EIA attributes the gap between actual and forecasted gas prices to a host of factors:

As regulatory reforms that increased the role of competitive markets were implemented in the mid-1980s, the behavior of natural gas was especially difficult to predict. The technological improvement expectations embedded in early AEOs [Annual Energy Outlooks] proved conservative and advances that made petroleum and natural gas less costly to produce were missed. After natural gas curtailments that artificially constrained natural gas use were eased in the mid-1980s, natural gas was an increasingly attractive fuel source, particularly for electricity generation and industrial uses. Historically, natural gas price instability was strongly influenced by natural gas resource estimates, which steadily rose, and by the world oil price. More recently, the AEO reference case has overestimated natural gas consumption due to the use of natural gas wellhead price projections that proved to be significantly lower than what actually occurred.⁶⁸

EIA's analysis illustrates how the confluence of technological, regulatory, resource, and domestic and international market factors make fuel forecasts so problematic. Fuel price uncertainty is especially important in evaluating the economics of natural gas-fired combined cycle plants. For the base assumptions used in this study, fuel constitutes half of the total cost of power from a new combined cycle plant, compared to 18% for a coal plant and 6% for a nuclear plant.

Figure 3. Coal and Natural Gas Constant Dollar Price Trends

Sources: EIA, Monthly Energy Review on-line data, Table 9.10, converted to constant dollars by CRS.

The price of the uranium used to make nuclear fuel has, like coal and natural gas, increased sharply and has been volatile (Figure 4). Although prices have recently dropped, they are still far above historic levels.⁶⁹ Over the long term, EIA expects nuclear fuel prices to increase in real terms from $0.58 per mmbtu in 2007 to $0.77 per mmbtu in 2023, and then slowly decline.⁷⁰ Even prices twice as high would not have a major impact on nuclear plant economics, which are dominated by the capital cost of building the plant.

Figure 4. Uranium Price Trends

Sources: Trade Tech Exchange Values, as reported in Platts Nuclear Fuel and http://www.uranium.info/.

Air Emissions Controls for Coal and Gas Plants

Regulations that limit air emissions from coal and natural gas plants can impose two types of costs: The cost of installing and operating control equipment, and the cost of allowances⁷¹ that permit plants to emit pollutants. The following emissions are discussed below:

Emissions from coal:

• Sulfur dioxide (SO₂), a precursor to acid rain and the formation in the atmosphere of secondary particulates⁷² that are unhealthy to breathe and can impair visibility.
• Mercury, a toxic heavy metal.
• Primary particulates (soot) entrained in the power plant's flue gas.

Emissions from coal and natural gas:

• Nitrogen oxides (NOx), a precursor to ground level ozone, acid rain, and the formation in the atmosphere of secondary particulates.
• Carbon dioxide (CO₂), a greenhouse gas produced by the combustion of fossil fuels.

The regulations and control technologies for SO₂, NOx, particulates, and mercury are discussed briefly under the category of "conventional emissions." These pollutants are subject to either existing regulations or regulations being developed under current law, and can be controlled with well-understood, commercially-available technologies. CO₂ is discussed in more detail because control technologies are still under development and may be far more costly than controls for conventional emissions.⁷³ While CO₂ is not currently subject to federal regulation, control legislation is being actively considered by the Congress and some states are taking action to limit CO₂ emissions.

More information on air emissions, particularly on regulatory and policy issues, is available in numerous CRS reports. The reports can be accessed through the "Energy, Environment, and Resources" link on the CRS website, http://www.crs.gov.

Conventional Emissions

The Environmental Protection Agency (EPA) has established National Ambient Air Quality Standards (NAAQS) for several pollutants, including SO₂, NOx, ozone, and particulates. New coal and natural gas plants built in areas in compliance with a NAAQS standard must install Best Available Control Technology (BACT) pollution control equipment that will keep emissions sufficiently low that the area will stay in compliance. Plants built in areas not in compliance with a NAAQS (referred to as "non-attainment" areas) must meet a tighter Lowest Achievable Emission Rate (LAER) standard.⁷⁴ In practice, air permit emissions are negotiated case-by-case between the developer and state air authorities. Federal standards set a ceiling; state permits can specify lower emission limits.

In addition to technology control costs, new plants that emit SO₂ must buy SO₂ emission allowances under the acid rain control program established by Title IV of the Clean Air Act.⁷⁵ Depending on the location of a new plant, it may also need to purchase NOx allowances.⁷⁶

Regulation of mercury is unsettled. On February 8, 2008, the U.S. Court of Appeals for the D.C. Circuit vacated the Bush administration's Clean Air Mercury Rule, which would have allowed new coal plants to comply with mercury emission limits by purchasing mercury allowances. Because of the court's action, coal plant mercury emissions are now categorized as a hazardous air pollutant. If the decision stands,⁷⁷ it will trigger a requirement for all coal plants, old and new, to install mercury control equipment that meets a Maximum Available Control Technology (MACT) standard. EPA has not yet defined a MACT standard for mercury, but state air officials will probably require new plants to meet tight mercury emission limits.⁷⁸

The technology and costs for controlling sulfur, NOx, particulate, and mercury emissions are briefly described below. For additional information on emission control technologies see the International Energy Agency Clean Coal Center at http://www.iea-coal.org/site/ieacoal/databases/clean-coal-technologies.

