1
Introduction

GENESIS OF THE STUDY

Energy is essential to the functioning of society. From coal for electricity production to oil products for transportation to natural gas for space heating, every aspect of modern life depends on energy. Yet, as beneficial as energy is, its production, distribution, and consumption also have negative impacts especially on human health and the environment.

Most effects of energy are negative, but that does not imply that energy use has an overall negative impact on society. Quite the contrary; the benefits to society of U.S. energy systems are enormous. However, it was not the committee’s task to estimate those benefits that are considered largely to be “internal” because they are reflected in energy prices or government policies.

The purpose of this study was to define and evaluate external effects of energy production, use, and consumption, which refer to costs and benefits not taken into account in making decisions (such as the siting of a power plant) or not reflected in market prices (for example, the price of gasoline at the pump). Under such conditions, the actions that follow might be sub-optimal—in the sense that the full social costs of the actions are not recognized—resulting in a loss of social welfare. When market failures like these occur, there is a case for government intervention in the form of regulation, taxes, fees, tradable permits, or other instruments that will cause economic agents to recognize the external effects in their decision making.

Before such public policies are pursued, the external effects of energy



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 22
1 Introduction GENESIS OF THE STUDY Energy is essential to the functioning of society. From coal for electric- ity production to oil products for transportation to natural gas for space heating, every aspect of modern life depends on energy. Yet, as beneficial as energy is, its production, distribution, and consumption also have negative impacts especially on human health and the environment. Most effects of energy are negative, but that does not imply that energy use has an overall negative impact on society. Quite the contrary; the ben- efits to society of U.S. energy systems are enormous. However, it was not the committee’s task to estimate those benefits that are considered largely to be “internal” because they are reflected in energy prices or government policies. The purpose of this study was to define and evaluate external effects of energy production, use, and consumption, which refer to costs and benefits not taken into account in making decisions (such as the siting of a power plant) or not reflected in market prices (for example, the price of gasoline at the pump). Under such conditions, the actions that follow might be sub- optimal—in the sense that the full social costs of the actions are not recog- nized—resulting in a loss of social welfare. When market failures like these occur, there is a case for government intervention in the form of regulation, taxes, fees, tradable permits, or other instruments that will cause economic agents to recognize the external effects in their decision making. Before such public policies are pursued, the external effects of energy 22

OCR for page 22
23 INTRODUCTION and their monetary values should be known. Thus, Congress directed the U.S. Department of the Treasury in the Energy Policy Act of 2005 (P.L. 109-58), Section 1352, to commission a study by the National Academy of Sciences that would “define and evaluate the health, environmental, security, and infrastructure external costs and benefits associated with the production and consumption of energy that are not or may not be fully incorporated into the market price of such energy, or into the Federal rev- enue measures related to that production or consumption.” Funding for the study was later provided through the Consolidated Appropriations Act of 2008 (P.L. 110-161). STATEMENT OF TASK In response to this mandate from Congress and the request from the Department of the Treasury, the National Research Council (NRC) estab- lished the Committee on Health, Environmental, and Other External Costs and Benefits of Energy Production and Consumption (see Appendix A). The Statement of Task (Box 1-1) was developed and served as the point of departure and guide for the committee’s work. In the remainder of this chapter, we define key terms in the Statement of Task and explain the gen- eral procedures followed in executing the task. This study is one of many related to energy that the NRC has recently undertaken. In the next section, we briefly discuss those NRC studies that have informed our work, especially the America’s Energy Future (AEF) ini- tiative, which is identified in the Statement of Task. We also briefly review previous studies on the external costs of energy. Also in this chapter, we provide the definition of an externality—the focus and core concept of this study—and provide some examples. The Statement of Task directed us to evaluate the externalities “asso- ciated with the production, distribution and consumption of energy from various selected sources.” We explain how we selected the sources and the particular elements of the energy system on which we focused. The approach that we took for identifying, quantifying, and evaluating externalities “in economic terms” is explained. A discussion of “appropri- ate metrics from each externality category” is included. Although the committee was not asked to “recommend specific strate- gies for correcting observable externalities, because those choices will entail policy judgments”—a position with which we agree—it is important to understand and to keep in mind the policy contexts in which our results may be used. The Statement of Task anticipated some of the methodological chal- lenges of evaluating externalities. We discuss the particular difficulties of

OCR for page 22
24 HIDDEN COSTS OF ENERGY BOX 1-1 Statement of Task An NRC committee will define and evaluate key external costs and benefits— health, environmental, security, and infrastructure—associated with the produc- tion, distribution, and consumption of energy from various selected sources that are not or may not be fully incorporated into the market price of such energy, or into the federal tax or fee or other applicable revenue measures related to such production, distribution, or consumption. Although the committee will carry out its task from a U.S. perspective, it will consider broader geographic implications of externalities when warranted and feasible. The committee will not recom- mend specific strategies for internalizing observable externalities, because those choices would entail policy judgments that transcend scientific and technological considerations. In carrying out its task, the committee will include the following activities: S eek to build upon the results of the NRC initiative America’s Energy Future: Technology, Opportunities, Risks, and Tradeoffs. I dentify key externalities to be assessed in the categories of human health, environment, security (including quality, abundance, and reliability of energy sources), and infrastructure (such as transportation and waste disposal sys- tems not sufficiently taken into account by producers or consumers). C onsider externalities associated with producing, distributing, and consuming energy imported from foreign sources. D efine appropriate metrics for each externality category considered. I dentify state-of-the-science approaches for assessing external effects (actual or expected) and expressing their effects in economic terms. D evelop an approach for estimating externalities related to greenhouse gas emissions and climate change. Estimate externalities related to those changes. P resent qualitative and, to the extent practicable, quantitative estimates of externalities and associated uncertainties within a consistent framework that makes the discussion of externalities and uncertainties associated with energy production, distribution, and consumption more transparent. W hen it is not feasible to assess specific externalities comprehensively, the committee will recommend assessment approaches and identify key informa- tion needs to inform future assessments. dealing with space, time and uncertainty. The committee sought to build on the work of companion studies within the NRC, particularly the AEF and America’s Climate Choices studies.

