Ask the Scholar

Document scope · 1 page
doc
Scholar
Ask about this object, its catalog metadata, its source description, or the page inventory. For page-specific OCR and visual context, open one of the page chats.

Scholar Source Context

Document identity
localId
225358479
label
Partial Staff Only Review Docs - [Climate Change] [3]
core
doc
dtoType
document
pageCount
1
Source metadata
Source extras
naId
225358479
levelOfDescription
fileUnit
otherTitles
42-t-5216833-20171095F-017-005-2019
recordType
description
ocrSource
nara-archive
Single page context
seq
1
pageIndex
0
type
document
mediaId
7b86b6d8cda9a5df
ocrText
FOIA Number: 2017-1095-F FOIA MARKER This is not a textual record. This is used as an administrative marker by the William J. Clinton Presidential Library Staff. Collection/Record Group: Clinton Presidential Records Subgroup/Office of Origin: Council of Economic Advisers Series/Staff Member: Michele Jolin Subseries: OA/ID Number: 23919 FolderID: Folder Title: Partial Staff Only Review Docs - [Climate Change] [3] Stack: Row: Section: Shelf: Position: S 21 1 11 2 Review of DOE Labs Study Unspecified Policies: No Program Descriptions: The authors do not estimate the costs or even describe the programs necessary to stimulate the adoption of the new energy-efficient technology they find could reduce emissions of approximately 200 million metric tons of carbon equivalent (MMTCE) by 2010. These reductions correspond to about ½ of the reductions needed to attain 1990 levels by 2010. In the absence of cost estimates this information is of little use in the evaluation of alternative policy options. Costs of Programs: Given reasonable assumptions, government programs may cost many tens of billions of dollars per year above the costs of existing programs to reduce 200+ million tons of carbon. The marginal cost of emissions reductions, including the cost of the government program, is likely to greatly exceed the permit price. As a result these programs may be less cost-effective than a cap-and-trade emissions control program. Luck: The transportation sector requires "luck" in terms of technological innovation to achieve the projected emissions reductions. Without a permit price, this luck appears likely only through stringent standards, such as increasing CAFE. Overestimates Reductions/Underestimates Costs: Behavioral Changes Ignored: The analysis appears to ignore the behavioral changes that offset the energy efficiencies promised by the new technologies. For example, more efficient windows lead to houses with more glass, and more efficient cars lead to more vehicle miles traveled. Thus these innovations offer real value to consumers, but smaller energy savings than engineers forecast. Cost Effectiveness Calculated Using Average, Not Marginal Prices: Many of the innovations in electrical use appear to be assumed to be cost effective based on an assumption of average, not marginal electricity prices. The marginal prices may be half as much as the average prices. Price Declines Resulting from Reductions in Energy Demand Are Not Taken into Account: The study assumes a 12% decrease in energy use, but does not account for this effect on price when calculating cost effectiveness for energy efficient technologies. Study Double Counts Emissions Reductions Occurring Under Business-As-Usual: The labs study double counts 22 MMTCE in reductions in the efficiency case that will occur under the business-as-usual case. Interactions Ignored, Double Counting "Likely" in Utilities Sector Analysis: The double counting of dispatching and retooling coal plants to natural gas results in about 9 MMTCE of overestimated reductions. Industry Sector Analysis Ignores the Costs of Accelerating Retirement. The Five Lab Study Five National Labs Assess Potential of Energy Technologies to Reduce Carbon Emissions - LABS: Lawrence Berkeley, Oak Ridge, Argonne, Renewable Energy, Pacific Northwest - Externally Peer Reviewed: U.Tenn, Monsanto, EPRI, GRI, Harvard, NAS, Stanford (Huntington), UNC (Link), and UCSB (DeCanio) Assumes - Expanded Technology Strategy (R&D and Diffusion) - Carbon Dioxide has a price and is traded Lab Study Results US Carbon Emissions in MMTCE 1800 1800 1700 1700 EIA Carbon Estimate Business as Usual 1600 - 1600 1500 - 1500 * $ 25/T 2010 Impact of High Efficiency + 1400 Low Carbon Technologies for 1400 Two Permit Trading Prices * $50/T 1300 1300 1200 1200 1990 1997 2010 Low-Carbon Technologies Technology Cost to Incremental Carbon