• Sulfur. Commercial technologies can remove 95% to 99% of the SO₂ formed by burning coal in pulverized coal plants, and over 99% of the sulfur in IGCC synthesis gas before it is burned. To the degree that a new pulverized coal unit or IGCC plant releases SO₂ to the atmosphere, it must buy SO₂ emission allowances. Because SO₂ emissions by plants with controls are so small, allowances are not a major expense compared to the other costs of running a power plant. At mid-2008 allowance and fuel prices, the annual cost of SO₂ allowances for a coal plant burning eastern coal would be on the order of $1 million, compared to over $220 million just for fuel.⁷⁹ The cost of the control equipment is more significant. An SO₂ control system will account for about 12% of the capital cost of a new pulverized coal plant and 29% of non-fuel operating costs (Table 2). (It is difficult to isolate environmental control costs for an IGCC plant because emissions control is largely integral with cleanup of the synthesis gas that is necessary, irrespective of environmental rules, prior to combustion.)
• Mercury. Some pulverized coal plants can achieve 90% removal of mercury as a co-benefit of operating SO₂ and particulate control equipment. Other plants will have to install a powdered activated carbon injection system (accounting for about 1% of the plant's capital cost and 9% of non-fuel operating costs). IGCC plants would remove 90% to 95% of the mercury from the synthesis gas using another technology also based on activated carbon.
• NOx. Commercial technologies can reduce NOx emissions to very low levels for pulverized coal and IGCC plants. Depending on a plant's location, it may have to purchase NOx emission allowances. As in the case of SO₂ allowances, because the controlled emission rates for new plants are so low the total cost of allowances is small compared to other plant operating costs. The cost of the control equipment for a pulverized coal plant is about 2% of capital expense and 9% of non-fuel operating costs.
• Particulates. Primary particulates are controlled using removal systems that have been a standard feature of pulverized coal plants for many years. Removal efficiencies exceed 99%. Primary particulate removal rates for IGCC plants are expected to be similar. Secondary particulates are controlled by reducing NOx and SO₂ emissions, as discussed above.

Carbon Dioxide

This section of the report discusses the technical and cost characteristics of carbon control technologies for coal and natural gas plants. The estimates of the cost and performance affects of installing carbon controls are uncertain because no power plants have been built with full-scale carbon capture. For additional information on carbon control technologies, see CRS Report RL34621, Capturing CO₂ from Coal-Fired Power Plants: Challenges for a Comprehensive Strategy, by [author name scrubbed], [author name scrubbed], and [author name scrubbed]; and Steve Blankinship, "The Evolution of Carbon Capture Technology, Parts 1 and 2," Power Engineering, March and May 2008.⁸⁰

CO₂ Removal for Pulverized Coal and Natural Gas Plants

Technology developed by the petrochemical industry, using a class of chemicals called amines, can be used to scrub CO₂ from flue gas. Amine scrubbing is currently used to extract CO₂ from part of the flue gas at a handful of coal-fired plants, to produce CO₂ for enhanced oil recovery and the food industry, but the scale is about a tenth of what would be needed to scrub 90% of the CO₂ from the entire flue gas stream of a large power plant.⁸¹ Scaling up amine technology to handle much larger gas flows at a power plant may be technically challenging.

Amine scrubbing is energy intensive. It diverts steam from power production and uses part of the plant's electricity production to compress the CO₂ for pipeline transportation to its final disposition. Amine scrubbing is estimated to cut a coal plant's electricity output by about 30% to 40%.⁸² The equipment is also costly. According to one study, the cost for building a new coal plant with amine scrubbing is an estimated 61% higher than building the a plant without carbon controls.⁸³ The same study estimated the cost for a coal plant retrofit installation, without taking into account the recent rapid increase in power plant construction costs, at about $1,600 per kW of net capacity, or almost $1 billion for a 600 MW plant.⁸⁴

The cost and performance impacts for adding amine scrubbing to a natural gas-fired combined cycle are also large. The estimated reduction in net electricity output is 14%, and the estimated increase in the plant capital cost is about 100%.⁸⁵ Researchers are attempting to commercialize less costly carbon capture technologies for conventional coal and gas plants, but these are still in early development.

CO₂ Removal for IGCC Coal Plants

Carbon capture for an IGCC plant involves multi-step treatment of the synthesis gas using technology originally developed for the petrochemical industry. Estimates of the cost and performance impact of incorporating carbon capture into a IGCC design vary widely. For the sample of studies shown in Table 3, the estimated increase in capital costs ranges from 32% to 51%. The estimated loss in generating capacity varies by more than a factor of two, from 13% to 28%. This wide variation reflects in part factors specific to different IGCC technologies, but is also an indication of limited experience with IGCC technology generally and the integration of carbon capture in particular.

While IGCC technology is arguably better-suited for carbon capture than pulverized coal systems, it does not currently provide a simple or inexpensive path to carbon control. In addition to the cost and performance penalties and uncertainties, other factors complicate implementing IGCC carbon control. For example, the nation's largest and least expensive coal supply is western subbituminous coal. However, the IGCC technologies best suited for using this coal also appear to incur the largest cost and performance penalties from adding carbon control technology.⁸⁶

CO₂ Allowance Costs

Congress has considered legislation that would put a cost on carbon emissions, such as the Lieberman-Warner Climate Security Act of 2007 (S. 2191). If Congress ultimately legislates allowance-based carbon controls, the estimated costs of such allowances are very uncertain. As an illustration of this uncertainty, Figure 5 shows EIA's alternative projections of CO₂ allowance prices under S. 2191. Depending on assumptions for such factors as the speed with which new technologies are deployed and their costs, and the availability for purchase of international CO₂ emission offsets, EIA's estimate of the price of allowances by 2030 ranges from about $60 to $160 per metric ton of CO₂ (2006 dollars).

Figure 5. EIA's Projections of S. 2191 CO2 Allowance Prices (2006$ per Metric Ton of CO2 Equivalent)

Sources: Supporting spreadsheets for EIA, Energy Market and Economic Impacts of S.2191, the Lieberman-Warner Climate Security Act of 2007, April 2008.

Even the low end of EIA's allowance price forecasts would impose costs far beyond those of existing air emissions regulations. Figure 6 compares the price of coal in EIA's long-term Reference Case projection (which assumes only current law, and therefore no carbon controls) to EIA's "core" case estimate of allowance prices from the S. 2191 study. Based on EIA's forecasts, by 2030 the allowance price is the equivalent of triple the coal price.⁸⁷ (As noted above, the outlook for CO₂ allowance prices is uncertain. Different legislative approaches and changes to other forecasting assumptions can produce very different estimates from those shown here.)

