Nuclear Energy: Economics

Introduction

In an entry published in the Encyclopedia of Energy Engineering and Technology, I attempted to answer the following question: Is nuclear power economic? Unfortunately, I was unable to give a definitive answer to this question. Most of the work on that entry was done in the 2004-2005 time frame, and at that time, there was very little information on the realized costs of building nuclear power plants in the United States. There was some information on the realized cost of building nuclear power plants abroad (mainly in the Far East). However, that information was at best “sketchy” and difficult to use to estimate costs in the United States. Additionally, in the United States, a number of cost estimates of building hypothetical (“paper”) plants at hypothetical (“Middletown USA”) sites were published. History has shown that such estimates are always too low.14 In short, at that time, there was a large amount of uncertainty about nuclear power plant construction costs. Thus, the best that could be done was to conclude that nuclear capital costs would have to be less than $1500 per kW before nuclear power would be economic. (This was substantially less than the cost of building nuclear power plants in the Far East.)

Over the last few years, the economic and political environment has changed, and thus, the question of whether nuclear power is economic needs to be reevaluated. First, a number of estimates of building actual nuclear power plants at actual sites have recently become available. This, by itself, would reduce some of the uncertainty about nuclear power plant construction costs. However, these estimates are two to three times higher than the ones made in the mid-2000s. Additionally, power plant construction costs in general have increased substantially. For example, the realized costs of building coal-fired power plants and wind farms have increased by about 80%—100%J²1 Such increases in construction costs would have major effects on the economics of nuclear power.

Second, unlike when the original entry was written, for a variety of reasons, utilities are now far more interested in nuclear power. In 2006, licensing activity at the U.S. Nuclear Regulatory Commission (NRC) was limited to four early site permit approvals. By issuing an early site permit (ESP), the U.S. NRC approves one or more sites for a nuclear power plant, independent of an application for a combined license to build and operate a power plant. This ESP can be valid for up to 40years. Since then, 16 utilities have filed applications with the NRC to build and operate a total of 28 nuclear units. Given this increased licensing activity, a reexamination of the underlying economics of nuclear power is certainly in order.

Third, in the 2004-2005 time frame, most forecasts of coal and natural gas prices in 2015-2020 time frame were about $1.45 and $6 ($2009) per mmBtu, respectively.¹³¹ Since then, fossil fuel prices increased substantially and then fell. As of 2010, most forecasts of fossil fuel prices are now greater than the ones made 5-6 years ago. Increased fossil fuel prices will clearly affect the economics of nuclear power.

The purpose of this entry is to reexamine the economies of nuclear power in light of the changed economic environment. Before proceeding with the analysis, two comments about the scope of the analysis will be made. First, in this entry, it is assumed that the investment decision—e.g., the decision to build a nuclear unit or coal-fired power plant—will be made by the owner of a traditional utility. This utility owns and operates other power plants, and the operation of all of them is interrelated. That is, if the utility chooses to operate power plant X less, it would have to operate power plant Y more to meet demand. Otherwise, “the lights would go out.” In such cases, the decision maker’s objective is to build the unit that minimizes the total cost of building the plant in question and operating all power plants. Suppose, for example, that the decision maker has the choice of building a nuclear or coal-fired power plant. The decision maker will calculate the total system costs if the nuclear plant is built and compare that estimate with the total system costs if the coal plant is built. The decision maker would choose to build the plant that yields the lower total system costs.

The available software to estimate total system costs is complex and expensive and requires many assumptions, and thus, using this approach is beyond the scope of this entry. However, if the alternative to the nuclear power plant is another baseload plant type operating in the same portion of the merit order (baseload demand), total system costs would be minimized by choosing the plant type that has the lower “stand-alone” or levelized cost. (Levelized costs will be defined below. Additionally, because electricity is costly to store, demand will vary over the day, month, and/or year. The portion of total demand that does not vary is called baseload, and the units that are used to meet this demand will run at close to full capacity over the entire year.) This is because the operation of both units under consideration will have the same effect on the operation of the other units. For example, suppose that the levelized cost of building and operating a nuclear power plant for 40years is 6 cents per kWh, and the levelized cost of building and operating a combined-cycle natural gas-fired power plant is 8 cents per kWh. If both units are assumed to operate in the baseload mode, then the operation of both units will have the same effect on the operation of the other units. In such a case, total system costs would be minimized by building the nuclear plant.

Thus, in this entry, the alternatives to the nuclear power plant are two other baseload plant types— namely, coalfired and combined-cycle natural gas-fired units. By limiting the comparisons to other baseload plant types, the analysis becomes much more tractable and transparent. Unfortunately, by just computing levelized stand-alone costs, many of the renewable technologies must be excluded from the analysis. The stand-alone cost of building and operating a wind farm, for example, can be computed. However, total system costs may not be minimized by building the wind farm even though that plant type has lower stand-alone costs. This is because the effects of the operation of the wind farm and the nuclear unit on the operation of the other units will not be the same.

Second, since fossil fuel prices will probably increase over time, there is a time dimension to the question of whether nuclear power is economic. Because of the recession and utility conservation programs, additional baseload capacity will probably not be needed until around 2020. Thus, in this analysis, the first year of a unit’s operation is assumed to be 2020. By focusing on the mid-term, the carbon capture and storage (CCS) technologies will not be considered. Recently, a CCS task force was formed with a goal of bringing 5 to 10 commercial-size CCS units online by 2016.01 Even if this goal, which is very ambitious, is met, to demonstrate that the technology works, the units would have to operate for probably 5-10 years. In all probably, it would take 15-20 years before the CCS technologies would be commercially available on a widespread basis.

Capital Costs for Nuclear and Coal-Fired Powerplants

One of the major uncertainties in any analysis of the economics of nuclear power deals with the construction cost estimates. This section begins with a discussion about nuclear and coal-fired powerplants’ overnight capital cost estimates. Overnight cost is defined as the cost of building a power plant instantaneously at some point in time. It is also a direct measure of the value of the land, labor, and materials needed to build a nuclear power plant. Thus, differences in overnight costs reflect differences in the values of the land, labor, and material needed to build the same unit.

It is obviously impossible to build a plant overnight, so the second part of this section describes how the total project costs are derived from the overnight costs. To do this, a number of important assumptions are needed, and in many studies, they are not articulated. This section will also show why comparisons of total project costs must be made with great care. The fuel costs for coal and natural gas-fired power plants are discussed in Section 4. The other assumptions will be briefly discussed in Section 6. These include nonfuel operating costs and nuclear fuel costs.