Reduction Generate Cost Potential *cents/kWh $/ton Carbon (MMT) Utilities - Carbon Dispatch -- $30 55 - Gas Repowering 2.5 - 3.2 $30 40 (24-83+) - Biomass co-firing 2.7 - 3.2 $38 17 (16-24+) - Wind 2.5 - 3.5 $42 7 (6-20+) - Other -- $25 9 (7-12) Industry - Advanced Turbine 2.5 - 3.5 $40 17 (15-26) - Industry Specific -- $40 14 (13-16) Buildings - Fuel Cell 5.0 - 6.0 $30 3 Transportation - Non-corn Ethanol -- -- 16 Total (rounded) 180 (160-250) *Average costs as of 2005 Energy Efficient Technologies Technology Carbon Reduction Potential (MMT) Utilities - Generation Efficiency 8 (7-13) Industry - R&D and Diffusion 51 Buildings - Standards and Diffusion 59 Transportation - Passenger Cars 28 - Light Trucks 28 - Heavy Trucks 14 - Aircraft 14 Total (rounded) 200 Business and Consumer Annual Costs and Cost Savings in 2010 40 35 Costs 30 Cost Savings $ billions 25 20 15 10 5 0 Utilities Industry Buildings Transportation 08/12/97 TUE 12:33 FAX 202 6222633 001 Office of Economic Policv Department of the Treasurv Washington. D. C. am FAX mors 5 labs Date: August 12, 1997 Number of pages including cover sheet: Luku Eromo Name FAX Number Phone Number To: Jeff Frankel 395-6947 Jay Shogren 395- From: David Wilcox 202-622-2633 2-2200 REMARKS: Urgent For your review Reply ASAP Please comment Attached are partial comments. Having an advance look may be useful. 08/12/97 TUE 12:33 FAA 202 P.002/011 AUG-11-1997 14:11 CEA Wilcox MEMORANDUM TO: Jay Shogren FROM: Joe Aldy and Quindi Franco DATE: August 11, 1997 RE: Review of DOE Labs Study, June 10 Draft We provide an assessment of U.S. Carbon Reductions by 2010 and Beyond: The Potential Impact of Energy-Efficient and Low-Carbon Technologies (DOE Labs Study, June 10 draft). The report identifies carbon reductions and their associated costs in the buildings, industry, transportation, and electric power sectors under two hypothetical climate policy scenarios. The study finds that, with unspecified "very aggressive policies" and a $50/ton permit price, carbon emissions can be stabilized at 1990 levels by 2010. In should reviewing be the hisklighted we believe it is beneficial to highlight several limitations of the study would First, the authors do not provide describe descriptions of the government programs that stimulate the necessary technology adoption to reduce emissions of 180 million metric tons of carbon equivalent (MMTCE). Based Itind on the actual history of CCAP. we estimate the costs of government programs to achieve these this reductions could total $45 billion (1995$), depending on the form of the marginal cost curve. alarmingly Small We annuitized this figure and found that government programs could cost $1.1 billion annually into perpetuity. These are costs in addition to existing program costs to improve energy yes efficiency technology development and adoption. Second, the analyses across sectors are not integrated. Therefore, the penetration rate of energy efficient technologies may be overestimated because the separate sectoral analyses do not account for declines in fuel prices. Third, the industry sector analysis ignores the costs of accelerating capital retirement, resulting in either an overestimate of reductions or 20 underestimate of costs. Fourth, the transportation sector requires "luck" in terms of technological innovation to achieve the projected emissions reductions. Fifth, the utilities sector analysis resulted in double counting of emissions reductions. examples of the qualifications can provide additional insight into the projections of emissions reductions and costs. Upon receiving the final version of the report and documentation of the analyses, we will update this review. Doesn't the analysis also ignora Overview of Report implementation costs? The Labs Study estimated the carbon emissions reduction opportunities available through technology development and adoption and fuel switching. The Study employs the Energy Information Administration's 1997 baseline (AEO97) for the buildings and industry sectors. For the transportation sector, the authors changed the AEO97 assumption of increasing fuel efficiency in automobiles to constant fuel efficiency. For the utility sector, the baseline was modified to reflect a fully competitive bulk-power market in the year 2010. Discharge, 1 08/12/97 TUE 12:34 FAX 202 6222633 003 AUG-11-1997 14:11 CEA P.003/011 shutdown, repowering, and new construction decisions were optimized to select the amount of capacity that minimized