Figure 6. Comparison of EIA's Reference Case Coal Prices and S. 2191 Core Case CO2 Allowance Prices

Sources: Supporting spreadsheets for EIA, Energy Market and Economic Impacts of S. 2191, the Lieberman-Warner Climate Security Act of 2007, April 2008; CRS calculations (assumes 20 MMBtus per ton of coal and 209 lbs. of CO2 per MMBtu if coal consumed).

Financial Analysis Methodology and Key Assumptions

This financial analysis of new power plants provides estimates of the operating costs and required capital recovery of each generating technology through 2050. Plant operating costs will vary from year to year depending, for example, on changes in fuel prices and the start or end of government incentive programs. To simplify the comparison of alternatives, these varying yearly expenses are converted to a uniform annualized cost expressed as 2008 present value dollars.

Converting a series of cash flows to a financially equivalent uniform annual payment is a two-step process. First, the cash flows for the project are converted to a 2008 "present value." The present value is the total cost for the analysis period, adjusted ("discounted" using a "discount factor") to account for the time value of money and the risk that projected costs will not occur as expected. This lump-sum 2008 present value is then converted to an equivalent annual payment using a uniform payments factor.⁸⁸

The capital costs for the generating technologies are also converted to annualized payments. An investor-owned utility or independent power producer must recover the cost of its investment and a return on the investment, accounting for income taxes, depreciation rates, and the cost of money. These variables are encapsulated within an annualized capital cost for a project computed using a "capital charge rate." The financial model used for this study computes a project-specific capital charge rate that reflects the assumed cost of money, depreciation schedule, book project life, financing structure (percent debt and percent equity), and composite federal and state income tax rate. For a POU project, which is 100% debt financed, a "capital recovery factor" reflecting each project's cost of money is computed and used to calculate a mortgage-type annual payment.⁸⁹

Combining the annualized capital cost with the annualized operating costs yields the total estimated annualized cost of a project. This annualized cost is divided by the projected yearly output of electricity to produce a cost per Mwh for each technology. By annualizing the costs in this manner, it is possible to compare alternatives with different year-to-year cost patterns on an apples-to-apples basis.

Inputs to the financial model include financing costs, forecasted fuel prices, non-fuel operations and maintenance expense, the efficiency with which fossil-fueled plants convert fuel to electricity, and typical utilization rates (see Appendix D, Table D-1 through Table D-4, below). Most of these inputs are taken from published sources, such as the assumptions EIA used to produce its 2007 and 2008 long-term energy forecasts. The power plant capital costs are estimated by CRS based on a review of public information on recent projects. Appendixes B and C of the report displays the data used for the capital costs estimates.

Analysis of Power Project Costs

This section of the report analyzes the cost of power from the generating technologies discussed above. Results are first presented for a Base Case analysis. Results are then presented for four additional cases, each of which explores a key variable that influences power plant costs. These cases are:

• Influence of federal and state incentives.
• Higher natural gas price.
• Uncertainty in capital costs.
• Carbon controls and costs.

In each case the cost of power from a natural gas-fired combined cycle plant is used as a benchmark for evaluating the cost of power from the other generating technologies. The gas-fired combined cycle plant is used as a benchmark because of the dominant role it has played, and may continue to play, as the source of new generating capacity capable of meeting baseload and intermediate demand. The closer a generating technology comes to meeting or beating the power cost of the combined cycle, the better its chances of competing in the market for new power plants.

The Base Case is a starting point for comparing how different assumptions, such as for fuel and construction costs, change estimated power costs. None of the cases is a "most likely" estimate of future costs. Future power costs are subject to so many variables with high degrees of uncertainty that projecting a most likely case is impractical. The object of the analysis is provide insight into how key factors influence the costs of power plants, including factors under congressional control such as incentive programs.

These estimates are approximations subject to a high degree of uncertainty. The rankings of the technologies by cost are therefore also an approximation and should not be viewed as definitive estimates of the relative cost-competitiveness of each option. Also note that project-specific factors would weigh into an actual developer's decisions, including how close a fossil plant would be to fuel sources, local climate (for wind and solar), the need for and cost of transmission upgrades, the developer's appetite for risk, and the developer's financial resources.

Case 1: Base Case

Key Observations

• The lowest cost generating technologies in the Base Case are pulverized coal, geothermal, and natural gas combined cycle plants. All have costs around $60 per Mwh (2008 dollars). Based on the assumptions in this report, other technologies are at least a third more expensive.
• Of the three lowest cost technologies, geothermal plants are limited to available sites in the West that typically support only small plants, and coal plants have become harder to build due to cost and environmental issues. The gas-fired combined cycle plant is currently a technology that can be built at a large scale, for cycling or baseload service, throughout the United States.
• The above projections are based on private (IOU or IPP) funding of power projects. The cost per Mwh drops precipitously if the developer is assumed to be a POU with low-cost financing. However, most POUs are small and do not have the financial or managerial resources to build large power projects.

Discussion

As noted earlier in the report, power plants can be built by investor-owned utilities (IOUs), publicly owned utilities (POUs), or independent power producers (IPPs). The Base Case assumes that coal and nuclear plants are constructed by IOUs because they are most likely to have the financial resources and regulatory support to undertake these very large and expensive projects. The natural gas combined cycle plant is assumed to be built by an IPP. IPPs often prefer to build and operate gas-fired projects because of their relatively low capital costs. The wind, solar, and geothermal plants are also assumed to be IPP projects. The most common current practice is for IPPs to develop renewable projects and sell the power to regulated utilities.

