Editor Note: Demand-Side Management (DSM) is an electric-utility program where all ratepayers subsidize the energy conservation investments of those ratepayers who participate in the program. Dan Simmons, director of state affairs for the American Energy Alliance and for the Institute for Energy Research in Washington, D.C., submitted the following testimony in opposition to Georgia Power Company’s request to implement a DSM program.
Summary from Testimony: Market vs. Non-Market Conservation
Advocates of demand side management frequently overlook the human dimension. People make conservation decisions every day. Families are constantly striving to maximize the benefits they receive from using energy with the cost of using that energy.
Some economic analyses try to argue that ratepayers systematically misestimate the proper discount rate and therefore DSM is required to fix this error. But those analyses fall victim to assuming they know what’s best for ratepayers. The people who know best are ratepayers themselves. Only the individual ratepayer can decide how he or she best allocates their scare resources.
DSM leads to inefficient outcomes. It increases rates for all ratepayers, but only benefits a few of the ratepayers. Because history shows that DSM programs do not typically lead to the demand reductions claimed, they do not lead to lower electricity rates overall.
DSM is not new. It has been tried and studied for years. DSM is subject to the well known problems of the rebound effect, free riders, moral hazard, to name a few. Georgia Power’s 2010 Integrated Resource Plan and Application for Certification of Demand Side Management Programs analysis does not explain how this program will overcome these well-known economic problems nore does it explain how it will overcome the problems with Georgia Power’s DSM program from the 1990s. Without an analysis, it is difficult to believe Georgia Power’s DSM programs will achieve the benefits Georgia Power projects.
TESTIMONY
Q. PLEASE STATE YOUR NAME, TITLE AND BUSINESS ADDRESS.
A. My name is Daniel R. Simmons. I am the Director of State Affairs for the American Energy Alliance and the Institute for Energy Research, non-profit organizations that educate the public on energy policy. Our offices are located at 1100 H Street, Suite 400, Washington, DC 20005.
Q. PLEASE SUMMARIZE YOUR EDUCATION AND PROFESSIONAL EXPERIENCE.
A. I received a Bachelor of Science in Economics from Utah State University in 1998 and a Juris Doctorate from George Mason University School of Law in 2003. In 1998, I moved to Washington, D.C. and worked for the Competitive Enterprise Institute as an environmental policy analyst. In 2000, I went to work for the Committee on Resources of the United States House of Representative as a professional staff member. From 2002 to 2005, I was the Emmett McCoy Research Fellow at the Mercatus Center and from 2005 to 2008, I was the Director of the Natural Resources Task Director at the American Legislative Exchange Council. From 2008 and presently, I am the Director of State Affairs at the American Energy Alliance and the Institute for Energy Research.
I have over a decade of experience analyzing the impact of environmental regulation and legislation at both the federal and state level. The majority of my research and writing has been on the impact of regulation and legislation on the consumer.
Q. WHAT IS THE PURPOSE OF YOUR TESTIMONY?
A. The purpose of my testimony is to provide the results of my review and analysis of the Demand Side Management (DSM) portions of Georgia Power Company’s 2010 Integrated Resource Plan, and to make recommendations pertaining to DSM.
Q. WHAT ARE YOUR RECOMMENDATIONS?
A. I recommend that the Commission not approve the certification of Georgia Power’s Application for Certification of Demand Side Management Programs. The purpose of Georgia Power’s Integrated Resource Plan is to “meet customer needs.”[1] It is not at all clear that Demand Side Management (DSM) programs meet customers’ needs.
Georgia Power’s DSM Programs will increase electricity rates. And economic theory and past history have shown that the DSM programs will likely result in smaller demand reductions than claimed. DSM programs are not new. Many of these programs have been tried for nearly 30 years including a DSM program in Georgia from 1992–1994. That program was a failure.
Georgia Power’s DSM application does not explain why this plan, contrary to theory and actual history, will result in cost-effective demand decreases. Georgia Power’s DSM application provides no justification to impose the program on Georgia ratepayers and fails to support the claim that “these demand side resources have been selected because they represent the most cost-effective plan for Georgia Power’s customers.”[2]
Furthermore, the DSM program is fundamentally unfair. The program punishes all ratepayers through higher rates. Meanwhile, it favors a few ratepayers that receive preferential treatment through the program by receiving subsidized energy efficient products such as light bulbs, refrigerators, consultation services, just to name a few.
