Growing competition in the electric power industry is raising questions regarding the future of utility demand-side management (DSM) programs. This article¹ addresses changes in the growth and character of electric utility DSM and how growing competition and the imminent restructuring of the electric power industry may affect utility DSM practices.

From 1989 through 1993, data collected by the Energy Information Administration (EIA) showed a steady increase in utility DSM spending and in energy and demand savings. The most recent data collected (1994) show that the industry is reducing DSM spending and experiencing a reduction in the rate of growth in energy savings. In 1994, utilities reported modest reductions in energy savings and potential peak reductions. However, utility projections for 1995 show approximately a 40-percent reduction in the growth of energy savings and lower potential peak load reduc tions from DSM programs.

Among other factors, the potential for restructuring in the electric power industry could affect utilities’ interest in energy savings. In a deregulated market for genera tion services, vertically integrated utilities will have an interest in selling more energy at higher prices. DSM programs that reduce consumption may place down ward pressure on prices. Restructuring also may create new types of DSM activities. A growing number of utilities are experimenting with two-way communi cation systems that provide customers flexible time-of-use or real-time pricing and energy information services.

The Development of Utility DSM

Electric utility DSM refers to programs implemented by utilities to modify customer load profiles. Such programs have a variety of objectives.

• Energy-efficiency programs reduce energy use, both during peak and off-peak periods, typically without affecting the quality of services provided. Such programs substitute technologically more advanced equipment to produce the same (or a higher) level of end-use services (e.g., lighting, heating, cooling, drive power, or building shell) with less electricity.
• Peak load reduction programs focus on reducing load during periods of peak power consumption on a utility’s system or in selected areas of the transmission and distribution grid. This category includes interruptible load tariffs, time-of-use rates, direct load control, and other load management programs.
• Load shape flexibility can be achieved by programs that modify prices, cycle equipment, or interrupt service in response to specific changes in power costs or resource availability. These approaches include real-time pricing and time-of-use rates for pricing periods that have flexible hours. They also may include interruptible load tariffs, direct load control, and other load management programs when those activities are not limited to peak load periods.
• Load building programs are designed to increase use of electrical equipment or shift electricity consumption from peak to off-peak hours thereby increasing total electricity sales. This category includes valley filling programs that increase load during off-peak periods and programs that introduce new electric technologies and processes.

The Public Utility Regulatory Policies Act of 1978 (PURPA) identified and helped to focus attention on the benefits of “increased conservation of electric energy” and “load management techniques.”² A series of studies over the last 18 years identified and quantified a large potential to increase the efficiency of energy use.³ Responding to this potential, State regulators supported and utilities implemented rebate and other DSM programs. Many DSM programs areviewed as resources because they capture cost-effective energy savings that would not otherwise be achieved. Most DSM programs are planned in an integrated resource planning (IRP) framework in which utilities compare the benefits and costs of DSM with the cost of additional generation. Utility IRP’s are subject to State regulatory review. Approximately half of the State regulatory commissions seek to reduce disincentives to utilities implementing DSM programs that result from conventional rate design practices. Given conventional rate designs, volumetric rates often are set above utilities’ short-run marginal costs.⁴ As a result, when utilities lose potential sales as a result of consumers using energy more efficiently, revenues and profits go down. State commissions address this problem by using: (1) net lost revenue adjustment mechanisms that allow utilities to recover revenues lost as a result of conservation programs net of any cost savings; (2) revenue decoupling that separates utilities’ profit ability from the levels of actual sales; or (3) DSM performance incentives that are paid to utilities based on the savings achieved⁵ (Figure FE1).

Electric energy savings and load reductions cannot actually be measured by metering and therefore must be estimated. Utilities report estimates of energy sav ings and peak load reductions based on engineering methodologies, statistical analysis of energy usage, and/or other estimation techniques. The estimated en ergy effects are subject to subsequent verification, as required by many State public service commissions. An EIA report⁶ concluded that while estimated savingsin some cases exceeded subsequently verified results, a large variance between estimated and verified savings was not found. The estimated data on DSM programs are reported to EIA annually on Schedule V, “Demand-Side Management Information,” of the Form EIA-861, “Annual Electric Utility Report.” For reporting purposes, DSM programs are categorized as energy efficiency, direct load control, interruptible load, other load management, other DSM programs, or load-building activities. Large utilities⁷ report for each program category and customer class estimated data for:

• Incremental energy effects and incremental peak load reductions—the effects caused by new program participants and new DSM programs during a given year. Incremental effects are “annualized”; that is, they are reported as if they were in effect for the entire year.
• Current and projected annual energy effects and peak load reductions—the total effects and peak load reductions caused by all participants (new and existing) in all DSM programs (new and existing) that are in effect during a given year. This includes the energy effects caused by programs initiated in prior years that are still in effect and programs that were terminated, but are still producing energy effects and/or peak load reductions. These data are reported for the reporting year, next year, and fifth year following the reporting year.
• Current and projected annual costs—the costs of DSM programs for the reporting year, the next year, and fifth year following the reporting year.

In addition, the type of energy-efficiency end-uses and programs offered in each customer class are collected.

From 1989 through 1993, utility DSM programs ex hibited steady or accelerating growth in energy savings and utility expenditures (Figure FE2). The largest share of utility expenditures and energy savings was associated with energy-efficiency programs. These programs supplied substantial peak load reductions, although large potential peak load reductions also occurred as a result of interruptible load programs.

Growing competition is becoming a major influence in the generation segment of the electric power industry. By the early 1990's, the exhaustion of economies-of-scale for large baseload generation,⁹ efficient modular generation technologies (particularly combined-cycle units and aero-derivative turbines), low natural gas prices, and emerging information and control technologies began to make competition possible. Changing regulatory policies facilitated competition among generation suppliers. By the end of 1992, competitive bidding for new power supplies was approved in 20 States and was under consideration in 9 others.¹⁰ Also, the Federal Energy Regulatory Commission (FERC) approved “market-based” pricing for some wholesale power sales,¹¹ and Congress broadened the scope of wholesale competition with the passage of the Energy Policy Act of 1992 (EPACT).¹² From 1989 to 1993, the number of qualifying facilities and other independent power production facilities (5 megawatts or more nameplate capacity) increased from 825 to 1,341, and their installed generating capacity increased from 36.6 to 59.1 gigawatts.¹³ In 1992, for the first time, generating capacity added by independent power producers exceeded capacity added by traditional electric utilities.¹⁴

Within this context of technological and regulatory change, proposals are being made by the members of the industry, regulators, and consumers to restructure the industry, potentially deregulating generation and allowing retail customers access to competitive generation markets. Three factors contribute significantly to the consideration of restructuring:

• Demand, primarily by industrial and large commercial customers, for lower prices and retail access: Differentials between embedded generation costs and wholesale spot prices for generation create the perception that consumer prices can be lower if customers gain access to wholesale power markets. Figure FE3 provides a comparison of the generating costs embedded in utility rates (highest cost utility, regional average, and least cost utility) and wholesale peak period spot prices for selected North American Electric Reliability Council regions.¹⁵ For most utilities, the embedded cost of generation that is built into their rates exceeds the wholesale spot price.¹⁶ Moreover, within any given region, there are significant differentials between the generation costs of high and low cost utilities. These differentials do not imply that utilities have been imprudent, but they do contribute to the per ception that retail prices include uneconomic generation costs.
• Implementation of the Energy Policy Act of 1992: EPACT provided Federal regulators the authority to order utilities to provide transmission access for the purpose of facilitating competition in wholesale power markets. FERC’s implementation of EPACT is illustrated by (1) its expansive notice of proposed rulemaking on wholesale competition;¹⁷ (2) its transmission access and pricing policy statement establishing a “golden rule” of comparability between transmission pricing for a utility’s own sales and transmission pricing for third parties;¹⁸ (3) its Notice of Proposed Rulemaking on Stranded Costs which addresses the treatment of historically incurred costs that cannot be recovered at market prices;¹⁹ (4) its encouragement for the formation of regional transmission groups; and (5) its require ment that transmission utilities, power pools, or electric reliability councils submit data on their transmission capabilities.²⁰
• The perception that competitive generation markets can work and produce economic efficiency benefits: Interest in electric industry restructuring is supported by the successful privatization or restructuring of electric utilities in the United Kingdom, Norway, New Zealand, Australia, Chile, and Argentina,²¹ and the relative success of restructuring in the natural gas, telecommunications, and other U.S. industries.²²

Electric industry restructuring is currently receiving active legislative or regulatory consideration in approximately three-quarters of the States.²³ The consideration of restructuring is focused on competition in the generation portion of the electric power industry. A retail access plan was approved by the California Public Utilities Commission. Modest retail wheeling experiments, in which large customers will be able to purchase generation services directly from competitive generation suppliers, were approved in Michigan and New Hampshire.

Full retail competition will mean that consumers may choose their generation suppliers and that there will be competition in generation services, in financial contracts used to hedge the risk of future volatility in generation prices, and perhaps in certain services related to coordinating the operation of generating units. Electric distribution, transmission, and at least certain dispatch and coordination services historically have been and will continue to be regulated.

Distinguishing functions of the industry in which there will be competition from those in which competition will be limited is important to understanding the potential opportunities for DSM in a restructured electric power industry. If restructuring proceeds, energy-efficiency incentive programs could be supported through non-bypassable charges paid by the customers of regulated transmission and distribution companies. Other DSM services could be paid for by participating customers and provided by competitive energy service companies or packaged with generation and financial services by competing power marketers. The packaging of energy management, generation, and financial hedging services might emerge as the basis for an independent retail business involving new participants in a competitive retail access market structure. However, this article will examine the narrower issue of impacts on electric utility DSM activity.

The latest data on DSM activities filed by electric utilities on Form EIA-861 are for 1994. Those filings also provided projected data for 1995 and 1999 for large utilities with sales to ultimate consumers or sales for resale greater than or equal to 120,000 megawatthours (MWh). Additionally, several utilities provided qualitative information on how increasing competition in the electric power industry is affecting their DSM programs.

The 1994 Program Year

Data compiled from responses on Form EIA-861 revealed moderate changes in utility DSM activity during the 1994 program year. Incremental energy savings decreased 8.4 percent from the 1993 level of 8,980 million kilowatthours (kWh) to 8,229 million kWh in 1994. Incremental potential peak load reductions decreased 17 percent from 7,137 megawatts (MW) in 1993 to 5,904 MW in 1994.²⁴ For the first time since EIA began tracking DSM activity, utility DSM expenditures decreased approximately 1 percent from $2.74 billion in 1993 to $2.72 billion in 1994. In 1993, utilities projected that 1994 DSM spending would exceed $3 billion.

A portion of the decreases in incremental energy savings and potential peak load reductions was anticipated in the utilities' 1993 projections of 1994 annual energy effects and peak load reductions. Annual energy savings in 1994 were 52,483 million kWh. In 1993, utilities projected 1994 annual energy savings of 52,655 million kWh. Annual potential peak load reductions in 1994 were 42,917 MW, exceeding the utilities' projections for 1994 of 42,220 MW. 1994 energy effects approached or exceeded the 1993 projections for 1994, suggesting that the reported decreases in incremental energy effects and peak load reductions represent a change in DSM activity, and are not the result of program evaluations completed since the filing of the prior year's Form EIA-861 data.

Most of the decreases in incremental energy savings occurred in energy-efficiency programs. However, all other program categories showed large percentage decreases in incremental energy savings. Interruptible load programs had the largest decreases in incremental potential peak load reductions, and percentage decreases in incremental potential peak load reductions also occurred in interruptible load, direct load control, and other load management programs. Other DSM programs showed an increase in incremental potential peak load reductions (Figure FE1)

Energy-efficiency programs accounted for 70.6 percent of direct DSM spending in 1994. The 1994 data continue to indicate that the cost to utilities of most energy-efficiency programs is competitive with or below the cost of new generating capacity. The cost of conserved energy in cents per kWh saved is a convenient index for making approximate comparisons between the cost of energy-efficiency programs and generic supply-side resources. The cost of conserved energy is the average life cycle cost of an efficiency measure or program expressed in cents per kWh saved over the life of the measures installed. Figure FE4 presents the average cost per kWh saved for the energy-efficiency programs of large utilities.²⁵

The DSM programs of 63 percent of reporting utilities had average costs of conserved energy under 3 cents per kWh (Figure FE4).

The modest reductions in 1994 DSM savings and expenditures might be explained by the fact that interest in restructuring accelerated rapidly after the issuance of the California Blue Book in April 1994, one of the first proposals for deregulation of generation and significant retail access.²⁶ By April, many utilities had already set DSM program budgets for 1994. The full impact of concerns about restructuring on DSM activity may be observed first in data for the 1995 program year.

Table FE1. Incremental Energy Effects and Potential Peak Load Reductions by Program Type

Program Type	Incremental Energy Savings (Gwh/Year)		Change In Incremental Energy Savings (percent)	Incremental Potential Peak Load Reductions (MW)		Change In Incremental Potential Peak Load Reductions (percent)
	1993	1994	1993-1994	1993	1994	1993-1994
Energy Efficiency__	8,472	8,054	-5	1,839	1,751	-5
Direct Load Control	25	15	-40	1,297	884	-32
Interruptible Load	75	12	-84	3,536	2,822	-20
Other Load Management	19	7	-63	371	282	-24
Other DSM__	389	141	-64	94	165	+76
Total	8,980	8,229	-8	7,137	5,904	-17
Source: Energy Information Administration, Form EIA-861, “Annual Electric Utility Report.”

Projections for the 1995 Program Year

The utilities' projections of annual energy effects and peak load reductions for 1995 suggest that substantial reductions in DSM activity could be under way (Figure FE5). There are, however, some important caveats re garding the reported data. Large utilities are asked to report projected annual energy savings, annual peak load reductions, and program costs for 1995 and 1999. “Annual effects” for 1995 and 1999 represent the continuing impacts of past, current, and projected years’ participation in DSM programs. Year-to-year changes in annual effects can approximate mod ifications in DSM programs, though they may be influenced by factors unrelated to DSM activity for that year (i.e., large customers going out of business, revisions as the result of evaluation of DSM programs, or economic factors). Utilities currently do not report projected incremental effects, which would more closely track the impacts of planned DSM activity occurring in the year that the data are reported.

Annual energy savings in 1995 are projected to equal 52,831 million kWh per year, 0.7 percent above the annual energy savings reported for 1994. Annual 1995 potential peak load reductions are projected to decline by 2.6 percent from 1994 levels to 41,784 MW.The projections of annual effects represent the cumula tive impacts of all prior DSM activity and new activity in 1995. The stagnation of annual effects in 1995 is a major departure from the year-to-year growth reported in prior years.