OCR for page 22
2 INTRODUCTION RELATED STUDIES National Research Council Studies With the National Academy of Sciences and National Academy of En- gineering having identified energy as a high-priority topic, it has received attention from many NRC committees, both past and current. These studies were relevant to the work of this committee. We briefly discuss two studies here and cite them throughout our report where appropriate. The AEF’s effort at the NRC was concerned with future technology and its potential for reducing U.S. dependence on oil imports and lower- ing greenhouse gas emissions, while ensuring that affordable energy will be available to sustain economic growth. The AEF’s task was to critically review recently completed major studies on the potential for change in energy technology and use; compare the studies’ assumptions; analyze the currency and quality of the information used; and assess the relative states of maturity of technologies for potential deployment in the next decade. A secondary focus was on technologies with longer times to deployment. A study committee and three panels produced an extensive analysis of energy technology options for consideration in an ongoing national dialogue. Collectively, they analyzed advanced coal technologies; nuclear power; renewable energy technologies (such as wind, solar photovoltaic, and geothermal); energy storage and infrastructure technologies; advanced transportation power-train technologies; technologies to improve energy efficiency in residential and commercial buildings, industry, and transporta- tion; and the technical potential for reducing reliance on petroleum-based fuels for transportation. These topics were addressed for three time frames: present-2020, 2020-2035, and beyond 2035. In response to a request from Congress concerning a related topic, the NRC also launched America’s Climate Choices, a suite of studies designed to inform and guide responses to climate change across the nation (see NAS/NAE/NRC 2009a). The results of the studies were intended to address the following key questions: • What short-term actions could be taken to respond effectively to climate change? • What promising long-term strategies, investments, and opportuni- ties could be pursued to respond to climate change? • What scientific and technological advances (for example, new ob- servations, improved models, and research priorities) are needed to better understand and respond effectively to climate change?

OCR for page 22
26 HIDDEN COSTS OF ENERGY • What are the major impediments—for example, practical, institu- tional, economic, ethical, and intergenerational—to responding effectively to climate change, and what can be done to overcome them? The AEF and America’s Climate Choices studies are important initia- tives that have provided and will continue to provide valuable information on energy technology and policy options for the nation. Indeed, our own study has been informed by the AEF’s analysis of future technology. How- ever, neither of these efforts was designed to focus on the monetary alue of energy’s external effects, including climate change. Prior Externality Studies The concept of externalities dates at least to the early twentieth century (Pigou 1920) and was discussed extensively in the post-war economics lit- erature (Meade 1952; Scitovsky 1954; Mishan 1965; Arrow 1975). Interest in the externalities of energy production and use gathered momentum in the following decades. Hohmeyer (1988) was one of the more prominent studies during this period. He took a top-down approach in which he es- timated the “toxicity weighted” emissions from electricity generation with fossil fuels and then multiplied this fraction by Wicke’s (1986) estimates of total damages from pollution to various end points (such as those on health, forests, and animals). The most prominent study in the United States during this period (Ottinger et al. 1990) used estimates from previous studies that quantified the environmental costs from electric power generation. Results of Niemi et al. (1984, 1987) were among those used by Ottinger et al.; those studies focused on visibility and health effects of airborne particulate matter. Ottinger et al. followed a five-step procedure in using these studies to value environmental damages: emissions, dispersion, exposure, impacts, and damages. Research in estimating the external costs of energy peaked in the early- to-mid-1990s when public utility commissions in the United States were interested in tilting electric utility investment choices toward sources with lower negative externalities, such as renewable energy. This policy was to be done with an “adder” equal to the marginal damages associated with each type of electricity generation. During this wave of interest, major stud- ies were done by Oak Ridge National Laboratory and Resources for the Future (ORNL/RFF) for the U.S. Department of Energy, by Hagler Bailly for the New York State Energy Research and Development Authority (NY- SERDA), by Research Triangle Institute for the State of Wisconsin (one of several states mounting these studies), and by several teams of European research organizations for the European Commission (EC). This latter

OCR for page 22
2 INTRODUCTION study, called ExternE, worked in concert with the ORNL/RFF team to use similar protocols. Around the same time that work began on the ORNL/RFF and ExternE studies, other studies were completed by the Union of Concerned Scientists (1992); Pearce et al. (1992) in their report to the U.K. Department of Trade and Industry; Triangle Economic Research (1995) for Minnesota; National Economic Research Associates (1993) in its study of Nevada; Regional Economic Research (1991) on California; and Consumer Energy Council of America Research Foundation (1993). Later, the EC began a companion study on the external costs of trans- portation as well as other research efforts—the most recent being the New Energy Externalities Development for Sustainability (NEEDS) pro- gram—that further refined and developed methodologies extending those developed in ExternE as well as extending those to other energy technolo- gies and study locations. Also in the transportation area, a series of studies were conducted by Greene et al. (1997) and Delucchi (2004), Parry et al. (2007), among several other studies performed since the mid-1990s. The more notable differences between our committee’s study and previ- ous studies, particularly the major studies done in the early to mid-1990s, are in the different approaches to, and the extent to which, the studies ad- dressed the following: a. Number of power plants—our study considered almost all coal and natural gas power plants in the country, whereas most other studies focused on a few sites or on plants within a state. b. Different power-generation options—our study considered fewer technologies than several of the previous larger studies; in particular, our study did not address the nuclear fuel cycle in the detail done in the ORNL/ RFF and ExternE studies, which carried out extensive probabilistic risk modeling. c. The manner in which the dispersion of airborne pollutants and formation of secondary pollutants were modeled—our study used a reduced form approximation of these processes, whereas some previous models used more site-specific, detailed air dispersion and transformation models, albeit for a small number of power plant sites and regulatory scenarios. The early studies also had no or limited modeling and analysis of ozone and fine (2.5 microns or less in diameter) particulate matter formed from the chemical transformation of pollutants emitted by a power plant. d. Consideration of greenhouse gas emissions and their damages—like all previous externality studies, our study reviewed recent literature rather than undertaking new scientific research; our study reviewed more recent literature than most of the previous studies, although recent studies within the ExternE program used similar literature.