Because Georgia Power’s application does not contain an analysis of relevant economic theory, it is difficult to agree with their projections of the impact of the DSM programs. The energy reduction estimates are almost certainly overestimates of the true impacts of the DSM programs.
Q. PLEASE DESCRIBE YOUR ANALYSIS THAT FORMS THE BASIS FOR YOUR RECOMMENATIONS.
A. First, the underlying premise upon which demand side management is based is flawed. Demand side management presumes that a utility understands their customers’ wants and needs better than the customers themselves. The assumption is that customers are not properly availing themselves of energy efficiency investments. In order to correct this alleged “market failure,” the utility collects a DSM rider every month and then provides products and services to allow its ratepayers to “properly” value energy efficiency improvements.
There are a number of problems with the underlying premise of DSM and the application of DSM programs. First, a utility cannot know or understand costs and benefits better than the customers themselves. As James Buchanan and other economists have explained, cost and benefits are subjective to the individual and cannot be divined by outside observers.[3] Each person knows what matters most to himself or herself.
Even though costs and benefits are subjective, some people argue that consumers incorrectly value the costs and benefits of energy efficiency, especially the appropriate discount rate, and therefore they justify the existence of DSM programs.
Two of the most important considerations in evaluating the present monetary value of improvements in energy efficiency are future interest rates and the price of electricity. These can be volatile, and consumers quite rationally may not place much weight on expected savings from energy efficiency occurring several years in the future.
In other words, consumers may rationally have much higher discount rates than Georgia Power and its consultants assume they should have. Consider this research presented by economist Ronald J. Sutherland in the context of an automobile fuel economy rulemaking by the National Highway Transportation Safety Administration:[4]
Corporations frequently require high hurdle rates in excess of 12 percent to undertake capital investments. Dixit and Pindyck present a compelling analysis of observed high discount rates for irreversible investments.[5] The technical literature indicates that irreversible investments may require hurdle rates two to four times the average discount rate in order to trigger an investment.[6] However, fuel economy standards have the unattractive investment properties of being irreversible, whereas common stocks are highly liquid. Metcalf and Rosenthal[7] and Hassett and Metcalf[8] explain how this irreversibility property warrants discount rates of at least two or three times higher than may be expected. Allowing for the irreversibility property of such investments, a required rate of return of at least 20% appears reasonable for high-income households.
Energy saving investments are typically irreversible investments and therefore require an even higher premium. The proposed fuel economy standards for light trucks are irreversible investments. The investment in fuel economy is a sunk cost at time of purchase. The investment cannot be reversed, should the consumer decide that the investment is unwarranted. Hassett and Metcalf apply the irreversible investment model to investments in energy conservation and conclude that an appropriate hurdle rate would be about four times greater than the standard discount rate. Metcalf and Rosenthal reach a similar conclusion in applying the model to commercial lighting and to energy efficient refrigerators. If the government imposed discount rate of 7 percent is considered standard, an appropriate discount rate for the fuel economy benefits would be at least 14 percent, but probably closer to 21 percent or event 28 percent.
The application of higher hurdle rates indicates that the benefits from fuel economy standards should be revised downward. The NHSTA study calculates consumer benefits as the present value of future energy saving using a 7 percent discount rate. However, the evidence on discount rates, as well as revealed consumer preferences, indicates that an appropriate discount rate is at least 2 or 3 times higher that [sic] the government imposed rate.[9]
Consumers may not act as Georgia Power assumes they should act, but that is no proof that consumers act against their rational economic self-interest or that there is a “market failure.” Consumers may be maximizing other dimensions that Georgia Power is not considering. This is especially true in areas such as housing and major appliances where there are substantial sunk costs coupled with uncertainty over future conservation savings.[10]
This presents a serious challenge for DSM programs. DSM charges all ratepayers and then rewards a few. It serves as a hidden tax that transfers benefits from nonpartisans to participants.[11] This is not economically efficient.