the cost of the power-supply system plus the cost of unserved energy. ? Compared to AEO97, the final baseline has slightly lower energy prices, larger electricity sales, and a greater share of gas generation. The study provides emissions reductions from this baseline for two scenarios: 1) "efficiency and 2)"high efficiency/low carbon" over the period 1998- 2010. The study describes neither the policies necessary to achieve emissions reductions nor the costs of federal programs for these scenarios, only their outcomes: increased adoption rates and better technologies. more electricity and more gasgeneration we Efficiency Scenario and with more gasoline consuming consumption, a lot more amen't energy in This case assumes that all technologies adopted are at zero or negative net cost to the user given the baselve more aggressive federal policies to stimulate development and diffusion of energy efficient Does This technologies. The Study estimates that 120 MMTCE will be reduced at negative net cost in this make sense wiR scenario. None of these reductions occur in the utility sector as a function of fuel switching or lower what is technology adoption to increase combustion efficiency. HigH saying! Efficiency/Low Carbon Scenario prices? this sensence This case assumes a "greater commitment" to reduce emissions through federal policies, in concert with state and private activities and a $50/ton permit price that stimulates low carbon technologies (read: primarily fuel switching). The policy is announced in 2000 and the policy's restrictions are phased-in through 2010. There is an assumed announcement effect (although an ? AEE is not specified), increased domestic and international R&D in low-carbon technologies. and a "change in psychology". The Study estimates reductions of 180 MMTCE from the baseline through energy efficient technology adoption at negative net cost (see high efficiency without low carbon technology in chart below). The carbon permits will yield an additional 150- 200 MMTCE of reductions through low-carbon technology adoption. Given that these occur at a permit price of $50/ton, the Study estimates the upper limits of these costs to be $10 billion per year in 2010. Carbon Emissions (MMTCE): 1990 2010 1990 2010 2010 2010 High Efficiency 2010 High Baseline Efficiency (w/o Low Carbon Tech) Efficiency/Low Carbon 1340 1720 1600 1540 1340 1390 Review The Lab Study employs several important assumptions in the analyses. Learning about these qualifications can enhance one $ understanding of the estimates regarding cost-effectiveness and- 2 08/12/97 TUE 12:35 FAX 202 6222633 004 P.004/011 AUG-11-1997 14:12 CEA discuss emissions reductions in this study. We present several of the most significant assumptions end- discussions of their impact below. The Study Does Not Specify Policies to Achieve Carbon Reductions "Cost effectiveness is improved because R&D, in combination with increased deployment efforts. result in declining capital costs. We do not specify the policies, economic conditions, or exogenous events that could precipitate such changes" (p. 1.4). The entire study is premised on two sets of undefined policies: an "aggressive" or "invigorated" public sector effort to stimulate energy efficient technology adoption (efficiency case) and a "very aggressive" public sector effort to stimulate technology development and adoption (high efficiency/low carbon case). The only specific policy instrument mentioned in the report is a $50/ton tradeable permit. The study makes vague references to policies and only provides details morely Tea what these unidentified policies would result in: 1) "better technology" (p. 1.5); 2) "higher penetration rates" (p. 1.6); 3) "changing the capital recovery factor [in industry sector] from 33% to 15%" (p. 4.8); and 4) "technological breakthroughs" (p. 5.3). These policy effects would result in reductions of 180 MMTCE for "free": firms and individuals receive energy cost-savings in excess of technology adoption and implementation costs. I thought these were ignored? The report specifically states that the efficiency case reduces, but does not eliminate, market barriers (p. x), and that implementation costs are ignored (p. xvi). First, we will discuss the implications of reducing market barriers, and then WE will address the cost issue. Market barriers" can occur even in well-functioning but occasionally complex, private markets. Information is fully This is transmitted among all those in the market, although some information may seem incomplete to too text- non-participants (read: policy or market analysts) A person addresses all real costs, but these Even in The costs can go undetected by analysts, hence the term "hidden costs". For market barriers, private lookish. costs match social costs, and barners resource allocation is efficient. Policy intervention to eliminate a * market barrier cannot make society better off and usually makes society worse off, regardless of could inserture the implementation mechanism.