The reduced growth in annual effects is partially attrib utable to the reporting practices of utilities. Significant declines in annual energy savings from 1994 to 1995 were noted on a number of individual utility reports. This was unexpected because “annual” energy savings reflect the cumulative effects of prior program years. These utilities were contacted for clarification of their reported data. In some cases, utilities had stopped in cluding annual energy savings of measures that remained in place, but were installed under DSM programs that were terminated. The extent of this under-reporting of annual energy savings for 1995 could be as great as 3,500 million kWh. Even assuming under-reporting of this magnitude, the rate of growth in annual energy savings in 1995 would decline by 40 percent. Utilities that reported significant decreases in potential peak load reductions also were contacted. Under-reporting of the continuing effects of terminated energy-efficiency programs had a much smaller impact on potential peak load reductions. Even after cor recting for possible under-reporting, potential peak load reductions declined in 1995. The remaining decreases in growth of annual effects after adjusting for reporting issues suggest that when 1995 data are reported later this year, significant decreases may be observed in incremental energy savings and peak load reductions.

DSM spending is projected to fall at a much slower rate than the growth in annual energy and peak load effects. DSM spending for 1995 is projected to decline from 1994 levels by 4.5 percent to $2.6 billion. This modest decline suggests that utilities are retaining the capa bility to implement DSM programs. Another possible explanation is that DSM budgets are perhaps being reassigned to customer service functions that are as of yet not clearly defined.

Annual energy savings from energy-efficiency pro grams are projected to continue growing, although at a slower rate, from 49,720 million kWh per year in 1994 to 51,221 million kWh in 1995. The reductions in DSM are not limited to energy-efficiency programs. Annual peak load reductions from energy-efficiency programs are expected to increase from 11,662 MW to 11,731 MW. For interruptible load programs and other DSM, utilities project reductions in annual peak load and energy effects in 1995. For direct load control programs, decreased potential peak load reductions are projected for 1995 (Figure FE6) and (Figure FE7).

These findings show a greater decline in energy savings and peak load reductions than suggested by an earlier study.²⁷ The study projected that the 1994 to 1998 decline in the rate of growth of cumulative energy savings would be less dramatic than the decline in DSM expenditures and that the growth in cumulative peak load reductions would come closer to matching recent historical experience. The study, completed in early 1995, relied on a smaller survey of 37 selected utilities and 22 State regulatory commissions. Each of the 37 utilities included in the survey spent at least $5 million on DSM in 1994, making them among the largest in the industry. The study did not regard the sample as representative of all U.S. utilities.

Possible explanations for a decline in DSM activity in 1995, supported by the qualitative data provided by electric utilities, include:

• Low avoided costs may make some DSM options no longer cost-effective. This explanation is con sistent with the increase in annual effects that is projected for 1999, when some utilities will require additional capacity.
• To reduce rate impacts of DSM programs, utilities may be lowering energy savings targets or placing more emphasis on benefit/cost tests that measure rate impacts, as opposed to reductions in customer or societal costs. For many utilities, negative rate impacts are primarily the result of revenue losses created by existing rate design practices whenever sales decline.
• Some utilities report that they are shifting from rebate, low-cost loans, and other financial incen tive programs to information and conventional financial programs. Information and conventional financing programs simply may be less effective than rebate and financial incentive programs in achieving savings over and above the savings that naturally occur in the absence of DSM programs.

The annual effects projected for 1995 raise serious questions about utilities’ commitments to cost-effective DSM opportunities. In a qualitative assessment of the impact of increasing competition on their DSM programs, several utilities suggested that, to date, competition is having little or no impact on their current DSM activities. Other utilities indicated that programs were being cut and that they were reducing or eliminating programs that incorporated rebates or other financial incentives. Additional data collection and analysis are needed to fully explain the decline in the growth of annual effects projected for 1995.

Projections for the 1999 Program Year

Year-to-year growth in annual effects is predicted by electric utilities to rebound to some extent by 1999. Projections exhibit growth in both annual energy savings and annual potential peak load reductions, compared with 1994 and 1995. This may reflect that some utilities are approaching the time when new capacity will be required.

The projected growth in annual energy savings is open to question, however, because of possible under-reporting of energy savings from terminated DSM programs. It is difficult to estimate to what extent under-reporting affects 1999 data, given that some previously installed measures may reach the end of their useful lives between 1995 and 1999. To the extent under-reporting had a greater impact on 1995 than on 1999 projections, the represented data may overstate the average annual 1995 to 1999 rate of growth in annual energy savings. As was the case for 1995, only projected “annual effects” data are available to represent 1999 energy savings and peak load reductions.

Utilities projected 1999 annual energy savings of 71,883 million kWh per year and potential peak load reductions of 51,487 MW. This represents an 8.0 percent average annual rate of growth in energy savings, and a 5.4 percent average annual rate of growth in potential peak load reductions from reported 1995 levels. These projections are lower than the projections made by the same utilities in 1993 for 1998 energy savings (88,978 million kWh in 1998) and potential peak load reductions (55,163 MW in 1998). Projected annual energy savings for energy-efficiency programs increased from 51,221 million kWh for 1995 to 69,825 million kWh for 1999.

DSM spending is projected to continue to decline, from $2.6 billion in 1995 to $2.5 billion in 1999. During the same period, utilities project a 13-percent reduction in direct utility expenditures on energy-efficiency programs.

The electric power industry has entered a period of rapid change. Predicting DSM effects 5 years into the future can be difficult. The extent to which changes have been fully or accurately anticipated by utilities in their 1999 DSM projections can be uncertain.

Summary of DSM Trends 1994 to 1999

The major trends in DSM data reported on Form EIA-861 for 1994 are:

• In 1994, utilities experienced moderate reductions in DSM activity.
• For 1995, utilities projected substantial reductions in the growth of annual energy savings and lower potential peak load reductions. These reductions are partially explained by data reporting issues.
• Although energy savings and peak load reductions from energy-efficiency programs were impacted, other types of DSM programs were affected to a comparable or greater extent by reductions in DSM activity in 1994 and reductions projected for 1995.
• DSM spending is projected to decline moderately, suggesting that utilities intend to retain a DSM capability.
• Utilities are projecting growth in annual energy savings and annual potential peak load reductions for 1999, although that growth will be at a more modest rate than over the last 5 years.

The restructuring of the electric power industry may change electric utility DSM. Utilities that anticipate little growth in the use of DSM resources attribute this to increasing competition in the electric power industry.²⁸ The fundamental characteristics of a restructured industry are:

• Generation revenues will be based on market prices for generation services, instead of through cost-of-service regulation.
• Customers increasingly will have access to flexible prices that reflect fluctuations in spot-market prices for generation.

These are characteristics of most models of a restructured electric power industry. The economic forces released by such changes could have significant impacts on 3 types of electric utility DSM: energy efficiency, load building, and real-time pricing and other flexible load-shape programs.

Energy Efficiency in a Competitive Electric Power Market

Energy-efficiency programs were designed in an IRP framework in which regulators required utilities to consider the benefits and costs of substituting such programs for the acquisition of new generation resources. In a deregulated competitive market, generating capacity will likely be added or retired based upon its marketability. Resource planning will become a competitive business function. This change is leading some commentators to question the continuing role of energy-efficiency programs. The resulting debate focuses on three issues:

• The ability of markets to capture cost-effective energy-efficiency opportunities.
• The costs of energy-efficiency programs in a competitive electric power market and the benefits of the programs to consumers and society.
• The rate impacts of energy-efficiency programs.