The Base Case has the following characteristics:

• The analysis is for new projects beginning operation in 2015.
• Estimates of fuel prices, allowance prices, and most operational characteristics are from EIA's Reference Case assumptions for the 2008 Annual Energy Outlook.⁹⁰
• The 2008 overnight capital costs for each technology are estimated by CRS from public information on recent projects (see Appendix B).
• The Base Case excludes "discretionary" incentives: The federal loan guarantee program and clean coal tax credit programs, state utility commission decisions to allow CWIP in rates, and the federal renewable energy production tax credit, which is scheduled to expire at the end of 2010. These incentives are excluded because they are granted by government entities based on a case-by-case analysis of individual projects, and/or are dependent on congressional action to fund or extend the incentives. Accordingly, there is no certainty that most projects will receive these incentives. For example, as of November 2008, DOE had received requests from nuclear plant developers for $122 billion in loan guarantees, compared to congressional approval of only $18.5 billion for nuclear projects.⁹¹
• The only incentives included in the Base Case are (1) the 30% investment tax credit for solar and geothermal energy systems, which has been extended to 2017 and is automatically available to any qualifying facility; and (2) the nuclear production tax credit, which is available to any qualifying facility. As discussed above, the assumed value of the nuclear credit is 1.35 cents per kWh.
• The Base Case includes no carbon emission controls or costs.

Given these assumptions, Table 4 presents the resulting annualized cost of power per Mwh for each technology.

Under the Base Case assumptions, the lowest-cost options are pulverized coal, natural gas combined cycle, and geothermal generation, all in the $60 per Mwh (2008 dollars) range (column 10). These results are attributable to the following factors:

• Pulverized coal is a mature technology that relies on a relatively low cost fuel.
• Natural gas is an expensive fuel, but combined cycle technology is highly efficient and has a low construction cost.
• Geothermal energy has no fuel cost and unlike variable renewable technologies, such as wind and solar, can operate at very high utilization rates (high utilization allows the plant to spread fixed operating costs and capital recovery charges over many megawatt-hours of sales).

Although all three technologies have similar power costs, the coal and geothermal technologies have limitations and risks that the natural gas combined cycle does not face. Geothermal plants are limited to relatively small facilities (about 50 MW) at western sites. As discussed above, many coal projects have been canceled due to environmental opposition and escalating construction costs. In contrast, the gas-fired combined cycle plant has limited environmental impacts, can be located wherever a gas pipeline with sufficient capacity is available, and plants can be built with generating capacities in the hundreds of megawatts. Probably the main risk factor for a combined cycle plant is uncertainty over the long term price and supply of natural gas.

In the Base Case, wind power, IGCC coal, and nuclear energy have costs in the $80 per Mwh range. IGCC and nuclear plants are very expensive to build, with estimated overnight capital costs of, respectively, $3,359 and $3,682 per kW of capacity (2008 dollars; see Table D-2). Because the plants are expensive and take years to construct (an estimated four years for an IGCC plant and six years for a nuclear plant) these technologies also incur large charges for interest during construction that must be recovered in power costs.

Wind has a relatively high cost per Mwh because wind projects have high capital costs ($2,100 per kW of capacity) and are assumed to operate with a capacity factor of only 34%. The low capacity factor means that the plant is the equivalent of idle two-thirds of the year. Consequently, the capital costs for the plant must be recovered over a relatively small number of units of electricity production, driving up the cost per Mwh. High capital costs and low rates of utilization also drive up the costs of the solar thermal and solar PV plants to, respectively, $100 per Mwh and $255 per Mwh.

Comparison to a Benchmark Price of Electricity

Another way of viewing the results is to compare each technology's costs to a benchmark cost of electricity. As discussed above, the benchmark used is the cost of power from a natural gas combined cycle plant.

Column 3 of Table 5 shows the difference between the Base Case power cost for each technology and the Base Case cost of power from the gas-fired combined cycle. Geothermal energy and pulverized coal are the only technologies that have power costs similar to the natural gas combined cycle plant. Nuclear, wind, and coal IGCC power are projected to have costs 31% to 35% higher, and solar thermal has a projected power cost 62% higher. Solar photovoltaic is over 300% higher.

Effect of Financing Costs

The cost of money can have a significant impact on the cost of power. As discussed earlier, POUs have access to lower cost financing than IOUs or IPPs. The significance of lower cost financing is illustrated in Table 6, which compares the cost of power assuming IOU and IPP financing (column 3) with the cost of power assuming POU financing (column 4). Excluding for the moment the solar technologies, the reduction in the cost of power ranges from 14% for the combined cycle plant (the least capital-intensive option, which makes it least sensitive to financing costs) to 37% for the capital-intensive IGCC and nuclear plants (column 5). The low cost of public financing helps explain why many capital intensive coal and nuclear projects have POU co-owners.⁹²

The reduction in cost by using public financing is only 11% for the solar thermal plant and 14% for the solar photovoltaic plant. The reductions are small because when the plants are publicly financed they lose the 30% renewable energy investment tax credit (POUs do not pay taxes and so cannot take advantage of any tax-based incentives). The loss of the tax credit largely negates the benefit of lower cost POU financing for solar projects.

Case 2: Influence of Federal and State Incentives

Key Observations

• Government financial incentives can make high-cost technologies into low-cost options. The incentive with the greatest impact is the federal loan guarantee, which reduces the cost of financing capital-intensive technologies. With a loan guarantee the cost of nuclear power flips from a high-cost option ($83.22 per Mwh) to one of the low cost ($63.73 per Mwh).
• Even when competing technologies have the advantage of the discretionary government incentives, no technology currently has a significant cost advantage over the natural gas combined cycle.

Discussion

The Base Case includes only non-discretionary incentives: The renewable energy investment tax credit and the nuclear production tax credit. This analysis includes the following discretionary incentives:

• Federal loan guarantees for nuclear power.
• A clean coal tax credit for the IGCC plant.
• A production tax credit for wind (assumes continuation of the terms and conditions of the current production tax credit).
• Return on construction work in progress (CWIP) in rates for IOUs.