OCR for page 22
28 HIDDEN COSTS OF ENERGY e. Consideration of externalities associated with infrastructure and energy security—of the latter, ORNL/RFF and ExtenE focused only on oil security. f. The extent to which the entire life cycle of an energy technology (from feedstock through conversion through fuel distribution through en- ergy service) was analyzed—the ways in which the different technology fuel cycles were analyzed—the ORNL/RFF, ExternE, and our study considered life-cycle impacts, whereas most other studies focused on electricity genera- tion or use of vehicles in transportation and not on the upstream activities in the life cycle. g. The extent that externalities, other than those associated with elec- tricity generation, were addressed in the same study (that is, transportation and energy used for heat). With some exceptions, the “adders” studies of the early and mid-1990s took a place-based approach to damage estimation of energy. They would posit the construction of a new electricity-generating plant of a particular type at a given location. Each study considered a small number of alterna- tive locations for each plant, generally from two to five. In those studies, the different results calculated for the different plants would reflect the influence of the specific location of the plant on the magnitude of the dam- ages. In contrast, our study calculates the health-related and some of the environmental damages for most of the power plants in the United States and estimates the damages from each. In this respect, it is similar to the U.S. Environmental Protection Agency’s regulatory impact analyses that use the BenMAP model and to the efforts of Muller and Mendelsohn (2007), which, although not studies of externalities per se, are comprehensive in their level of spatial resolution, such as addressing the specific location of all power-plant emissions. The ORNL/RFF and ExternE studies included relatively detailed engi- neering descriptions of the technologies of the power plants, whereas our study and most other studies focused on estimates of emissions from power plants and not on the underlying technologies. In estimating the health and environmental damages, the ORNL/RFF and ExternE studies used different detailed models to predict the dispersion of primary pollutants from the power plants and the atmospheric forma- tion of secondary pollutants, specifically ozone and fine particulate matter. Studies of externalities associated with greenhouse gas emissions gener- ally either focus exclusively on these emissions and the associated climate change, as exemplified by the authoritative reports of the Intergovernmental Panel on Climate Change, or focus on other pollutants. Our study, on the other hand, provides a range of quantitative estimates of the damages from climate change in monetary terms. The previous externality studies either

OCR for page 22
2 INTRODUCTION do not include this important issue in their analysis or, as in the case of the ORNL/RFF and ExternE studies, draw on a more dated and limited scientific literature in this area. Externalities associated with infrastructure and energy security are usually not addressed in other studies. Our report considers electric grid externalities, infrastructure vulnerability to attacks and accidents, and na- tional security. The ORNL/RFF study provided estimates of damages from dependence on foreign oil, and other studies focused on this issue provided updated estimates. However, as discussed in our report, although damages might result from global dependence on oil in a cartel-dominated market, such damages are not considered externalities. The ORNL/RFF and ExternE’s consideration of damages from different parts of the life cycle, for example, coal mining, sets them apart from most other studies, which do not consider externalities on a life-cycle basis. Sev- eral studies have estimated life-cycle emissions of some fuels currently, or prospectively, used in ground transportation; these studies did not attempt to estimate the impacts and associated externalities of these emissions. Our study, on the other hand, takes an energy life-cycle approach somewhat similar to the ORNL/RFF and ExternE studies, but with more updated considerations and data. Although many studies have addressed different aspects of the exter- nalities from energy production, distribution, or use to varying degrees, they have focused on one type of externality (such as health effects), or one particular sector (usually electricity generation or transportation). Also, they generally focused on one part of the energy cycle. In contrast, our study has a relatively comprehensive scope that includes all types of exter- nalities from energy life cycles of both electricity and transportation, as well as from production and use of energy for heat in residential, commercial, and industrial sectors. DEFINING AND MEASURING EXTERNALITIES Defining Externalities External effects or “externalities” are important because failure to ac- count for them can result in distortions in making decisions and in reduc- tions in the welfare of some of society’s members. An externality, which can be positive or negative, is an activity of one agent (for example, an individual or an organization, such as a company) that affects the well-being of another agent and occurs outside the market mechanism. In the absence of government intervention, externalities associ- ated with energy production and use are generally not taken into account in decision making. Box 1-2 provides definitions of the technical terms used in