The Ability of Markets to Capture Cost-Effective Energy-Efficiency Opportunities

Technology-based evaluations suggest that many cost-effective energy-efficiency improvements are not rapidly adopted in the marketplace. For example, in 1990, the Electric Power Research Institute estimated that 20 percent of total U.S. electricity consumption could be saved with energy-efficiency measures costing less than 3.5 cents per kWh saved.²⁹ Others suggest much higher potential savings.³⁰ Given the measures considered in such studies, it appears that consumers acting on their own do not adopt many commercially available and cost-effective efficiency measures. This finding is consistent with a second group of studies of actual consumer purchasing practices indicating that residential consumers act as if they severely discount the value of future energy savings when making energy-efficiency investments.³¹ A third group of studies examining commercial and industrial customer behavior found that such customers seldom undertake major energy-efficiency investments with more than a 2-year simple payback.³² For many measures, a 2-year payback implies that energy-efficiency investments have to produce an after tax return on investment of 30 percent or higher.

Economists, technologists, and social science researchers are engaged in a debate concerning the source of this non-cost-effective consumer behavior.³³ Such behavior may be the result of barriers to the adoption of efficiency measures which represent real costs of efficiency improvements or failures of markets to operate efficiently. Energy-efficiency programs that remedy or offset genuine market failures could increase overall economic efficiency in comparison to competitive market outcomes. Three primary perspectives are being advanced in this debate.

First, some economists argue that there must be “hidden costs” associated with the adoption of efficiency measures.³⁴ In some cases, this argument is offered as a simple tautology: markets are presumed to operate efficiently; therefore, the failure of markets to adopt efficiency measures must be attributable to some cost not considered in conventional benefit/cost analysis. At this level, the hidden cost position adds little to the debate since the answer is assumed in the premise of the argument. There may be hidden costs such as minor inconveniences or differences in perform ance associated with the adoption of some efficiency measures. There may also be hidden benefits such as small improvements in performance or conveniences that are not considered in conventional benefit /cost studies. The hidden cost hypothesis is at best incom plete in that there are cases, such as efficient lighting ballasts, refrigerators, personal computers, and televisions, in which there is little or no possibility of hidden costs, yet cost-effective efficiency measures are not widely adopted.³⁵

Second, some commentators relate the efficiency gap to uncertainty about future energy prices or other market conditions.³⁶ In the face of uncertainty, an efficient consumer may put off making deferrable investments. Most energy-efficiency improvements are made as part of a decision to invest in new equipment or a new building. If decisions to adopt efficiency measures are not made at the time a building is designed or equipment purchased, the opportunity is effectively lost. For example, it is not practical to change the orientation of a building to reduce summer heat gains after it is built. Nor can the consumer obtain a more efficient refrigerator without purchasing a new one. The opportunity to make energy-efficiency improvements exists when a building or appliance is acquired. Such efficiency investments are not deferrable. In these circumstances, efficient consumers must make decisions at the time of purchase based on the expected outcome of their choices regardless of the extent of uncertainty about market conditions.

A third view advanced by other economists, supported by social science researchers, and implicit in the positions of many technologists is that part of the efficiency gap may result from market failures related to the nature of the information involved in evaluating energy-efficiency investments. Economists identify two types of market failures in consumer evaluations of energy-efficiency investments:

• Information on the energy use of many products and services is not readily available or evident to many consumers when making energy efficient investments.³⁷ This also contributes to the difficulty of communicating the benefits of energy-efficient investments.³⁸ Energy use can be a low priority for some commercial and industrial establishments where energy costs represent approximately 3 percent of their total costs.
• Consumers may lack the expertise necessary to gather, process, and apply information to make optimal energy-efficient choices.³⁹ Additionally, recent experiments in economics show that consumers tend to repeat prior decisions when faced with unfamiliar choices and to avoid cost minimizing choices that have higher first costs.⁴⁰ In the market, such behavior impedes the commercialization of new energy-efficient technologies.

Such market failures may disproportionately impact the acceptance of new technology, limiting the ability of suppliers to achieve economies of scale, reduce product prices, and make energy-efficient technologies more competitive and widely available. They also may contribute to a more general market failure—new technology frequently has spillover benefits, making it difficult for the original developer to capture the full value of development and commercialization.

To the extent that market failures retard the commer cialization of energy-efficient technologies, utility or government energy-efficiency programs can play an essential role in pulling new technologies into the market place.

The Benefits and Costs of Energy Efficiency in a Competitive Generation Market

Short-term prices are significantly below the avoided costs of generating capacity assumed in DSM benefit/cost analysis just a few years ago. This could result in the discontinuance of DSM programs that are no longer cost-effective. This may account for part of the reduction in DSM activity. Increased competition is expected to improve the productivity and production efficiency of existing generation, delay retirement of some existing capacity, and lead to pricing that could flatten the difference between peak and off-peak loads. These effects can perpetuate surpluses and temporarily hold down market prices for generation. Given short-term capacity surpluses, the benefits of efficiency and other new resources could be more limited than assumed earlier in the decade. Even in the short-term, however, prices will not be uniformly low for all hours and locations. In the long run, restructuring might produce higher prices for generation services. In a restructured industry, the marketability of power can govern the addition of new capacity. New generating capacity will not be added until prices have risen sufficiently above the cost of new facilities to ensure generation suppliers a reasonable return at variable and uncertain market prices.⁴¹ Additionally, utilities are discovering that targeting DSM to optimize or defer transmission and distribution capacity investments canproduce substantial benefits, not previously considered in DSM benefit/cost analysis.⁴²

One of the benefits of energy efficiency is that reduced consumption avoids environmental impacts associated with electric generation. In the last few years, a series of studies were completed that attempt to place damage cost valuations on emissions from electric power plants. Some of these studies have tried to quantify externality values. However, they do not include estimates of environmental damage associated with global climate change.⁴³ If concerns about climate change and other environmental impacts of electric generation grow, this could lead to renewed interest in energy efficiency, one of the few low-cost approaches to reducing carbon dioxide emissions.

Overall, utility energy-efficiency programs are successful. In 1994, the mean utility cost for efficiency programs fell to 2.9 cents per kWh saved. A number of utilities were able to achieve substantial energy savings at costs below 2 cents per kWh saved⁴⁴ (Figure FE4). Some analysts question the costs of energy-efficiency rebate programs and the apparent disparity between high and low cost programs.⁴⁵ They point out that utility accounting, measurement, and reporting practices vary and that in some cases, customer costs are not included in reported program costs. More recent and detailed reviews of utility program evaluations adjust for inconsistent practices in response to these concerns.

In a detailed analysis of verified savings achieved, 20 utility commercial lighting programs were reviewed. All 20 programs were found to be cost-effective when compared to program-specific avoided costs.⁴⁶ A more comprehensive review of evaluations for 40 large commercial programs that accounted for one-third of 1992 utility DSM spending was recently completed for the Department of Energy. Most of these programs, which account for 88 percent of utility and consumer spending on programs included in the study, were cost-effective. For all the programs analyzed, the savings weighted average ratio of total resource benefits to total resource costs was 3.2 to 1.⁴⁷ Eight programs had total resource costs at or below 2« cents per kWh. There are examples of programs, particularly smaller programs, that are not cost-effective. Overall, however, utilities demonstrate a capability to undertake highly cost-effective large energy-efficiency programs.