Table 7 shows the effect of the discretionary incentives compared to the Base Case. The additional incentives have the greatest effect on nuclear power. The annualized cost of nuclear generation drops by 23% (column 7), from one of the highest to one of the lowest costs. The most important driver for the nuclear plant is the federal loan guarantee, which allows a developer to fund a project with 80% debt at a much reduced interest rate. The loan guarantee alone cuts the cost of nuclear power by 20% ($15.44 per Mwh).

The renewable production tax credit reduces the cost of wind power by 10%. Geothermal and combined cycle plants (with no additional incentives) and coal (with a 5% reduction in cost due to CWIP in rates) remain low-cost options.

Table 8 compares the combined cycle benchmark cost of power (column 3) to the cost of power with discretionary incentives (column 4). The table is limited to the technologies that receive the additional incentives: Pulverized coal (CWIP in rates), IGCC coal (CWIP and an investment tax credit), wind (production tax credit), and nuclear (loan guarantee and CWIP). With discretionary incentives, nuclear power swings from a 35% higher cost than the combined cycle to only a 3% difference (comparing columns 3 and 4). The cost advantage of the combined cycle over wind and IGCC coal drops from more than 30% to just under 20%. The cost of power from pulverized coal remains similar to that of the combined cycle.

Case 3: Higher Natural Gas Prices

Key Observations

• If the price of natural gas is assumed to be 50% higher than in the Base Case, geothermal and pulverized coal power are clearly less costly than the combined cycle. However, the use of the geothermal power is limited to available sites in the western United States, and pulverized coal by construction cost and environmental issues.
• In the higher gas price case, the cost of power from the natural gas combined cycle plant converges with wind, nuclear, and IGCC coal. The combined cycle plant no longer has a clear economic advantage over these technologies, but neither is it at a great disadvantage.

Discussion

The economics of natural gas-fired generation pivot on fuel prices. For the base assumptions used in this study, fuel constitutes half of the total cost of power from a new combined cycle power plant, compared to 18% for a coal plant and 6% for a nuclear plant. In addition to being critical to the cost of gas-fired power, natural gas prices are also one of the most uncertain elements in this analysis. As discussed earlier in this report, natural gas prices have been exceptionally difficult to forecast. If the United States becomes more dependent in the future on imports of liquefied natural gas, the domestic and international natural gas markets will be increasingly linked, adding an additional element of uncertainty to the natural gas price outlook.⁹³

Underestimates of natural gas prices were pervasive among government and private forecasters in the 1990s and contributed to over-investment in gas-fired generating capacity.⁹⁴ If future gas prices are higher than assumed in this report's Base Case, the economics of gas-fired generation could change substantially. The gas market has historically been volatile. Gas prices increased more than 200% from the early 1990s through 2007, and annual increases sometimes exceeded 50% (Figure 7).

Figure 7. Natural Gas Price Trends (Henry Hub Spot Price)

Source: St. Louis Federal Reserve Bank FRED database.

Figure 8 illustrates the Base Case gas price projection and an alternative that ramps up to a level 50% higher than in the Base Case. In the Base Case the annualized cost of power from a natural gas combined cycle plant is $61.77 per Mwh. With a 50% higher gas price, the combined cycle power cost is $77.05 per Mwh. At this power cost the combined cycle is substantially more costly than pulverized coal or geothermal power, and has a clear economic advantage only over the solar technologies (Table 9, column 4). On the other hand, even with this much higher fuel price projection, the cost of power from the combined cycle is still comparable to that of wind, nuclear, and IGCC coal generation; and while pulverized coal and geothermal power have lower costs, as discussed above the former is increasingly hard to build for cost and environmental reasons, and the latter is limited to small plants at western sites. Therefore, even with a 50% increase in fuel prices, the gas-fired combined cycle is still a competitive option for new generating capacity.

Figure 8. Projection of Natural Gas Prices to Electric Power Plants, 2006 $ per MMBtu

Source: EIA, Annual Energy Outlook 2008, and CRS estimates.

Another perspective is to determine the increase in the Base Case natural gas price projection required for the cost of power from the natural gas combined cycle plant to equal the cost of power from an alternative technology. This is illustrated in Table 10. The table shows that the price of gas would have to be between 62% to 69% higher than in the Base Case for the cost of power from a combined cycle to equal the projected cost of electricity from nuclear, wind, or coal IGCC technologies (column 3). Natural gas prices would have to increase by about 125% to 635% for the cost of combined cycle power to match solar thermal or solar photovoltaic electricity costs.

Case 4: Uncertainty in Capital Costs

Key Observations

• Because of its low capital costs and assumed high utilization rate, the power cost of the gas-fired combined cycle plant is about half as sensitive to changes in capital costs as the other technologies.
• The implication is that if power plant capital costs continue to increase rapidly, the competitive position of the combined cycle will improve compared to all other technologies.
• If capital costs decline, the competitive position of the other technologies will substantially improve versus the combined cycle. However, even assuming a 25% drop in capital costs compared to the Base Case, the combined cycle is still competitive with all other technologies.

Discussion

As noted above, the cost of building power plants has recently increased dramatically. Whether costs will continue to increase, remain steady in real dollar terms, or decline is unknown. Table 11 illustrates the effect on the cost of power of assuming a uniform 25% increase or decrease in capital costs for all technologies compared to the Base Case. Power costs change by about +/-20% for each technology except for the gas-fired combined cycle plant (+/-12%; see column 3). This is because the combined cycle has a relatively low capital cost and a high capacity factor.