OCR for page 22
30 HIDDEN COSTS OF ENERGY BOX 1-2 Definitions of Key Terms Much of the nomenclature for the key terms is taken from the damage function ap- proach, which has been the standard approach to examine the costs and benefits of environmental regulations, required by OMB (NRC 2002a). This approach begins with some burden, say emissions, which ultimately has some physical effect; this effect is then monetized and termed damage. The monetary value of reductions in burdens is termed benefits (the opposite of damages). Burdens: Externalities from economic activities are always by-products of those activities, some of which are useful and some of which cause health and environmental effects, for example. The by-products themselves are termed burdens. Emissions of air pollutants and discharges of pollutants into a river are examples of burdens. Effects and Impacts: These burdens have a real effect in the environment, that is, they have a physical component that affects health, damages ecosystems or reduces visibility, for example. Sometimes, as with energy security, the physical component is not directly present. In any event, these physical effects are termed effects or impacts. Damages: Damages are the monetary value of the physical effects, in its simplest form calculated by multiplying the quantity of physical effects of interest by a monetary value for that effect. This monetary value represents, ideally, the population average of the maximum willingness to pay for a unit improvement in this physical metric. That is, it reflects the preferences people have for reducing this physical effect, given their income and wealth. It is analogous to the price people are willing to pay for a product for sale in a market. Benefits are the opposite of damages. this report that bear on externalities. Some examples of externalities using these terms are presented below. An additional illustration is presented in Appendix B. Example 1. A coal-fired electricity-generating plant, which is in compli- ance with current environmental regulations, releases various pollutants to the atmosphere that adversely affect the health of residents. The pollution released by the plant is an example of a negative externality because it con- tributes to health problems for residents. The damage from this pollution is an additional cost of production to society (a “social cost”). If these social costs were not adequately taken into account in selecting the plant’s site or the air pollution control technology that it uses, the true costs of the plant have not have been reflected in these decisions. Example 2. Many thermal power plants use water for cooling; there- fore, they emit heated effluent, which sometimes benefits anglers because fish in cold regions are attracted to the warmer water. Therefore, the fishing is better in the effluent plume. This is an example of a positive externality.

OCR for page 22
31 INTRODUCTION Total versus Average versus Marginal Damages: To be most helpful for policy, we want to estimate marginal damages and compare these to marginal costs to reduce these damages. The marginal damage is the damage that arises from the last unit of emissions or other type of burden. In many cases, marginal damages are constant over the relevant range of emissions. That is, the damage from the last unit is no different than the damage from the first unit of emissions. But, in some cases, say for pollutants that accumulate in the environment, marginal damages grow with more emissions. In any event, for policy purposes, if the marginal damages from the last unit of emissions exceed the marginal costs from eliminating that unit, then it would benefit society to eliminate that last unit, since the damage prevented would exceed the cost of preventing that damage. Total damages, in contrast, are the sum of marginal damages for all units of emis- sions. Average damage is the total damage divided by the number of units of emissions or other burdens in question. Average and marginal damages may equal one another under certain conditions, but in general they are different. In this report, sometimes we assume that they are equal because it is easier to calculate average rather than marginal damages and actual differences are expected to generally be within error margins. Externalities: An externality, which can be positive or negative, is an activity of one agent (that is, an individual or an organization like a company) that affects the well- being of another agent and occurs outside the market mechanism. In the absence of government intervention, externalities associated with energy production are generally not taken into account in decision making. The improved angling is a societal benefit that probably was not reflected in the utility’s decisions about where to site the plant and effluent. This societal benefit does not take into account any other ecosystem changes, which might or might not be seen as beneficial. Example 3. A company is building a new coal-fired power plant in a small community and hires a large number of construction workers. The increase in demand for construction workers drives up the local wage rate and adversely affects homeowners who wish to hire workers to remodel their homes. The price of their remodeling projects has gone up with the increase in wages. This is not an externality, however, because the activity of one agent (the company building the power plant) affects other agents (homeowners wishing to remodel their homes) through a market mecha- nism—the labor market. The company takes the increase in wages into account because it must also pay the higher wages to attract construction workers. Example 4. Farmers respond to a demand for corn-based ethanol by diverting land from food production to fuel production. The reduction in

OCR for page 22
32 HIDDEN COSTS OF ENERGY the supply of feed corn and other grains drives up the prices of grains and meat, thereby making consumers worse off. This is not an externality since the activity of one agent (fuel buyers bidding up the price of corn for etha- nol production) affects other agents (the food-buying public) through the markets for corn and other food products. Example 5. Workers in high-risk occupations receive a higher wage than workers doing similar tasks in jobs with lower risks. This is not an externality because those bearing the risks are freely choosing within a mar- ket to accept this risk and are compensated for the risks they face through higher wages. Increased costs faced by a firm do not by themselves indicate whether the firm’s activities are an externality. For example, electricity- generating plants participating in the Acid Rain Program of the U.S. Clean Air Act face higher costs because they must surrender valuable permits for each ton of sulfur dioxide (SO2) emitted. This higher cost is the result of a government program to reduce the externality associated with acid rain. In the case of elevated costs to compensate for high-risk jobs, no government policy is involved in addressing the risks faced by employees due to the nature of the work they are offered. Externalities matter because, when they are not accounted for, they can lead to a lower quality of life for at least some members of society. For example, suppose that the power plant in the first example has access to technology that, at a cost of $40/ton, can cut its emissions by 10 tons. Suppose further that the full cost of the effects that residents suffer (for example, health and psychological costs) is $50/ton. If the plant were to install the technology, total social welfare would increase—the additional cost to the plant would be $400, but the “savings” to the residents (that is, the reduction in adverse effects they suffer) would be $500. Human well- being would be increased by this change. However, if the externality had not been accounted for in the plant’s decisions, aggregate well-being of all members of society would be lowered. It is important to distinguish true negative externalities from unfortu- nate market signals (such as higher prices of food) that hurt some members of society but are not externalities. The reason for this distinction is that, in the case of a true externality, the possible well-being of society can be raised by accounting for it—the “pie” that represents the value of society’s goods, services, and related intangibles is enlarged. If it is not a true ex- ternality, market intervention cannot alter the size of the pie but can only reallocate it. Note the following additional points about externalities: • The agents that produce externalities can be organizations or in- dividuals. For instance, a restaurant diner who smokes an after-dinner