These results are significant because: (1) they reflect only the direct effects of utility conservation programs and ignore secondary impacts on the availability of newtechnology and market behavior; and (2) large-scale utility energy-efficiency programs are relatively new and their performance continues to improve.

Some recent utility programs focused on creating a lasting transformation in regional or national energy markets by bringing new technologies into the market place or changing standard practices. For example, a national consortium of 24 utilities sponsored the “Golden Carrot” Super-Efficient Refrigerator Program that awarded $30 million in manufacturer incentives to the manufacturer introducing and marketing the most efficient new refrigerators. Whirlpool Corporation's winning bid resulted in the introduction, in 1994, of CFC-free refrigerators that used 29.4 percent less energy than the 1993 Federal Appliance Efficiency Standard. The objective of such programs is to introduce new technologies and practices that subsequently could retain and expand market share without the need for continuing financial incentives. Such programs can reduce utility costs per kWh saved. They also begin to address the equity questions that are raised because participants may benefit more than non-participants from rebate programs. By changing the products available in the market place, such programs produce benefits both for direct participants and other customers who may later take advantage of the availability of improved technology.

Rate Impacts of Energy-Efficiency Programs

Utilities and regulators cite the rate impacts of energy-efficiency programs as a reason for reducing savings targets or avoiding reliance on large rebates. These rate impacts reflect the net impact of revenue losses associated with reduced utility sales, direct and indirect program costs to the utility, and the supply cost savings associated with reduced demand and energy consumption. For many utilities, the largest contributing factor is the revenue loss that occurs under conventional rate design practices. In a regulated environment, conventional rate design practices lead to energy and demand charges substantially in excess of utilities’ short-run marginal costs. The difference between a utility's energy charges and marginal costs reflects a contribution to the recovery of the utility’s fixed costs. When conservation programs reduce sales, conventional rate designs result in a net revenue loss to the utility. Utilities must adjust rates to recover the net lost revenues by spreading the recovery of fixed costs over a reduced sales volume.

As utilities move into a competitive environment, their energy charges will inevitably fall towards marginal costs. This already is evident in the rates that many utilities are offering their largest customers and will be essential to the utilities' ability to compete for incremental sales. As the industry continues to move towards restructuring, rates are likely to be unbundled with the price of competitive services separated from other components of the customers' bills and pushed towards their marginal costs. Any remaining fixed costs could be recovered through a fixed access, customer, or demand charge. A series of studies documented that changing rate design practices could dramatically reduce negative rate impacts, in some cases even producing a reduction in average rates over the life of the efficiency measures.⁴⁸ These studies suggest that large rate impacts from efficiency programs are a short-term consideration and could be substantially mitigated through optional rate designs and cost allocation prac tices. As competition increases, more efficient rate design practices will greatly reduce the rate impacts that have been associated with efficiency programs.

Consumer and Utility Interests in Energy Efficiency Programs

In evaluating whether the projected reductions in 1995 energy-efficiency programs represent a transitional or a longer-term phenomenon, it is useful to consider how restructuring may affect consumer and utility interests in energy-efficiency programs.

In a competitive market, the effects of significant efficiency programs will be to reduce demand and to lower the market price of generation services. These benefits would accrue to all electricity consumers in relevant market areas. Given that generation revenues in a fully competitive market will be recovered at market prices, instead of on a cost-of-service basis, the interests of utilities in operating such programs will change. In the regulated environment, utilities have an obligation to serve, including the obligation to build or acquire generation resources. Energy-efficiency programs offer an attractive way to avoid the need for investment in new capacity. In a fully competitive environment, the obligation to serve could become an obligation to provide access to the transmission and distribution grid. In a competitive market for genera tion services, it is in the vertically integrated utility’s interest, as competitive generation supplier, to sell more generation services at a higher market price.⁴⁹ Efficiency programs will bring this interest into conflict with the utility’s traditional service objective of helping customers reduce their total energy bills. Energy-efficiency programs typically reduce energy consumption and may place downward pressure on the price of generation. This downward pressure on generation prices could reduce utility profits. This shift in the interests of local utilities might help to explain reductions in savings from DSM programs.

Policymakers who wish to retain a broader set of efficiency programs face two challenges. First, a means of financing such programs that does not penalize the local utility in comparison to other generation suppliers has to be identified. Several commentators suggest a system-benefits charge to be paid by all consumers seeking to access the transmission and distribution grid.⁵⁰ Such charges might take the form of fixed access fees, usage-based charges, or an “uplift” equal to a percentage of electricity costs. Some States have adopted analogous universal service charges to address public policy objectives in competitive telecom munications markets. Such charges would be non-bypassable and competitively neutral, paid by all consumers with access to the grid regardless of their choice of generation supplier.

Second, policymakers have to address reluctance on the part of local utilities to implement programs that reduce demand and potentially reduce market prices for their generation. Several options are being dis cussed including divestiture of local distribution utilities’ interests in competitive generation, establish ment of conservation trusts, creation of separate conservation utilities, and/or an expanded competitive bidding process that allows product manufacturers, vendors, and others to compete for incentives to support technology commercialization and market transformation. These options avoid the situation in which only the incumbent generation supplier could offer efficiency programs paid for by all consumers.

Customer Service and Load Building Programs

Electric utilities' competitive interest in expanding sales does not mean that all energy efficiency and DSM opportunities will be ignored. When asked about the impacts of growing competition on DSM activities, several utilities indicated that they will increasingly focus on offering energy services to customers. Packaging generation with efficient electric devices, in some cases, may help utilities attract and retain customers. Some utilities are effective in using energy-efficiency programs as a way to attract or retain industrial customers.⁵¹ Many utilities are utilizing DSM to compete with natural gas or to market electro-technologies. In 1994, the annual energy effects of load building programs were projected to double from 3,059 giga watthours (GWh) in 1995 to 6,251 GWh in 1999.⁵²

Real-Time Pricing and Other Flexible Load- Shape Programs

Under current regulation, most customers are served under rates based on average embedded costs.⁵³ Customers receive a single, high level of service reliability. And, for most customers, the same rate applies throughout the year or large periods during the year, regardless of the actual cost to the utility of generating electricity in any given hour or of distributing electricity to any particular portion of the transmission and distribution grid. As a result, consumers have little opportunity to control their electricity costs by matching their preferences regarding the cost, timing, and reliability of service to the price and character of the services purchased. New communication technologies are making it practical to provide consumers variable price signals and a range of other demand-side services.

Time-of-use pricing, real-time pricing, and other flexible load-shape programs can take advantage of the substantial variation in generation prices by time and location that is expected in a competitive market. Utilities have started offering real-time pricing to their largest customers and residential pilot programs that involve automated energy management, two-way communication systems, and time-of-use prices. Spot-market prices will fluctuate based on load levels, the availability of major generating units, and transmission constraints. In some cases, generation prices could fluctuate from less than 2 cents to as much as 15 cents per kWh on a significant number of days per year. During capacity shortages, prices could increase to 50 cents per kWh or higher, reflecting the cost of building new generation to serve peak loads and the price signals that might be required to match demand to available supply.