Table 11 shows that the power cost of the combined cycle is about half as sensitive to changes in capital costs as the other generating technologies. The implication is that continued rapid escalation in the cost of building power plants will favor the economics of combined cycles. This is illustrated by Table 12. In the Base Case (Column 3), the power costs of wind, nuclear, and IGCC coal are about a third higher than the combined cycle. In the high capital cost case (Column 4) the difference widens to almost 50%. On the other hand, decreases in capital costs, whether the result of market forces or government incentives, would reduce the cost of power from the other technologies about twice as much as for the combined cycle. This is illustrated by the low capital cost case (Column 5), in which all the non-solar technologies are within 21% or less of the generating cost of the combined cycle.

Case 5: Carbon Controls and Costs

Key Observations

• The estimates of carbon-related allowance costs and control technology costs used in this analysis are subject to an exceptional degree of uncertainty, including whether Congress will actually pass carbon control legislation. The results of this analysis are therefore equally uncertain.
• With the carbon control assumptions used in this analysis, coal-fired generation is expensive, ranging from about $100 to almost $120 per Mwh. The least expensive options include zero-carbon emission technologies: Geothermal ($59.23 per Mwh), nuclear ($83.22) and wind ($80.74).
• The natural gas combined cycle plant without carbon capture is competitive with the other options, even with allowance costs, at $77.21 per Mwh.
• If the cost and efficiency penalties of carbon capture technologies are assumed to drop by 50%, the gas-fired combined cycle plant with capture has an electricity cost comparable to wind and nuclear power. However, a coal plant with capture is still more expensive than wind or nuclear power.

Discussion

Carbon control legislation is under consideration by the Congress, but there has been no agreement on the structure of a control regime or a timetable for implementation. No power plants have been built with full scale carbon capture equipment. The costs of CO₂ allowances and control systems are therefore very uncertain. Actual costs will depend on the content of final legislation (if any), the development of allowance markets in the United States and abroad, and the evolution of control technologies.

The carbon capture power cost analysis for this study is based on the following assumptions:

• Power plant cost and performance with carbon controls assume current (petrochemical industry based) technology capable of removing 90% of the CO₂. As discussed above, the cost of carbon capture for power plants using petrochemical industry derived technology will be very high. Table 13 provides estimates of how the capital costs and heat rates of coal and gas plants increase with the addition of carbon controls based on current technology. Capital costs increase by 42% to 97% (column 4), and heat rates increase by 21% to 27% (column 7) resulting in a decline in efficiency. Newer technologies may be less costly and more efficient, but these are still in development.

• The CO₂ allowance price projection is adapted from the EIA "core" case forecast from its analysis of S. 2191.⁹⁵ Allowance costs begin in 2012 at $17.70 per metric ton of CO₂ (2008 dollars); increase by 2020 and 2030 to, respectively, $31.34 and $63.99; and reach $266.80 by 2050 (see Table D-4). All allowances must be purchased (i.e., there is no free distribution of allowances to power plants).
• Fuel prices are the same prices used in the Base Case (see Table D-4).
• As in the Base Case, the only financial incentives included are the nuclear production tax credit and the investment tax credit for solar and geothermal plants.
• From a financing standpoint, units with carbon controls are assumed to be high risk projects that incur financing costs equivalent to below investment grade interest rates. This assumption is made because units coming on-line in 2015, as assumed for this study, would be part of the first wave of power plants with carbon controls.

Table 14, below, shows estimates of the levelized cost of power for a carbon capture case.

The results indicate:

• The power costs for coal plants using control technologies are high compared to the Base Case. The costs in the carbon case range from $100.69 per Mwh to almost $120 per Mwh (column 10), compared to $63.19 per Mwh for a pulverized coal unit in the Base Case (Table 14, column 10). This illustrates the impact of the high capital costs and efficiency penalties of current carbon capture technologies.
• With the imposition of carbon costs on fossil plants, three of the least expensive options are zero-carbon technologies: Geothermal ($59.23 per Mwh), nuclear ($83.22) and wind ($80.74). Because geothermal plants are limited to specific sites in the western states, nuclear power (a baseload technology) and wind power (a variable renewable resource) are the zero carbon options with relatively low costs and wide latitude for plant sites.
• A fourth relatively low-cost technology is the natural gas combined cycle plant without carbon capture ($77.21 per Mwh including allowance costs). The relatively low cost is due to the technology's low capital cost, high capacity factor, and relatively low emissions of CO₂ per megawatt-hour of power generated. As shown in Table 14, the natural gas combined cycle plant without carbon capture incurs allowance costs of $13.06 per Mwh, which is 61% less than the pulverized coal plant cost of $33.80 per Mwh (column 6). In other words, for every dollar of allowance costs incurred by a coal plant without capture technology, the combined cycle incurs only about 40 cents in costs.⁹⁶
• Solar thermal power ($100.32 per Mwh) has a lower cost than fossil plants with carbon capture technology, but is still estimated to be about 20% more expensive than nuclear and wind power.

The relatively low cost of power from the natural gas combined cycle plant is in part a function of the fuel price. As noted above, the carbon capture analysis uses the same fuel price projections as in the Base Case. It is possible that in a carbon-constrained world demand for gas will increase, driving up prices. As shown below in Table 15:

• A 12% increase in the price of gas would equalize the cost of electricity from the combined cycle plant without carbon capture with wind power (column 3);
• A 20% increase would equalize the power cost of the combined cycle plant and the nuclear plant;
• The price of natural gas would have to more than double for the power cost of the gas-fired combined cycle plant to equal the cost of coal power with carbon controls, or increase by 75% to match the cost of solar thermal power.

This scale of natural gas price increases has precedent. As shown in Figure 7, between the early 1990s and 2007 the market price of natural gas increased by about 200%.

As discussed above, the cost and efficiency impacts of current carbon capture technologies are high, and improved technologies are under development. Table 16 shows the estimated cost of power for plants with carbon capture assuming that capital cost and heat rate (efficiency) penalties are both reduced by 50%. In this case the combined cycle plant with capture has an electricity cost slightly less than wind and nuclear power, and the pulverized coal plant with capture closes to within 20% of wind power and 16% of nuclear (columns 8 and 9). The IGCC plant with capture is more expensive, with a power cost 28% higher than wind and 24% higher than nuclear; this result reflects the high cost of IGCC technology even before carbon capture is added.