Second, history shows that DSM programs in the past have not lived up to their billing. Many of these programs are three decades old, but Georgia Power’s Application for Certification of Demand Side Management Programs does not contain an appreciation of the past history of these programs. This is a serious flaw because the types of programs that Georgia Power seeks to implement have been implemented in the past and did not achieve the benefits claimed. Not only have these programs been tried elsewhere, in the early 1990s Georgia Power tried a DSM program, but it was a failure.
Some researchers have examined the past performance of DSM programs. One important Rand Corporation study found that DSM can indeed lower electricity sales, but the cost is far higher than utilities estimate.[12] In fact, the study found that the cost of the DSM programs was roughly two to three times as expensive, on average, as the energy the programs were attempting to conserve. Here is the conclusion to that study:
The within-utility estimates reported in Section 5 suggest that energy efficiency expenditures do lower electricity sales. The effect of these expenditures is small relative to what utilities themselves report in the same data, however. For the overall sample (324 utilities), we estimate DSM expenditures lowered mean electricity sales by between 0.3 and 0.4 percent at an average cost of $0.14-0.22/kWh. We estimate DSM expenditures had a larger effect on electricity sales for the sample of 119 utilities reporting positive DSM expenditures in every year. For these utilities, we estimate DSM expenditures lowered electricity sales by between 0.6 and 1.2 percent at an average cost of $0.06-0.12/kWh. By comparison, depending on the sample, utilities themselves estimated DSM expenditures lowered retail electricity sales by between 1.8 and 2.3 percent at an average cost of $0.02-0.03/kWh. We suspect utility estimates of DSM program effects are higher than our estimates because utilities generally do not fully control for selection bias.
Importantly, our estimates do not imply that investments in electricity efficiency at the firm or household level are ineffective; rather, they suggest that DSM expenditures do not perfectly target consumers on the margin of making energy efficiency investments. The firms and households most likely to make investments in electricity efficiency are also the most likely to take advantage of DSM programs. While the evidence presented in this study indicates that DSM expenditures in aggregate have done less to improve the electricity efficiency of the U.S. economy than what utilities themselves estimate, it may be that, from a societal standpoint, these DSM expenditures are nonetheless worthwhile. Moreover, some types of DSM programs may be more effective than others, a possibility our data, unfortunately, do not permit us to assess.
The lion’s share of utility-sponsored energy efficiency program expenditures pays for financial incentives like rebates, subsidies, and loan guarantees that aim to lower the cost of adopting energy efficient technologies. Cost, however, is but one of the determinants of technology adoption. Indeed, the well-established fact that many firms and individuals routinely forgo financially profitable investments in energy efficiency suggests that factors other than direct cost may be driving the investment decision. (24) Information and agency problems may be among them.
Much of the theoretical work on technology adoption emphasizes the process by which information on new technologies diffuses through the economy. Credible and useful information on new technologies does not spread instantaneously and so many potential adopters are uncertain about both the costs and benefits of adopting a given technology. There is also growing evidence that a variety of agency problems lead to suboptimal levels of energy efficiency investment (DeCanio, 1993, 1998). For example, tenants may not directly pay the electricity bill or the individual decision maker within a firm may have weak incentives to make energy efficiency investments despite the high expected value of these investments. Financial incentives will not necessarily help overcome these barriers to energy efficiency investment. Programs that collect and disseminate information on the costs and benefits of particular energy efficiency investments, demonstrate to particular firms how they can benefit from energy efficient investments, and resolve agency problems, on the other hand, may be more successful in inducing investments in energy efficiency. Future empirical research should address this question.[13]
In short, DSM can reduce energy use, but at what cost? According to this study, the cost of the DSM programs was two or three times as expensive as the electricity it was designed to conserve.
Not only do DSM projects have a poor record in the aggregate, but the DSM program Georgia Power tried in the early 1990s was also a failure. The experience of that program should be of particular relevance to the current application for DSM programs, but Georgia Power does not discuss any lessons learned from its early programs. This is a serious omission.
Third, economic theory suggests that demand side management programs are flawed in practice. As noted above, utilities overestimate the value of DSM programs. One reason is because utilities have no incentive for their DSM programs to actually reduce demand or be cost-effective. Indeed, the best situation for utilities is for demand to actually increase and for the utilities to be able to charge ratepayers not only for the increased electricity demanded, but also for the cost of the DSM programs.