OCR for page 22
3 INTRODUCTION centrations, climate effects, and, therefore, reduced habitat, but the funda- mental economic valuation question remains conceptually straightforward: How much of other goods and services (such as education, housing, and health care) would people be willing to give up in exchange for preserving polar bear habitat? The conceptual answer will differ depending on whether there are many polar bears left in the wild or whether there are only a few. Whether a monetary value can be estimated that is accurate enough for use in policy decisions remains a challenge for many ecological services. The committee has used WTP to monetize external effects wherever pos- sible, recognizing its limitations and controversies. Some effects are not monetized at all, and others are monetized with great uncertainty. Indeed, some effects cannot even be estimated, much less quantified, even though we know they exist. The committee is especially aware that ecological impacts, including impacts on ecosystem services, have not been monetized in this report. Evaluating these impacts economically has a long and challenging history (for example, EPASAB 2009; NRC 2004a; Cropper 2000). Ecological ef- fects that influence the production of economic goods, such as agricultural products, timber, fish, and recreational benefits, often have been monetized, although often incompletely. This report includes some aspects of agricul- tural production in its monetization of the damages from emissions from electricity generation that contribute to the formation of criteria air pol- lutants. However, changes in ecosystem services, such as nutrient cycling and provision of habitat, and more subtle changes in ecosystem functioning that can affect ecosystem performance have not generally been monetized, largely because it is difficult to quantify those changes at present (for ex- ample, Cropper 2000). Although the committee has described these impacts qualitatively, at least to some degree, they likely are significant monetarily and otherwise. Despite these limits, the commmittee believes that using its results will improve federal policy making. Consideration of External Benefits There are obviously considerable benefits to having energy. Most of these benefits are reflected in the prices paid for energy and are not external benefits. For the most part, external benefits are relatively few in number and small compared with the external damages that have been identified. For example, ORNL/RFF (1992-1998) identified the crop fertilization ben- efits of the nitrogen and sulfur from NOx and SO2, respectively; the crop fertilization benefits of CO2; and the recreational benefits of enhanced fish- ing opportunities in reservoirs formed from large hydro projects. Of those, our study explicitly considered the crop fertilization benefits

OCR for page 22
4 HIDDEN COSTS OF ENERGY of CO2; the results of the integrated assessment models considered in Chap- ter 5 account for the impacts. However, we did not explicitly consider other external benefits for the following reasons: ORNL/RFF (1992-1998) found the crop fertilization benefits of NOx and SO2 to be small compared with the health-effect damages. We did not consider reservoir recreational bene- fits because we did not consider hydropower as important, for the purposes of this study, as the other technologies considered (refer to Chapter 2). THE POLICY CONTEXT FOR THIS STUDY Externalities are important to analyze and understand because they provide an example of a situation where government involvement can potentially improve on the market outcome. Although the committee was not tasked to make recommendations for policy makers to address energy- related externalities, we did indicate how knowledge about the value of externalities can be used to improve market outcomes. This section relates the results of our study to existing policies that address externalities and discusses how the results of the study should and should not be used. The Nature of Externalities Evaluated in This Study As noted earlier in this chapter, the committee evaluates the externali- ties associated with energy production and consumption that have not been corrected through existing policies—that is, the externalities remaining after policies have been implemented. Therefore, the study does not document the substantial progress that has been made in reducing the external dam- ages associated with energy production and consumption over the past few decades. To illustrate, emissions from electric power plants that contribute to criteria pollutant formation are regulated by a variety of state and federal regulations. In particular, one of the goals of Title IV of the 1990 Clean Air Act Amendments was to reduce SO2 emissions from coal-fired power plants by 50% from 1985 levels by the year 2010. Most of the reductions were already achieved by 2005, the year of this study. We quantify the damages associated with remaining SO2 emissions from fossil-fueled power plants in 2005. A similar statement can be made regarding tailpipe emissions from motor vehicles. Emissions from cars per mile traveled have declined by 90% since the passage of the 1970 Clean Air Act as a result of various regula- tions. We calculate the remaining emissions from cars in 2005 and 2030. We evaluate the damages associated with emissions in the years of 2005 and 2030, relative to zero emissions. For example, in the case of coal-fired power plants, we characterize the per plant aggregate damages associated with SO2 emissions in 2005 compared with no SO2 emissions. The same is true of the air-pollution damages associated with motor vehicles: We

OCR for page 22
 INTRODUCTION evaluate the per vehicle total damages from current emissions relative to zero emissions. This is not because emissions should be reduced to zero but because any other baseline would be arbitrary. The appropriate level of SO2 emissions from power plants depends on the costs of reducing those emissions (see Figure 1-1), but estimating the appropriate level of emissions is beyond the scope of this study. The methods used to estimate air-pollution damages from fossil-fueled power plants and motor vehicles assume that the damages of each additional ton of pollution from a source are constant8—hence, we also compute the damages per ton of pollutant, which could be compared with control costs. In the case of power plants, we provide estimates of the distribution of air-pollution damages across power plants. This is important for two reasons: First, the damages associated with a plant depend on where the plant is located, so damages vary spatially; second, total damages vary greatly across plants because of differences in plant size and pollution control. Variation in damages across plants is useful information from the perspective of pollution control. Plants with large total damages may war- rant further air-pollution controls. We also distinguish damages by the stage of the life cycle at which they are generated. Although it is possible to aggregate NOx damages associ- ated with passenger transportation across all stages of the life cycle—oil exploration and extraction, oil refining, transportation of gasoline to the consumer, and consumption of gasoline by a car—regulations to limit NOx emissions will be targeted at different stages of the life cycle: Regulations to limit tailpipe emissions will differ from those to limit oil-refinery emissions. Similarly for damages associated with electricity generation, it is important for policy purposes to separate mining damages from those damages as- sociated with electricity generation because policies to control each set of externalities will differ. Thus, although we present aggregate estimates of damages—per kilowatt hour or per mile traveled—they should be placed into proper context for policy. Policies to Correct Externalities Policies to address or correct externalities include taxes, transferrable pollution permits, performance standards, and technology-based standards. Economic theory dictates that the most efficient policies for correcting 8 This is a common assumption in the air-pollution literature. The concentration-response functions in the literature are essentially linear over the relevant range of ambient air pollution in the committee’s study. Also, the emission-to-concentration relationship and unit costs of various health effects and other impacts are treated as constant. Unit costs are not necessarily constant across time and location.