In a restructured industry where consumers choose their generation suppliers, some utilities, generation suppliers, and intermediary supply coordinators could be expected to package energy and information services. The packaging of energy and telecommuni cations services makes it possible to expand the DSM and other services available to consumers, including:

• Time-of-Use and Real-Time Pricing:
• 
  Communication linkages can be used to send out variable price signals or schedule time periods when low, moderate, or high price levels will be in effect. The technology used to receive and respond to such price signals will be automated energy management systems that implement predetermined consumer preferences regarding tradeoffs between cost and comfort or convenience.
• Customer-Influenced Load Management:
• 
  Two-way communications permit utilities to determine the effects of load management at the premise and end-use levels. Utilities could offer load control services that include a customer override option, with billing dependent upon whether the option was exercised.
• Energy Information Services:
• 
  Communication and information management systems can be used to provide customers with an array of energy information services, including:
  1. Continuously updated breakdowns of monthly energy use by major appliance or end use and variable pricing category.
  2. Comparisons of energy use by appliance or end use in the current and prior periods.
  3. Projections of the monthly electricity bill based on partial month data.
  4. Comparisons of energy use to typical neighborhood profiles.
  5. DSM recommendations, including estimates of energy cost impacts of potential efficiency imp
  rovements.

Benefits from automated meter reading, remote connect/disconnect services, electronic billing, automated bill payment, theft or tampering detection, distribution automation, and non-energy services also may contribute to the cost-effectiveness of energy-related two-way communication systems.

In some cases, energy information services may be provided as part of a broad band communication net work that also makes available cable TV, telephone, internet, security system, video-on-demand, medical alert, and other telecommunications services. But, a choice of communication technologies, including use of existing telephone lines, wireless, and hybrid fiber optic/coaxial cable systems, will permit energy information services to develop at a pace that is independent of the construction of broad band tele communication networks.

There is significant interest within the industry in packaging flexible pricing, load management, energy information, and other services. The extent to which such approaches become cost-effective for small consumers will depend upon the degree of variation in spot prices, the number of hours per year in which spot prices are high, the willingness of customers to pay for energy information and other services, and the ability of manufacturers to continue to lower the cost of communication and energy management systems.

In conclusion, it appears that in 1994 DSM programs were impacted by increasing competition in the electric power industry, while decreases in potential peak load reductions and in the growth of annual energy savings were projected for most DSM programs for 1995. A part of the reported reduction in the growth in the annual energy savings was caused by under-reporting of energy savings from past installations of energy-efficiency measures that continue to provide savings, but were installed under programs that are no longer in existence. EIA is addressing this problem in its 1995 survey. After correcting for major instances of under-reporting, the growth in annual energy savings projected for 1995 remained below that achieved in prior years.

Reduced growth in energy savings and peak load reductions may be a reflection of a number of factors: lower avoided costs; concerns regarding competition and rate impacts; and regulatory uncertainty during a transition toward a competitive environment. Another factor may be the conflict between integrated utilities’ financial interests as suppliers of competitively priced generation and the potential of DSM programs to reduce load and market prices for generation. Electric utilities’ long-term projections show a resumption of growth in annual energy savings and peak load reductions by 1999. Projected DSM spending levels suggest that utilities plan to retain a substantial portion of their capability to implement DSM programs.

As the industry considers major restructuring, the scope and character of electric utility DSM are likely to change. Market interventions designed to accelerate the commercialization of new energy-efficient technologies or practices may continue to be justified as a means of reducing market failures. However, the trends evident in the Form EIA-861 data raise questions as to whether new program and institutional options should be considered to address this objective. At the same time, restructuring could greatly expand other demand-side activities including the use of real time pricing, time-of-use pricing, automated energy management, energy information services, and other services designed to expand the ability of customers to respond to changing price signals. Providing service packages that include generation, management of the price risks associated with competitive generation markets, and demand-side services could help attract and retain customers in acompetitive market. The future of DSM will be determined by the choices that consumers, utilities, other service providers, regulators, and legislators make during the transition to competitive electric power markets.

END NOTES

² The 1978 Public Utility Regulatory Policies Act, Public Law 95-617, 16 U.S.C. 2601 and 2621(d)(6).

³ M. D. Levine, et al., Mitigation Options for Human Settlements, International Panel on Climate Change Working Group II, Chapter III-D (August 23, 1994); A. Rosenfeld, et al., "Conserved Energy Supply Curves for U.S. Buildings," Contemporary Policy Issues (January 1993), p. 45; Alliance to Save Energy, American Council for Energy Efficient Economy, Natural Resource Defense Council, and Union of Concerned Scientists, America’s Energy Choices: Investing in a Strong Economy and a Clean Environment, Technical Appendices (Cambridge, MA, 1991); National Academy of Sciences, Policy Implications of Greenhouse Warming: Report of the Mitigation Panel (Washington, DC: National Academy Press 1991); Office of Technology Assessment, U.S. Congress, Changing by Degrees: Steps to Reduce Greenhouse Gases, OTA-O-482 (Washington, DC, 1991); Barakat and Chamberlin, Inc., Efficient Electricity Use: Estimates of Maximum Energy Savings, EPRI, CU-6747 (1990); R. Carlsmith, et al, Energy Efficiency: How Far Can We Go? (Oak Ridge, TN: Oak Ridge National Laboratory, 1990); U.S. Dept. of Energy, Office of Conservation, Energy Conservation Multi-Year Plan 1990-1994 (August 1988); H. Geller, et al, Pacific Gas & Electric Residential Conservation Power Plant Study (February 1986); A. Meier, et al, Supplying Energy Through Greater Efficiency: The Potential for Conservation in California's Residential Sector (1983); Solar Energy Research Institute, A New Prosperity -- Building a Sustainable Energy Future (Andover, MA: Brick House Publishing, 1981); The National Research Council, Alternative Energy Demand Futures to 2010 (1979).

⁴ Marginal Cost is the cost of producing a small additional increment of power. Short-run marginal costs reflect the cost of delivering that increment of power from existing generating capacity.

⁵ The 1990 Clean Air Act Amendments, 42 U.S.C. §7651c, and the Energy Policy Act of 1992, 16 U.S.C. §2621(d)(8), contain specific provisions designed to encourage States to adopt ratemaking mechanisms that remove the disincentives to effective implementation of energy-efficiency programs. EPACT also requires State utility commissions to consider standards that will require utilities to employ integrated resource planning .

⁶ Energy Information Administration, "Evaluation and Verification of Demand-Side Management Programs," U.S. Electric Utility Demand-Side Management 1993, DOE/EIA-0589(93) (Washington, DC, July 1995).

⁷ Large utilities are those with sales to ultimate consumers or sales for resale greater than or equal to 120,000 megawatthours per year.

⁸ Electric utilities report estimates of savings and peak load reductions. These reports are subject to a quality assurance review performed by EIA. The reports for major utilities are compared to utility filings with the utilities’ State regulators. Utilities are contacted for clarifications when reporting issues are identified. Utilities were asked to indicate whether energy savings or peak load reductions are subject to verification. For prior years, estimated savings have subsequently been compared to program evaluation results. While estimated savings in some cases exceeded subsequently verified results, a large variance between estimated and verified savings was not found (U.S. Electric Utility Demand-Side Management, 1993, "Estimation and Verification of Demand-Side Management Programs" ). Utilities report actual peak load reductions for energy-efficiency programs and both potential peak load reductions and actual peak load reductions for direct load control, interruptible load, other load management, and other DSM programs. Potential peak reductions reflect the installed load reduction capability of the utility. Actual effects reflect the load reductions achieved from programs in place at the time the utility experiences its annual peak load. For purposes of this paper, the sum of actual peak load reductions from energy-efficiency programs and potential peak load reductions from direct load control, interruptible load, other load management, and other DSM programs will be referred to as potential peak load reductions. Incremental energy savings are reported on an annualized basis, as if savings had been achieved for a full calendar year regardless of the date during the year on which individual measures were installed.