Viewed most cynically, utilities may promote DSM programs because they allow the utility to capture some of the revenue that would be lost if consumers made energy efficiency improvements in the absence of a DSM program. As explained below, with DSM programs, utilities get paid for getting consumers to make energy efficiency improvements that they would have made in the absence of the program. With a DSM program in place, instead of merely losing lost revenue from lost sales, the utility makes money on the efficiency improvements whether or not they were conducted because of the DSM program.[14]
As explained below, basic economic theory explains why DSM programs do not result in the energy efficiency gains or the economic efficiency gains frequently claimed. The economic efficiency of DSM does not necessarily matter to utilities. What matters are the regulatory constraints and incentives the utility faces.[15] As a result, utilities do not necessarily properly analyze the basic economic problems with DSM. The following is some of the economic problems of DSM.
The stated goal of DSM programs is to “reduce future energy consumption while lowering future utilities bills.”[16] The problem with this goal is that when technology increases efficiency, the end result can be that more of the resource is used, not less. This is a well-known and long understood outcome known as the rebound effect or the Jevons Paradox.
In 1865, William Stanley Jevons wrote The Coal Question about England’s reliance on coal. Jevons found that when James Watt discovered a new, more efficient steam engine, the efficiency gains actually led to more coal use, not less. When steam engines became more efficient, they were used for more tasks.
People generally assume that greater energy efficiency should result in lower energy use. But the Jevons Paradox explains that is not necessarily the case. This rebound effect is a restatement of the law of demand—when prices fall, demand rises.[17]
This rebound effect is just as applicable to DSM programs today as to using coal to run coal-fired steam engines. When consumers buy more energy efficient appliances, for example, they may use the appliances more, improving the customer’s economic welfare, but not reducing overall energy use. This is a flaw in DSM programs that seek to reduce overall electricity demand.[18]
Georgia Power’s DSM Application does not demonstrate that it has evaluated the rebound effect. Furthermore, a stated goal of some of the DSM programs is to “reduce future energy consumption while lowering future utilities bills.”[19] This further complicates achieving demand reductions. When prices are lower, people will likely use more electricity.
Another well-known challenge to the effectiveness of DSM programs to actually achieve economically-efficient demand reduction is free riding, otherwise known as adverse selection. Simply-stated, DSM programs frequently fail to achieve their goals of increasing energy efficiency because the people that opt-in for the programs would have done so even in the absence of the DSM programs.
As one analyst wrote about a refrigerator exchange program, “Those will participate who have to replace their refrigerators anyway and those who expect to have to replace their refrigerators in the near future.”[20] Worse, free riders increase the administrative and subsidy cost of the DSM program without increasing energy efficiency.[21]
The problem with free riders can be pervasive. Surveying the literature on the impact of free riders on DSM programs, economist Matthew Hoffman wrote:
Virginia Kreitler of Synegetic Resources Company has conducted the most extensive survey of free-rider estimates in the U.S. She found that residential appliance programs contained 63.3 percent free riders on average, and commercial programs contained 37.7 percent free riders. More recent surveys have found free-riders rates ranging from 6 to 95 percent for various DSM programs.[22]
With Georgia Power’s DSM application, there is the potential for free riders with many of the programs. The Residential Lighting and Appliance program is ripe to be used by people who would buy more efficient appliances or CFL in the absence of the program. The Residential High Efficiency New Homes program, the Residential Existing Homes program, the Residential Refrigerator Recycling program, the Residential Water Heating program, the Commercial Audit program, the Commercial Custom Incentive program, the Commercial Prescriptive Incentive, and the Industrial Audit all are more likely to be utilized by people and companies who would utilize these types of services and energy efficiency measures in the absence of the DSM programs.
Georgia Power’s DSM Application does not demonstrate that it has evaluated the free rider problem. As a result, it is likely that the Application overestimates the amount of conservation that will actually occur as a result of the DSM programs.