Appendix A. Power Generation Technology Process Diagrams and Images

Pulverized Coal

Figure A-1. Process Schematic: Pulverized Coal without Carbon Capture

Source: Adapted from MIT, The Future of Coal, 2007.

Figure A-2. Process Schematic: Pulverized Coal with Carbon Capture

Source: Adapted from MIT, The Future of Coal, 2007.

Figure A-3. Representative Pulverized Coal Plant: Gavin Plant (Ohio)

Source: Image courtesy of Industcards.com.

Integrated Gasification Combined Cycle Coal (IGCC)

Figure A-4. Process Schematic: IGCC without Carbon Capture

Source: Adapted from MIT, The Future of Coal, 2007.

Figure A-5. Process Schematic: IGCC with Carbon Capture

Source: Adapted from MIT, The Future of Coal, 2007.

Figure A-6. Representative IGCC Plant: Polk Plant (Florida)

Source: Image courtesy of Industcards.com.

Natural Gas Combined Cycle

Figure A-7. Process Schematic: Combined Cycle Power Plant

Source: Diagram from Siemens Energy http://www.powergeneration.siemens.com/products-solutions-services/power-plant-soln/combined-cycle-power-plants/CCPP.htm

Figure A-8. Representative Combined Cycle: McClain Plant (Oklahoma)

Source: image courtesy of Industcards.com.

Nuclear Power

Figure A-11. Representative Gen III/III+ Nuclear Plant: Rendering of the Westinghouse AP1000 (Levy County Project, Florida)

Source: AP1000 image from Progress Energy (http://www.progress-energy.com/aboutenergy/poweringthefuture_florida/levy/ap1000.jpg).

Wind

Figure A-12. Schematic of a Wind Turbine

Source: Schematic from California Energy Commission EnergyQuest website (http://www.energyquest.ca.gov/story/chapter16.html)

Figure A-13. Representative Wind Farm: Gray County Wind Farm (Kansas)

Source: Image of Gray County wind farm from http://www.kansastravel.org/graycountywindfarm.htm.

Figure A-14. Wind Turbine Size and Scale (FPL Energy)

Source: Image of wind turbine scale from FPL Energy (http://www.fplenergy.com/renewable/pdf/NatLeaderWind.pdf)

Geothermal

Figure A-15. Process Schematic: Binary Cycle Geothermal Plant

Source: Diagram from Steven Lawrence, presentation on "Geothermal Energy," University of Colorado, undated, citing Godfrey Boyle, Renewable Energy, 2nd Edition, 2004 http://leeds-faculty.colorado.edu/lawrence/syst6820/Lectures/Geothermal%20Energy.ppt.

Figure A-16. Representative Geothermal Plant: Raft River Plant (Idaho)

Source: image courtesy of Industcards.com.

Solar Thermal Power

Figure A-17. Process Schematic: Parabolic Trough Solar Thermal Plant

Source: Diagram from http://www.solarserver.de/solarmagazin/solar-report_0207_e.html.

Solar Photovoltaic Power

Figure A-20. Process Schematic: Central Station Solar Photovoltaic Power

Source: Diagram from California Energy Commission, Comparative Costs of California Central Station Electricity Generation Technologies, Appendix B, p. 61

Figure A-21. Representative Solar PV Plant: Nellis Air Force Base (Nevada)

Source: Image from the Nellis Air Force Base website at http://www.nellis.af.mil/shared/media/document/AFD-080117-039.pdf.

Figure A-22. Nellis AFB Photovoltaic Array Detail

Source: Image from the Nellis Air Force Base website at http://www.nellis.af.mil/shared/media/document/AFD-080117-039.pdf.

Appendix B. Estimates of Power Plant Overnight Costs

The financial analysis model used in this study calculates the capital component of power prices based on the "overnight" cost of a power plant. The overnight cost is the cost that would be incurred if a power plant could be built instantly. The overnight cost therefore excludes escalation in equipment, labor, and commodity prices that could occur during the time a plant is under construction. It also excludes the financing charges, often referred to as interest during construction (IDC), incurred while the plant is being built.

With the exception of plants using carbon control technology (see Appendix C) the overnight costs were estimated for this study from public information on actual power projects. The costs were estimated as follows:

• CRS developed a database of information on 161 power projects and cost estimates covering the fossil, nuclear, and renewable energy technologies included in this report.
• A subset of the projects in the database were used to estimate overnight costs. Projects were excluded for many reasons, including because the projects were too old to reflect current construction costs, did not use standard technology, were extreme high or low outliers and no information was available to explain the costs, or had other unusual characteristics (e.g., some plants reduced costs by purchasing used or surplus equipment).
• The remaining projects were sorted by technology (e.g., nuclear, wind, etc.). The reported cost per kilowatt of capacity for the projects in each group were then averaged to estimate the overnight cost for each technology.

To the extent possible the information for the database was taken from information filed by utilities with state public service commissions. The advantage of using this source is that utilities seeking permission to construct new plants are often required to disgorge cost details. With these details the project cost estimate can be adjusted to exclude IDC and other expenses not directly associated with the cost of the plant, such as major transmission system upgrades distant from the plant site.

When utility commission filings for a project were not available, as was almost always true for IPP and POU projects, other public sources were used, including press releases and trade journal articles. In most cases it was possible to determine whether or not a cost estimate included IDC. However, it was rarely possible, with or without utility commission filings, to determine how much cost escalation was built into a project estimate. Because it was not possible to extract the escalation costs from the project estimates, as a rough correction the financial model assumed no cost escalation to avoid a double count. The model does compute the IDC charges.