¹ ial here from next pacf. may help mater The economics literature has identified several barriers that explain the slow rate of technology adoption:- 1) qualitative attributes differ across technologies; 2) private attributes of information; 3) private discount rate diverges from social rate) and 4) crogeneous usage levels. Qualitative artributes of technologies can affect adoption Consumers of technology prefer Energy Heiency is not one only thing that ma Mcrs to consumers. Indent technologies because of a set of characteristics, not just energy efficiency. For example, some technology consumers may purchase a product with a lower efficiency than another product because the former is more reliable than the efficiency-superior product. I In contrast, policy intervention to eliminate a market failure (such as carbon emissions) may make society better off, depending on the costs of the intervention. 3 08/12/97 TUE 12:35 FAX 202 6222633 005 AUG-11-1997 14:12 CEA P.000/011 New technologies may be costly to integrate with existing dues. until Private attributes of information include the tasks of learning how a new technology works with all compt one's existing suite of technologies and identifying technology suppliers. Transaction costs exist complex ofa would expect arbitrage to occur where middlemen consultants would assess the needs of syskmare with these tasks. If opportunities for economic gains in supplying this information exist, one potential adopters and match them with the appropriate suppliers. it may make sense fully depreciated replace an energy-inetticient component with another likenge- The social discount rates to evaluate new technologies may not fully capture the uncertainty and inefficient N° risks faced by individuals and firms. These private actors may use a much higher discount rate component reflecting the uncertainty of future energy prices, the reliability of the technology, the irreversibility of a technology investment, and the constraints on their ability to borrow capital. If innest ment is irreversible, secial = private, doesn't it? Heterogeneous use of energy intensive technology can affect who benefits from does efficiency not mean that it is investments and who does not. Just If a technology is cost-effective on average may cost- effective for individuals, who do not use much energy. Those who may use the device less 1 intensurely may find it optimal to purchase a less efficient model. These four types of market barriers illustrate the economic behavior that explains the apparent under adoption of energy efficiency technology (the "efficiency gap") These barriers clearly demonstrate the difference between the technological maximum adoption (which only assumes technology costs) and the economic maximum adoption (which assumes total costs). set of limate change policies to stimulate technology adoption could increase efficiency, but only because they also decrease the effect of a market failure (carbon emissions). Social welfare is improved because the net benefits of indirectly fixing the market failure exceed the costs of could move removing the barriers If carbon emissions are appropriately priced (assuming in this case that $50/ton is the correct carbon price), a policy to remove a barrier (e.g., lower the capital recovery of this factor from 33% to 15%), will generate negative net benefits. High adoption rates will be realized once prices rise enough for some technologies to elear the barriers. pase. to From the literature on market barriers, we know that the costs of reducing barriers are greater than their benefits. In this the authers of the Labs Study da not provide an estimate of the costs to government of to lewer market barriers to achieve these substantial duations only provide benefits in terms of projected emissions reductions). We calculated estimates of the costs of emissions reduction programs based on the government's experience with the Climate