OCR for page 22
6 HIDDEN COSTS OF ENERGY externalities are those targeted at the externality itself—for example, a tax on SO2 emissions rather than a tax on the electricity associated with those emissions or a tax on NOx emissions from motor vehicles rather than a tax on gasoline.9 Taxing SO2 emissions (or regulating them through a permit market or performance standard) provides an incentive to reduce SO2 by using pollution-control equipment, by switching to low-sulfur coal, or by reducing the level of electricity produced. A tax on electricity generation does not provide the incentives to reduce SO2 emissions per se. The same is true of a gasoline tax and NOx emissions. For emissions related to criteria pollutants, the committee therefore notes that its estimates of externalities associated with emissions per kilo- watt-hour of electricity produced or per gallon of gasoline should not be interpreted as recommendations for electricity or gasoline taxes equal to these monetized damages. Economically efficient methods of correcting emissions that contribute to criteria air pollutants include taxes on the emissions themselves or permit markets in which rights to pollute are denominated in terms of damages.10 A similar statement can be made for CO2 emissions. For fossil-fueled power plants, we provide estimates of the damages per ton for key emissions that contribute to criteria air pollutants, as a function of plant location. For CO2 emissions, we provide ranges of estimates of marginal damages. Externalities and Technology Choice A frequent use of estimates of the externalities associated with elec- tric power generation and transportation is to inform technology choices when making public investment decisions. Should expansion of electricity- generating capacity take the form of coal, natural gas, nuclear power, or wind power? What technologies should be pursued as alternatives to gasoline-powered internal combustion engines for passenger vehicles? This study can help to inform such choices; however, it must be emphasized that we evaluate the externalities associated with various technologies indepen- dent of their costs. For example, an integrated gasification-combined cycle (IGCC) coal plant with carbon capture and storage is an extremely clean plant, but it is also an expensive one. Externalities are an important com- 9 Policies that associate a price with the externality—for example, a tax or a permit mar- ket—are, in general, more efficient than policies that dictate the method of correcting the externality; for example, requiring coal-fired power plants to install flue gas desulfurization units (scrubbers). 10 For example, if a power plant in a densely populated area creates more damages per ton of SO2 emitted than a power plant in a remote area, the former plant would require more damage-denominated permits than the latter to emit a ton of SO2.

OCR for page 22
 INTRODUCTION ponent of the choice among various technologies but must be supplemented by estimates of private costs. SOME METHODOLOGICAL ISSUES: SPACE, TIME, AND UNCERTAINTY Defining and evaluating externalities is unavoidably complicated by their spatial nature, by the fact that they manifest themselves over time, perhaps very far into the future, and by uncertainty. We discuss each of these issues in this section. Spatial Scales of Analysis The external effects of energy, by their very nature, vary spatially. Some individuals and groups experience far greater effects from energy produc- tion and use than is reflected by the average amount—that is, than if the effects were evenly distributed—and others experience far less. In carrying out the committee’s task, we focused on the spatial distribution of dam- ages caused by coal-fired and gas-fired power plants wherever they were located and by transportation emissions in each of the U.S. counties in the 48 contiguous states. Note, however, that a lack of location data for stages upstream of the power plants and vehicle operations prevented us from estimating these kinds of damages in a spatially explicit manner. Consideration of Effects on U.S. s. Global Scales Although the committee’s task stipulated that the external costs and benefits of energy be analyzed from a U.S. perspective, we were also charged with consideration of broader, more global, implications when warranted and feasible. Some effects considered by the committee occur mainly in the United Sates, such as effects related to ozone-forming emissions from motor vehicles. However, other effects, such as those related to CO2 emis- sions and climate change, will occur on a global scale. Likewise, for some of the security-related issues, or for transportation, which relies on energy production and distribution occurring outside the United States, ignoring the global consequences would result in substantial distortions. Moreover, as is apparent for climate change-related effects, some parts of the world are likely to suffer inherently different, and to some extent larger, burdens of these effects than the United States. In such situations, we have elected to characterize effects both in the United States and on a global scale, as consideration of them on different spatial scales might have an impact on policy choice. For practical reasons, we have provided sparing detail regard- ing differential impacts among non-U.S. regions.

OCR for page 22
8 HIDDEN COSTS OF ENERGY Consideration of Differential Effects on Local and Regional Scales Within the United States Within different locations in the United States, many external costs and benefits related to energy are heterogeneously distributed as well—for reasons inherent to the nature of the economic activity or the geography or as a consequence of one or another policy choice. For example, one of the substantive health consequences of climate change in the United States is the impact on heat- and cold-related morbidity and mortality, for ex- ample, heat waves. These impacts are far stronger in northern cities with moderate climates within which temperatures fluctuate widely year to year. Because of differences in the extent of human physiologic adaptation to higher temperatures, more people die in heat waves in Chicago than, say, in Birmingham, Alabama, and rising average temperatures will accentuate that disparity further. Likewise, because of greater population density and prevailing winds, the distribution of harmful effects from emissions that form criteria air pollutants is highly nonhomogeneous. For example, popu- lations in eastern seaboard counties bear more of the health-related external costs of this external impact of electricity production from fossil fuels than do populations in upwind areas and will continue to irrespective of any short-term policy choices. Thus, when aggregate damages are presented, the differential impact may be partly obscured. For other impacts, such as the local—potentially devastating—effects of a power-plant disaster or disruption occurring in a distribution line (for example, an oil or gas line), local choices may be extremely important in determining “who pays.” Often, siting of these types of facilities is partially determined by geographical factors, such as where production and utiliza- tion actually occur. In some situations, aggregate damages may be juxtaposed against local damages, creating not only heterogeneity but also complex policy alterna- tives. For example, there is at least some evidence that centralized, rather than decentralized, management of spent nuclear fuel results in an inher- ently lower risk of adverse external consequences; yet arguably for the site or sites chosen for a centralized activity the local “costs” can be higher. Another similar consideration is that the damages of power-plant emissions vary by the population affected by the emissions. Differences in Susceptibility oer Spatial Scales Even within the same locations, there is compelling evidence that some parts of the human population or that some species within an ecosystem are more vulnerable than others to a particular external effect. One of the factors responsible for differential effects is age; the very young and the very