⁹ From the turn of the century until about 1970, electric utilities were able to reduce generation costs by building larger generating units -- some as large as 1,300 megawatts. Thereafter, further increases in maximum unit size failed to provide economic advantages given technical, construction lead time, and reliability constraints. (R. Hirsh, Technology and Transformation in the American Electric Utility Industry (Cambridge, U.K.: Cambridge University Press 1989).) Today the optimum size for new generating capacity may be 150 megawatts or less. (C. Bayless, "Less is More: Why Gas Turbines Will Transform Electric Utilities," Public Utilities Fortnightly, (December 1, 1994) p.21.)

¹⁰ National Association of Regulatory Utility Commissioners, Utility Regulatory Policy in the United States and Canada: Compilation 1992-1993 (Washington, DC, 1993), p. 421.

¹¹ In Re TECO Power Services and Tampa Electric Company, 52 FERC ¶61,191 (1990); In Re Ocean State Power II, 59 FERC ¶61,360 at 62,323-4 (1992).

¹² Energy Policy Act of 1992, Pub. L. No. 102-486, Stat. 2776 (1992).

¹³ Energy Information Administration Form EIA-867, "Annual Nonutility Power Producer Report."

¹⁴ Energy Information Administration, Electric Power Annual 1992, DOE/EIA-0348(92) (Washington, DC, January 1994), p. 12.

¹⁵ Federal Energy Regulatory Commission (FERC) Form 1 "Annual Report of Major Electric Utilities, Licensees and Others" (1994) and McGraw-Hill, Power Markets Week. (1994).

¹⁶ Spot-market prices during peak and mid-peak periods are low because they reflect the current surpluses of generating capacity in many parts of the country. The generation costs built into utility rates are higher because they are based on the utility’s embedded or historical costs and reflect surplus capacity, high cost plants completed in the 1980's and early 1990's, and other fixed or already incurred costs.

¹⁷ Federal Energy Regulatory Commission, Promoting Wholesale Competition through Open Access Non-Discrimination Transmission Services by Public Utilities and Recovery of Stranded Costs by Public Utilities and Transmitting Utilities, Docket Nos. RM95-8-000 and RM94-7- 001, Notice of Proposed Rulemaking and Supplemental Notice of Proposed Rulemaking (March 29, 1995).

¹⁸ Federal Energy Regulatory Commission, Inquiry Concerning the Commission’s Pricing Policy for Transmission Services Provided by Public Utilities under the Federal Power Act, 69 FERC ¶61,086 (1994).

¹⁹ Federal Energy Regulatory Commission, Recovery of Stranded Costs by Public Utilities and Transmitting Utilities, Notice of Public Proposed Rulemaking, 59 Federal Register 35274 (July 11, 1994).

²⁰ Federal Energy Regulatory Commission, New Reporting Requirement Implementing Section 213(b) of the Federal Power Act and Supporting Expanded Regulatory Responsibilities under the Energy Policy Act of 1992, and Conforming and Other Changes to Form No. FERC-714, FERC Docket No. RM 93-10-000, Final Rule (September 30, 1993).

²¹ E. Kahn & S. Stoft, Organization of Bulk Power Markets (Berkeley, CA: Lawrence Berkeley Laboratory 1996).

²² C. Winston, "Economic Deregulation: Days of Reckoning for Microeconomists," Journal of Economic Literature, 31 (9) (September 1993) pp. 1,263-1,289.

²³ Legislative Energy Advisory Program, Quarterly Legislative Letter (December 1, 1995).

²⁴ For purposes of this article, actual peak load reductions from energy-efficiency programs are included in potential peak load reductions.

²⁵ This calculation of the cost of conserved energy is based upon 1994 reported incremental savings from efficiency programs, direct costs of efficiency programs, the allocation of indirect costs in proportion to direct costs by DSM program type, a conservative assumption of a 10-year average life, and discounting the value of future savings at a 5-percent real discount rate. Cost of conserved energy was calculated as follows:

²⁶ In the Matter of the Order Instituting Rulemaking on the Commission’s Proposed Policies Governing Restructuring California’s Electric Services Industry and Reforming Regulation, Docket No. R. 94-04-031 (April 20, 1994).

²⁷ M. Schweitzer and M. Pye, Key Factors Responsible for Changes in Electric-Utility DSM Usage (Oak Ridge, TN: Oak Ridge National Laboratory, Sept. 1995).

²⁸ M. Schweitzer and M. Pye, Key Factors Responsible for Changes in Electric-Utility DSM Usage (Oak Ridge, TN: Oak Ridge National Laboratory, Sept. 1995).

²⁹ A. Fickett, C. Gellings, & A. Lovins, "Efficient Use of Electricity," Scientific American (1990).

³⁰ See Endnote 3.

³¹ Berkovec, J. et al., Heating System and Appliance Choice, Working Paper, MIT Energy Laboratory, Report #MIT-EL 83-004WP (January 1983); Cole, H. and R. Fuller, Residential Energy Decision Making: An Overview with Emphasis on Individual Discount Rates and Responsiveness to Household Income and Prices, Hillman Assoc., U.S. Department of Energy (February, 1981); Corum, K. and D. O’Neal, Investment in Energy-Efficiency Houses: An Estimate of Discount Rates Implicit in New Home Construction Practices, Energy, Volume 7, No. 4 (1982), pp. 389-400; Dubin, J., Econometrics Theory and Estimation of the Demand for Consumer Durable Goods and Their Utilization: Appliance Choice and the Demand for Electricity, MIT Energy Laboratory Discussion Paper No. 23, MIT-EP 82-035WP (May 1982); Gately, D., Individual Discount Rates and the Purchase and Utilization of Energy-Using Durables: Comment, Bell Journal of Economics, Volume 11, No. 1 (Spring 1980), pp. 373-374; Goett, A. and Moss, W., Implicit Discount Rates in Residential Customer Choices, Investments in Conservation Measures, EM-5587, Research Project 2574-1, Electric Power Research Institute, Palo Alto, CA (February 1988); Hausman, J., Individual Discount Rates and the Purchase and Utilization of Energy-Using Durables, Bell Journal of Economics, Volume 10, No. 1 (Spring 1979), pp. 33-54; McRae, D., Rational Models for Consumer Energy Conservation, Burby and Marsden (eds.), Energy and Housing, Cambridge: Oelgeschleger, Gunn, and Hain, Publishers, Inc. (1980); Meier, A. and J. Whittier, Consumer Discount Rates Implied by Purchases of Energy-Efficient Refrigerators, Energy, Volume 8, No. 12 (1983), pp. 957-962; Ruderman, H., et al., The Behavior of the Market for Energy Efficiency in Residential Appliances Including Heating and Cooling Equipment, LBL-15304, Energy Analysis Program, Lawrence Berkeley National Laboratory (September 1984).