Change Action Plan (CCAP). CCAP promotes carbon reductions through a broad array of voluntary programs that stimulate "cost-effective"wechnolog financially adoption by in private firms. Participation hva finn in a CCAP program implies that carbon IS reduced and the firm gains reducing financially. emig CCAP received appropriations totaling $494 million (1995$) during the FY95 FY97 period. During this period, the Department of Energy and the Environmental Protection Agency can account for reductions of 14 MMTCE (see attached table). Each ton of carbon reduced in CCAP cost the federal government. on average, $35.29 (1995$) over this three year period. Assuming a flat marginal cost curve (MC = AC of CCAP) to reduce 180 MMTCE beyond the baseline, the government costs of the unidentified policies in the Labs Study would Need footnote Note that CCAP efforts are incorporated In the AE097 baseline. I think 4 questroning why any goit action is required to attain truly cost Heating reductions 08/12/97 TUE 12:37 FAX 202 6222633 P.006/011 AUG-11-1997 14:13 CEA attaining the reduction 180 MMTCE PER year A exceed $6.3 billion. To account for the opportunity COST of capital through time, we annuhized this value and found that overnment costs would about the S150 million annuallysisto perpetality However, the marginal IS costs of technology adoption in the efficiency and high efficiency/low carbon scenarios are definitely higher than the 1075 average could cost of be 14 million much metric tons reduced in the baseline scenario, especially agaregate as more and more efforts to reduce emissions are higher. undertaken. Assuming that costs increase 10% for every 10 million metric tons of carbon reduced. then the government cost would total $16.1 billion (annuffy of $375 million). This is a slowly increasing marginal cost curve given baseline assumptions of gains in energy efficiency. Assuming that costs increase 20% for every 10 million metric tons of carbon reduced, then the cost to government would exceed $45.2 billion (annuity of of ST.T bilien) (see attached chart). The Sectoral Analyses Are Not Integrated "The model runs for each of the three end-use sectors were not integrated and therefore may overstate the effects of technology penetration. In an integrated modeling effort, fuel prices might fall as consumption declines, resulting in less penetration of energy-conserving technologies" (p. 1.1). The analyses of the various sectors assume that the reduction of 180 MMTCE occurs through energy efficient technology adoption without any price effect. This implies that, with government programs, private agents will adopt technologies because the benefits (energy cost- savings) exceed the costs of adoption. The authors claim that these reductions occur at zero or negative net costs. By failing to account for the effect of decreased energy demand due to energy efficiency technology adoption, some of those adoption decisions that occur on the margin would no longer generate positive net benefits for private consumers. The Labs Study indicates that energy consumption will decline from the baseline by 11.6% under the high efficiency/low carbon case (p. xi). This decrease in demand should result in a decrease in the price of energy, and cause some energy efficient technologies to become unattractive to private agents. In addition. the absence of an integrated analysis precludes an assessment of the economy-wide effects of a tradeable permit system. As previous studies have indicated, the nature of the permit allocation (e.g., grandfathering, auction, or a hybrid) and the characteristics of the revenue recycling (e.g., various adjustments to existing taxes) can affect economy-wide investment. Understanding the effects on investment is instrumental in assessing the likelihood of success of these scenarios, given that they rely on substantial R&D and technology adoption. The Industry Sector Analysis Ignores the Costs of Accelerated Retirement "When the economic losses of accelerated retirement are accounted for this means that, at the margin, all investments are not likely to be cost effective at our assumed 15% CRF [capital recovery factor]. Since we do not have a model to 3 We employed a 7% discount rate to calculate the annuities in this paper. 