OCR for page 22
 INTRODUCTION old are more susceptible to energy-related burdens, such as those imposed by heat stress, water constraints, or pollution. Likewise, the underlying health status of individuals or groups creates large disparities in effects. In highly developed societies, risks of the nutritional consequences of climate change or diarrheal illness are essentially nil, while these impacts dominate in societies with lower levels of overall health status. Conversely, air pol- lutants from electricity or transportation tend to affect, to a greater degree, individuals and societies with higher underlying rates of cardiovascular and chronic disease, which are more prevalent in richer societies. This same factor differentiates the consequence to an individual with chronic disease from his/her healthy partner, even living in the same house. These conditions may be confounded by disparities created by differential access to resources, for example, socioeconomic differences, within a nation or region. For example, during the last highly publicized heat wave—Chicago 2003—almost all of the excess deaths occurred among poor minorities without air-conditioning or ready access to health or social services. Once again, aggregate cost data would tend to mask, rather than emphasize, such differences. Temporal Issues Some effects related to the production and use of energy may take years, decades, or longer to manifest themselves. For example, chronic health effects of air pollution attributable to fuel combustion are not the consequence of an exposure that occurred yesterday or a few weeks ago, but they are the cumulative result of conditions that develop over longer periods. As a more extreme example, health risks from the disposal of nuclear waste generated from electricity production may persist over mil- lennia because of the long-lived nature of the radioactive waste. This per- sistence presents challenges in making judgments about the performance of a waste repository, the behavior of human society, and other key factors over a very long period. One challenge is that it is very difficult to predict both the future physical effects and their monetary values because they depend on a host of uncertainties about how people in the future will live. A second chal- lenge arises in comparing effects that are quantified in monetary values at different times (such as expenditures on control equipment now and fewer adverse health effects in the future). In making such evaluations, two factors should be considered. One is that many opportunities exist for investing resources now to yield future benefits. The future benefits of a proposed action should be compared with the future benefits that could be achieved by investing the same resources in other ways. The other factor is that the

OCR for page 22
60 HIDDEN COSTS OF ENERGY people affected may differ, especially if the delays are long enough that they are necessarily members of different generations. It is conventional and appropriate to discount future values by a fac- tor that depends on the distance into the future and the discount rate. (It is also necessary to account for future inflation, usually accomplished by valuing all consequences in “real”—that is, “constant” or “inflation- adjusted”—dollars.) In addressing the difficulties that arise in identifying the appropriate discount rate, two approaches are often used. The first is commonly referred to as “descriptive”; the second, as “prescriptive.” The descriptive approach uses a discount rate that is similar to market interest rates, which are market prices that are determined by the interactions of individuals, firms, and other institutions seeking to borrow or save for vari- ous time periods. The prescriptive approach is often used for time periods of more than about 30 years, for which market interest rates rarely exist. This approach explicitly considers two factors: the rate at which future generations’ utility should be valued relative to the current generation’s utility (an ethical question), and the rate at which incremental resources will enhance the future generations’ utility (a descriptive question) (see Chapter 5 for further discussion). Estimates of the appropriate discount rate derived from the prescrip- tive approach are typically smaller than those derived from the descriptive approach. This divergence raises a number of ethical questions, such as whether individuals and governments currently are consuming too much and investing too little and how much individuals should sacrifice now to potentially benefit many future generations. For valuation of climate-change effects (see Chapter 5), the discounted value referred to as the social cost of carbon is often used. It is the present- day value of the combined damages and benefits that will occur over many future years if an additional ton of greenhouse gas is emitted today. Esti- mating the discounted cost involves consideration of current greenhouse gas emissions’ effects on climate over the next century or more, environmental and human welfare effects caused by climate change, how the effects may vary globally, the course of future economic development, the range and likelihood of economic and social effects arising from climate change, and the extent to which human society might adapt to climate change. Because the choice of a discount rate for such long periods involves great uncer- tainty, the committee does not recommend a particular discount rate for assessing the value of these effects. Model Selection and Evaluation The committee made extensive use of computational models to evalu- ate available knowledge, compare alternative technologies, and provide a