Another challenge for DSM programs to achieve real demand reduction is the problem of moral hazard. Moral hazard is a concept related to free riders.[23] With moral hazard, consumers intentionally make inefficient decisions because DSM programs will pay for energy efficient upgrades. As Franz Wirl explains, “the optimal response for a consumer facing the incentive . . . is to cut back on all his own conservation efforts and to let the utility pay for the entire upgrade.”[24]
Moral hazard, like free riding, makes it difficult to quantify the true impact of the DSM programs. Moral hazard makes it difficult to ascertain the true amount of conservation triggered by DSM programs. On paper it might appear that conservation has occurred when no conservation has occurred at all. That is because consumers, knowing that the utility will aid with the purchase of more energy efficient equipment, holds off for the aid instead of purchasing the more energy efficient equipment.[25]
Georgia Power’s DSM Application does not demonstrate that it has evaluated the moral hazard problem. As a result, it is likely that the Application overestimates the amount of conservation that will actually occur as a result of the DSM programs.
Q. WHAT SPECIFIC CONCERN DO YOU HAVE WITH THE RESIDENTIAL HIGH EFFICIENCY NEW HOMES PROGRAM?
A. One very curious aspect of the Residential High Efficiency New Homes program is the requirement that “in order to be eligible for the rebate, heating systems and water heating in the home must be electric.”[1] This charge seems to be in conflict with the logic chart supporting this program.
The logic chart states that the objective of the Residential High Efficiency New Homes program is to “maximize achievable energy efficiency in new home by helping homebuyers reduce future energy consumption while lowering their future utility bills.” The problem is that the vast majority of recent new electricity capacity additions have been combined-cycle natural gas fired generation. It is more energy efficient to use natural gas directly for home heating than to use natural gas in combined-cycle central electricity generation and then transport the electricity generated by natural gas to the home. But this more energy efficient outcome is foreclosed by the requirements of the High Efficiency New Homes program.
Q. WHAT IS YOUR SUMMARY OF YOUR CONCERNS?
A. Advocates of demand side management frequently overlook the human dimension. People make conservation decisions every day. Families are constantly striving to maximize the benefits they receive from using energy with the cost of using that energy.
Some economic analyses try to argue that ratepayers systematically misestimate the proper discount rate and therefore DSM is required to fix this error. But those analyses fall victim to assuming they know what’s best for ratepayers. The people who know best are ratepayers themselves. Only the individual ratepayer can decide how he or she best allocates their scare resources.
DSM leads to inefficient outcomes. It increases rates for all ratepayers, but only benefits a few of the ratepayers. Because history shows that DSM programs do not typically lead to the demand reductions claimed, they do not lead to lower electricity rates overall.
DSM is not new. It has been tried and studied for years. DSM is subject to the well known problems of the rebound effect, free riders, moral hazard, to name a few. Georgia Power’s 2010 Integrated Resource Plan and Application for Certification of Demand Side Management Programs analysis does not explain how this program will overcome these well-known economic problems nore does it explain how it will overcome the problems with Georgia Power’s DSM program from the 1990s. Without an analysis, it is difficult to believe Georgia Power’s DSM programs will achieve the benefits Georgia Power projects.
Q. DOES THIS CONCLUDE YOUR TESTIMONY?
A. Yes.
[2] Georgia Power, Application for the Certification of Demand Side Management Programs, Docket #31082, p. 2, Jan. 2010.
[3] See e.g. James Buchanan, Cost and Choice: An Inquiry Into Economic Theory (1969).
[4] Ronald J. Sutherland, Public Interest Comment on Light Truck Average Fuel Economy Standard Model Years 2008–11, Mercatus Center, http://www.mercatus.org/sites/default/files/publication/Fuel_Economy_Standards_for_Light_Trucksv2.pdf.
[5] Avinash Dixit and Robert S. Pindyck, Investments Under Uncertainty, Princeton, NJ, Princeton University Press, 1994.
[6] Avinash Dixit, “Investments and Hysteresis” The Journal of Economic Perspectives, 1992, 6(1), Winter,
pp. 107-132. Saman Majd and Robert S. Pindyck, “Time to Build, Option Value, and Investment Decisions,” Journal of Financial Economics, 1987, 18, pp. 7–27. Robert McDonald and Daniel Siegel, “The Value of Waiting to Invest,” Quarterly Journal of Economics, 1986, 101, pp. 707–728.