The 161 projects in the database includes information on 119 United States power plant projects. Some are still in the planning stage, and a few never progressed beyond paper studies and were canceled. The database also includes information on 31 generic and 11 foreign cost estimates for nuclear power plants. (A generic estimate is a cost estimate not associated with any real project or specific site. Generic estimates are usually made by vendors or found in government and academic studies.) The generic and foreign estimates are useful for illustrating cost trends because no nuclear plants have been built in the United States in many years, but none were used in the final estimate of the overnight nuclear plant cost.

Although the capital costs used in this study are based on these actual project estimates, the capital costs are still subject to significant uncertainty due to such as factors as cost escalation and evolution in power plant and construction technology. The uncertainty is greatest for the technologies which have the least commercial experience, such as advanced nuclear plants and IGCC coal plants.

Immediately following is information on the projects used to estimate overnight costs for this report. There is a table for each technology (e.g., pulverized coal) listing each project used to estimate the overnight cost for that technology. Accompanying each table is a graph showing the time trend for that technology's capital costs. The data points on the graph are marked to indicate whether a point represents a project used in estimating the overnight cost, or another project that was excluded from the estimate for one of the reasons discussed above. The time axis for these graphs is the actual or planned first year of commercial service.

The following acronyms are used in the tables:

Pulverized Coal

Figure B-1. Pulverized Coal Project Cost Trends

Integrated Gasification Combined Cycle (IGCC) Coal

Figure B-2. IGCC Project Cost Trends

Nuclear

Figure B-3. Nuclear Project Cost Trends

Natural Gas Combined Cycle

Figure B-4. Combined Cycle Project Cost Trends

Wind

Figure B-5. Wind Project Cost Trends

Geothermal

Figure B-6. Geothermal Project Cost Trends

Solar Thermal

Figure B-7. Solar Thermal Project Cost Trends

Solar Photovoltaic

Figure B-8. Solar PV Project Cost Trends

Appendix C. Estimates of Technology Costs and Efficiency with Carbon Capture

Pulverized Coal with Carbon Capture

The costs and heat rate for a supercritical pulverized coal plant with carbon capture is primarily based on information from MIT's 2007 study, The Future of Coal.⁹⁷ MIT estimated that a new supercritical plant built with amine scrubbing for CO₂ removal would have the following characteristics:

• CO₂ capture rate: 90%
• Change in efficiency compared to a new plant without carbon capture: -23.9% (from 38.5% to 29.3%). This equates to an increase in the heat rate of 31.3%.
• Increase in capital cost: 61%.⁹⁸

For a new plant with amine scrubbing to have the same 600 MW net capacity as a new plant without carbon controls, the size of the plant has to be scaled up to account for the electricity and steam demands of the capture system. The increase is proportional to the change in efficiency. Therefore, a developer would have to build the equivalent of a 788 MW plant with carbon capture to get 600 MW of net capacity, with the difference (188 MW) consumed by the amine scrubbing system, either in the form of steam diverted from power generation or electricity used to compress the CO₂.⁹⁹

MIT does not break out the variable and fixed O&M costs for carbon capture, as required by the financial model used in this study. These costs were calculated from a DOE study of the costs of retrofitting carbon capture to the Conesville Unit 5 coal-fired plant in Ohio. Based on this study, the incremental O&M costs for carbon capture are $8.24 per kW for fixed O&M and $7.79 per Mwh for variable O&M (2006 dollars).¹⁰⁰ These costs for operating the carbon capture system are added to the base O&M costs for a coal-fired plant, as estimated by EIA, to calculate the total O&M costs for the plant.

The estimated characteristics of a new supercritical pulverized coal plant with amine scrubbing are:

• Capacity: 600 MW.
• Heat rate: the base heat rate of 9,200 btus per kWh in 2008 increases by 31.3% to 12,080 btus per kWh.
• Overnight capital cost: $4,025 per kW (base 2008 cost of $2,500 per kW increased by 61%).
• Variable O&M costs (2006 dollars): a base value of $5.86 per Mwh plus the carbon control incremental cost of $7.79 per Mwh for a total of $13.65 per Mwh.
• Fixed O&M costs (2006 dollars): a base of $35.20 per kW plus the carbon control incremental cost of $8.24 per kW for a total of $43.44 per kW.¹⁰¹
• Capacity factor: 85%, same as for a new supercritical plant without carbon capture.
• Construction time: assumed to be four years, same as for a new supercritical plant without carbon capture.

IGCC Coal and Natural Gas Combined Cycle with Carbon Capture

The operating and cost characteristics of a coal IGCC plant built with carbon capture are taken from EIA assumptions for its 2008 long-term forecast,¹⁰² except for the capital cost. As shown in Appendix B, the cost estimate for an IGCC plant without carbon capture, based on public information on current projects, is $3,400 per kW in 2008. This is much higher than EIA's estimate for an IGCC plant without ($1,773 per kW) or with ($2,537) carbon controls.

To estimate the capital cost of an IGCC plant with carbon capture, the percentage difference in the EIA estimates of plants with and without capture (43%) was applied to the CRS estimate of $3,400 per kW without capture. This produces an estimated cost for an IGCC plant with carbon controls of $4,862.¹⁰³ EIA's other assumptions, such as for O&M costs and heat rates, are used without adjustment in this study.

The capital cost for a natural gas-fired combined cycle with carbon capture was estimated in the same way. Based on public data for current projects, the overnight cost estimate for a new combined cycle used in this study is $1,200 per kW in 2008 (see Appendix B). This compares to EIA's estimates of $706 per kW for a combined cycle without carbon capture and $1,409 with carbon capture, a premium of 100%.¹⁰⁴ The capital cost for a new combined cycle with carbon capture used in this study is therefore double the CRS base cost of $1,200 per kW, or $2,400 per kW. As with the coal IGCC, EIA's other assumptions for a combined cycle plant with carbon capture are used without adjustment.

Appendix D. Financial and Operating Assumptions

Appendix E. List of Acronyms and Abbreviations