OCR for page 22
61 INTRODUCTION framework to assess damage. The committee recognizes that all models face inherent uncertainties because human and natural systems are more complex and heterogeneous than can be captured in a model. Moreover, the committee also recognizes that once a model is selected and applied, large uncertainties remain regarding input selection and choices of scale. In its selection and use of models, the committee relied on a report of the NRC’s Committee on Regulatory Environmental Models (NRC 2007a), which recommended that models cannot be validated (declared true) but instead should be evaluated with regard to their suitability as tools to address a specific question. In following this approach, the committee first identified its specific questions, then identified the tools available, and finally made model selections. The committee recognized that its analysis involved five key activities: (1) characterizing a range of technologies that provide electricity, transpor- tation, and heating; (2) identifying the pollutant emissions (and other envi- ronmental hazards) attributable to each technology; (3) linking emissions (hazards) to exposures; (4) linking exposures to effects; and (5) translating effects into damages that can be monetized. Modeling was required for electricity production and heating—steps 3, 4, and 5—and for transporta- tion impacts—steps 2, 3, 4, and 5. Therefore, the committee reviewed a number of models that could support this task and considered several alter- natives, including several models to address the issues of model uncertainty. Ultimately, the committee determined that the use of a single model would make its results more transparent and open to evaluation than would trying to interpret results from several models. The committee selected the APEEP (Air Pollution Emission Experiments and Policy) model (see Chapter 2) for steps 3, 4, and 5 and the GREET (Greenhouse Gases, Regulated Emis- sions, and Energy Use in Transportation) model (see Chapter 3) for step 2 in transportation technologies. In making these choices, the committee did not consider these two models to be the only or even the best models for this task. Instead, the choice reflects the committee’s recognition that these models were clearly appropriate for the task, were accessible to the com- mittee, were transparent in their applications, and had received sufficient prior use and performance evaluation. To further evaluate the performance of these models for use in calculating external impacts, the committee car- ried out comparative evaluations where that was feasible. Intake Fraction and Other Tools for Model Ealuation The committee sought other studies with comparable results to evalu- ate the consistency of its model approach with approaches used by others engaged in similar research. In making these evaluations, the concept of “intake fraction” was useful and transparent. It is defined by Bennett et

OCR for page 22
62 HIDDEN COSTS OF ENERGY al. (2002) as the integrated incremental intake of a pollutant summed over all exposed individuals and occurring over a given exposure time, released from a specified source or source class, per unit of pollutant emitted. Since that time, numerous studies have estimated intake fractions for various source categories (such as power plants, mobile sources, residential wood burning, indoor cleaning products, and aircraft) and pollutants (such as particulate matter and toxic air pollutants). Most important, use of the intake fraction approach has increasingly become a tool for model perfor- mance evaluation and model comparisons. For source-receptor estimates from power plants, work by Nishioka et al. (2002) provided a model evaluation opportunity. To assess the health effects of increased pollution, Nishioka et al. (2002) modeled state-by-state exposures to fine particulate matter (PM2.5) originating from power-plant combustion and used intake fraction as an intermediate output. The com- mittee was able to compare its power-plant intake fraction obtained from APEEP with theirs and got consistent results. Moreover, Nishioka et al. (2002) multiplied their population-weighted exposures derived from in- take fractions by exposure-response functions for premature mortality and selected morbidity outcomes, providing the committee with further oppor- tunity to evaluate APEEP results. In the transportation impact modeling, there were two studies that provide key evaluation opportunities. In an effort to better characterize the relationship between mobile-source emissions and subsequent PM2.5 exposure, Greco et al. (2007) characterized PM2.5 exposure magnitude and geographic distribution using the intake fraction. They modeled total U.S. population exposure to emissions of primary PM2.5 as well as particle precursors SO2 and NOx from each of 3,080 counties in the United States. Their mean PM2.5 intake fraction was 1.6 per million with a range of 0.12 to 25 per million compared with 1.0 per million with a range of 0.04 to 33 per million obtained from APEEP. Greco et al. (2007) concluded that long- range dispersion models with coarse geographic resolution are appropriate for risk assessments of secondary PM2.5 or primary PM2.5 emitted from mobile sources in rural areas but that more-resolved dispersion models are warranted for primary PM2.5 in urban areas because of the substantial contribution of near-source populations. One of the advantages of APEEP is better spatial resolution in urban counties, but it may still lack the necessary level of spatial detail, giving rise to some uncertainty about results. Marshall et al. (2005) used three alternative methods to estimate intake fractions for vehicle emissions in U.S. urban areas. Their best estimate of the urban intake fraction for diesel particles was 4 per million, results that are consistent with the urban-county results in APEEP. However, the need for future efforts to provide exposure resolution below the county scale remains a priority.

OCR for page 22
63 INTRODUCTION Addressing Uncertainty Assessment of uncertainty in model outputs is central to the proper use of model results for decision making. There are a number of uncertainties that arise in the calculation of damages from energy use. The committee elected to confront uncertainty using approaches recommended by the NRC’s Committee on Regulatory Environmental Models (NRC 2007a). This committee considered the use of probabilistic (Monte Carlo) ap- proaches to quantify all uncertainties to be problematic in many situations, especially when uncertainty analysis is used to reduce large-scale analyses of complex environmental and human health effects to a single probability distribution or when uncertainty is dominated by decision variables, as is the case for this current study. In this study, uncertainty is dominated by such factors as the selected value of a statistical life, which cannot eas- ily be captured in a probability distribution. In situations where detailed probabilistic modeling is not appropriate, the models committee (NRC 2007a) recommended the use of scenario assessment and sensitivity analy- sis. The current committee chose to use this approach, and where feasible, it has used alternative scenarios and sensitivity analysis to characterize uncertainties. ORGANIZATION OF THE REPORT The discussion in Chapter 2 focuses on the external effects and their valuations, resulting from electricity generation. Chapter 3 addresses exter- nalities related to the production and use of transportation fuels. Chapter 4 discusses energy used to supply heat for industrial processes and to heat indoor spaces. Chapter 5 addresses effects attributable to climate change and their valuations. Chapter 6 discusses effects and valuations related to infrastructure and security. Chapter 7 presents overall conclusions from the committee’s evaluations, including a comparison of climate and nonclimate damage estimates, and discusses factors to keep in mind when interpret- ing the results of the evaluations. Chapter 7 also recommends research to inform future consideration of various issues in this report.