[7] Gilbert Metcalf and Donald Rosenthal, “The ‘New’ View of Investment Decisions and Public Policy Analysis: An Application of Green Lights and Cold Refrigerators,” Journal of Policy Analysis and Management, 1995, 14 (4), pp. 517-531.
[8] Kevin A. Hassett and Gilbert Metcalf, “Energy Conservation Investment: Do Consumers Discount the Future Correctly?,” Energy Policy, Vol. 21, June 1993, pp. 710-716.
[9] Ronald J. Sutherland, Public Interest Comment on Light Truck Average Fuel Economy Standard Model Years 2008–11, Mercatus Center, http://www.mercatus.org/sites/default/files/publication/Fuel_Economy_Standards_for_Light_Trucksv2.pdf.
[10] See Kevin A. Hassett & Gilbert Metcalf, Energy Conservation Investment: Do Consumers Discount the Future Correctly? 21 Energy Policy, June 1993.
[11] Matthew C. Hoffman, The Economic Fallacies of Demand-Side Management, Competitive Enterprise Institute, May 1995, http://cei.org/gencon/030,04661.cfm.
[12] David Loughran & Jonathan Kulick, Demand-Side Management and Energy Efficiency in the United States, 25 Energy Journal 19, 2004.
[13] David Loughran & Jonathan Kulick, Demand-Side Management and Energy Efficiency in the United States, 25 Energy Journal 19, 2004.
[14] Franz Wirl, Lessons from Utility Conservation Programs, 21 The Energy Journal 87, 102–103, 2000.
[15] Franz Wirl, Lessons from Utility Conservation Programs, 21 The Energy Journal 87, 88, 2000.
[16] See e.g. Georgia Power, Application for the Certification of Demand Side Management Programs, Docket #31082, Program Plan: Residential High Efficiency New Homes, p. 3, Jan. 2010.
[17] Gilbert Metcalf, Energy Conservation in the United States: Understanding its Role in Climate Policy, May 2006, http://ase.tufts.edu/econ/papers/200609.pdf.
[18] For more information see e.g. Franz Wirl, Lessons from Utility Conservation Programs, 21 The Energy Journal 87, 2000.
[19] See e.g. Georgia Power, Application for the Certification of Demand Side Management Programs, Docket #31082, Program Plan: Residential High Efficiency New Homes, p. 3, Jan. 2010.
[20] For more information see e.g. Franz Wirl, Lessons from Utility Conservation Programs, 21 The Energy Journal 87 at 95, 2000.
[21] Matthew C. Hoffman, The Economic Fallacies of Demand-Side Management, Competitive Enterprise Institute, May 1995, http://cei.org/gencon/030,04661.cfm.
[22] Matthew C. Hoffman, The Economic Fallacies of Demand-Side Management at 22, Competitive Enterprise Institute, May 1995, http://cei.org/gencon/030,04661.cfm.
[23] See e.g. Franz Wirl, Lessons from Utility Conservation Programs, 21 The Energy Journal 87 at 97, 2000; Steven Nadel, Lesson Learned: A Review of Utility Experience with Conservation and Load Management Programs for Commercial and Industrial Customers ( 1990).
[24] Franz Wirl, Lessons from Utility Conservation Programs, 21 The Energy Journal 87 at 97, 2000.
[25] Franz Wirl, Lessons from Utility Conservation Programs, 21 The Energy Journal 87 at 97, 2000.
Daniel:
What DSM really means is Deceptive Strategic Marketing; a term credited to Joel Gilbert (http://www.apogee.net/joelGilbert.aspx). DSM, as typically applied, allows electric utilities build load under the guise of energy efficiency. This is my area of expertise. The following links present two views from the trenches:
http://www.pur.com/pubs/2073.cfm
http://multinationalmonitor.org/hyper/mm0994.html#econ
Call me at 314-342-0714 if you care to discuss these perspectives further.
Mark
[…] (Institute for Energy Research) The Left’s Energy Efficiency Gamble (Heritage Foundation) Demand-Side Management: Government Planning, Not Market Conservation (Testimony before the Georgia Public Service Commission by IER’s Daniel […]
[…] have been around for a long time and they are found wanting. A well-referenced MasterResource article details the many pitfalls of such programs. They say that “DSM programs are fundamentally […]