³² W. Fuller, "Industrial DSM D What Works and What Doesn’t," Proceedings on the 1992 ACEEE Summer Study on Energy Efficiency in Buildings, at 5.75 (August 1992); G. Hatsopoulos, et al., "Capital Investment to Save Energy," Harvard Business Review 111 (March-April 1978); A. Evans and S. Zussman, Barriers to the Adoption of Energy-Conserving Technologies in the Textile Industry, U.S. Department of Energy (1979); H. Landsberg, ed., Selected Studies on Energy: Background Papers for Energy: The Next Twenty Years, Ballinger Publishing Company, Cambridge, MA (1980); R. Stobaugh and D. Yergin, eds., Energy Future: Report of the Energy Project at the Harvard Business School, Ballantine Books, New York, NY (1979); Alliance to Save Energy, Industrial Investments in Energy Efficiency: Opportunities, Management Practices, and Tax Incentives: Summary, Washington, D.C.; H. Herzog, et al., Energy Management in Central Maine Power Company’s Industrial Sector with Specific Emphasis in the Pulp and Paper Industry, Final Report for the Central Maine Power Company, Massachusetts Institute of Technology, MIT-EL 92-001 (March 1992); Niagara Mohawk Power Corporation, Niagara Mohawk Demand-Side Management (DSM) Subscription Option Status Report (April 12, 1994).

³³ E. Hirst and J. Eto, Justification for Electric-Utility Energy-Efficiency Programs, Oak Ridge National Laboratory (August 1995).

³⁴ L. Ruff, "Least-Cost Planning and Demand-Side Management: Six Common Fallacies and One Simple Truth," Public Utilities Fortnightly (April 28, 1988) p. 19; R. Sutherland, "Market Barriers to Energy-Efficiency Investment," The Energy Journal Vol. 12, No. 3 (1991) p. 15.

³⁵ M. Levine, et al., Energy Efficiency, Market Failures, and Government Policy, LBL-35376, ORNL/CON-383 (Berkeley, CA: Lawrence Berkeley Laboratory, March 1994).

³⁶ K. Hassett and G. Metcalf, "Energy Conservation Investment, Do Consumers Discount the Future Correctly?," Energy Policy (June 1993) p. 710.

³⁷ A. Sanstad and R. Hawarth, "‘Normal’ Markets, Market Imperfections and Energy Efficiency," Energy Policy, Vol. 22, No. 10 (1994) pp. 812-818; W. Kempton and L. Layne, “Consumer’s Energy Analysis Environment,” Energy Policy, Vol. 22, No. 1 (1994) p. 857; P. Komor, et al., Energy Use, Information, and Behavior in Small Commercial Buildings, PU/CEES-240 (Princeton, NY: Princeton University, July 1989).

³⁸ S. DeCanio, "Barriers within Firms to Energy-Efficient Investments," Energy Policy, (September 1993) pp. 906-914.

³⁹ Sanstad and Hawarth, "‘Normal’ Markets, Market Imperfections and Energy Efficiency."

⁴⁰ H. Huntington, "Been Top Down So Long, It Looks Like Bottom Up to Me," Energy Policy October (1994) p. 833; R. Thaler, The Winner’s Curse: Paradoxes and Anomalies of Economic Life (New York, NY: the Free Press (MacMillan), 1992); R. Thaler, "Toward a Positive Theory of Consumer Choice," Journal of Economic Behavior in Organization (March 1990) p. 39.

⁴¹ A. Dixit and R. Pindyck, Investment Under Uncertainty (Princeton, NJ: Princeton University Press 1994); T. Kaslow and R. Pindyck, "Valuing Flexibility In Utility Planning,: The Electricity Journal, Vol 7, No. 2 (March 1994) p. 60; P. Centolella, "Prices, Options, and Investment in Competitive Power Markets," Proceedings of the Third National Energy Summit (1995).

⁴² R. Orans, C. Woo, and J. Swisher, Targeting DSM for Transmission and Distribution Benefits: A Case Study of PG&E's Delta District (Palo Alto, CA: Electric Power Research Institute 1992); R. Pratt, et al., "Potential for Feeder Equipment Upgrade Deferrals in a Distributed Utility," Proceedings of the 1994 ACEEE Summer Study on Energy Efficiency in Buildings (August 1994) p. 2.229; S. Sparks, et al., "Producing More with Less: Evaluating the Impact of a T&D Agricultural DSM Program," Proceedings of the 1994 ACEEE Summer Study on Energy Efficiency in Buildings (August 1994) p. 2.49; R. Weijo and L. Ecker, "Acquiring T&D Benefits from DSM: A Utility Case Study," Proceedings of the 1994 ACEEE Summer Study on Energy Efficiency in Buildings (August 1994) p. 2.269.

⁴³ Energy Information Administration, Electric Generation and Environmental Externalities: Case Studies, DOE/EIA-0598 (Washington, DC, September 1995); Oak Ridge National Laboratory and Resources for the Future, Estimating Externalities of Coal Fuel Cycles, U.S. Department of Energy and The Commission of European Communities (September 1994); RCG Hagler-Bailley, New York Externalities Cost Study (1995); see also: California Energy Commission, Electricity Report, Appendix F (November 1992).

⁴⁴ Energy Information Administration, Form EIA-861, "Annual Electric Utility Report." This calculation of the cost of conserved energy is based upon 1994 reported incremental savings from efficiency programs, direct costs of efficiency programs, the allocation of indirect costs in proportion to direct costs by DSM program type, a conservative assumption of a 10-year average measure life, and discounting the value of future savings at a 5-percent real discount rate.

⁴⁵ P. Joskow and D. Marron, "What Does a Negawatt Really Cost? Evidence from Utility Conservation Programs," The Energy Journal April (1992), pp. 41-74.

⁴⁶ J. Eto, et al., The Cost and Performance of Utility Commercial Lighting Programs, LBL-34967, UC-350 Lawrence Berkeley Laboratories (Berkeley, CA, May 1994).

⁴⁷ J. Eto et al., Where Did the Money Go? The Cost and Performance of the Largest Commercial Sector DSM Programs. (Berkeley, CA: Lawrence Berkeley National Laboratory, December ,1995).

⁴⁸ P. Centolella, Direct Testimony, In the Matter of Georgia Power Company’s Application for Approval of an Integrated Resource Plan (1995); Niagara Mohawk Power Corporation, 1993 Update to the 1991 IERP (June 1993); E. Hirst and S. Hadley, Price Impacts of Electric-Utility DSM Programs, ORNL/CON-402 (November 1994) pp. 16-17.

⁴⁹ Utilities that do not own generation may avoid this conflict in objectives.

⁵⁰ See for example: P. Centolella, Testimony, In the Matter of the Obligation of the Association of Business Advocating Tariff Equity for Approval of an Experimental Retail Wheeling Tariff for Consumers Power Company, Case No. U-10143R, Michigan Public Service Commission (1994); R. Cavanagh, Usage Based System Benefit Charges: The New Regulatory Imperative for Avoiding Stranded Benefits (February 1995).

⁵¹ J. Jordan and S. Nadel, Industrial Demand-Side Management Programs: What's Happened, What Works, What's Needed, U.S. Department of Energy and the Energy Foundation (March 1993); M. Kellogg, Staying Power Program Impact Evaluation, Bonneville Power Administration (August 1991); and Testimony of S. Larson, In the Matter of the Naragansett Electric Company, Rhode Island Public Utilities Commission, Docket No. 1939 (1992).

⁵² It is possible that reductions in forecasted savings from efficiency programs could reflect some utilities reporting under the category of energy efficiency the net load impacts of programs designed to both attract load and improve the efficiency of customers currently using electricity. EIA has added instructions to the 1995 Form EIA-861 that address this issue.

⁵³ "Embedded costs" are the sum of current operating expenses, depreciation and amortization expenses associated with historical investments, and a reasonable return on the undepreciated and unamortized capital account balances associated with historical investments.



File last modified: April 10, 1997

Contact:

Channele Donald

cdonald@eia.doe.gov

Phone: (202) 426-1270