5 08/12/97 TUE 12:38 FAX 202 6222633 AUG-11-1997 14:13 CEA P.007/011 account for this potential early retirement and the economic losses, we must caveat our estimates of investment and net costs. the investment cost may be understated by the amount of loss due to any early retirement that may occur" (p. 4.15). This assumption implies that one of three results should be accounted for in the analysis. First, the projected emissions reduction should be revised downward, because the assessment of the industry sector overestimates technology adoption by ignoring these costs. Second, the cost of reducing emissions. if the reduced carbon from this sector remains constant, should increase, with some technology adoption decisions occurring at positive net costs. Third, if the authors maintain the same costs and emissions estimates, then the costs of government programs to somehow force down the CRF from 33% to 15% would increase. The Transportation Sector Requires "Luck" to Achieve Carbon Reductions "[B]ecause the [transportation sector] outcomes postulated in the high efficiency/low carbon scenario require technological breakthroughs, they require a certain degree of luck to be achieved by 2010. There are no credible methods to accurately gauge the probability of such breakthroughs - we believe they stand a decent chance of occurring with an intensification of research efforts, but we stop short of claiming they are a likely outcome of such an intensification" (p. 5.3). Note that this statement includes two qualifications: luck on top of intensified research efforts. To achieve just the efficiency level of emissions reductions for this sector (73 MMTCE), these intensified research efforts may require two to ten times existing funding on transportation (p. 5.3). To gain a sense of the magnitude of what such an increase in funding might be, consider that the Partnership for a New Generation of Vehicles (PNGV) alone is funded at the federal level at $263 million. Private sector costs of participating in the PNGV are in addition to this large sum. Further, the estimates of costless (on net) carbon reductions relies on a series of tenuous assumptions: commercial development of the fuel cell for passenger cars, commercial availability of cellulosic ethanol (and the elimination of the ethanol excise tax exemption), and an apparently arbitrary 30% reduction in costs for "certain key technologies" (p. 5.24). These technologies are not identified, and the report does not provide documentation for the NEMS model runs conducted for the transportation sector. Interactions Ignored, Double Counting "Likely" in Utilities Sector Analysis "Chapter 7 considers other electricity supply technology options. including: converting coal-based power plants to natural gas; cofiring coal with biomass; efficiency improvements in generation and transmission and distribution systems; extending the life of existing nuclear plants; increasing generation and capacity of existing hydropower plants; and constructing new powerplants using advanced coal technologies. Each of these options is assessed independently. Thus. 6 08/12/97 TUE 12:38 FAX 202 6222633 AUG-11-1997 14:14 CEA LIA/700'- interactions between the options are not taken into account, and the possibility of double counting is therefore likely" (p. 7.1). The emissions reductions in the electricity sector come from two sources. First, utilities undertake fuel switching at the $50/ton permit price through carbon-ordered dispatching. Second, utilities modify their capital stock through several technology options such as those listed in the above quotation. As electricity production moves away from coal in the first analysis, fewer and fewer plants (and therefore potential emissions reductions) will be available for conversion to natural gas or cofiring with biomass. The second set of analyses were "static" and did not optimize unit/plant production cost, dispatch, or system load. Conversations with one of the co-authors (Stanley Hadley of Oak Ridge National Laboratory) indicate that 10 GW of coal-to-gas conversions (approximately 9 million tons of carbon reductions) were double- counted. Furthermore, the analysis of specific options did not examine the effect of increasing fuel prices (from gas demand due to coal-to-gas repowering). Given the range of assumptions considered in the study, the authors actually estimated that repowering coal plants for natural gas could result in carbon reductions between 5 million and 269 million tons (p. 7.2), depending on the gas/coal price differential, the cost of carbon. and the costs of sulfur dioxide and nitrous oxide emissions. This is quite a substantial range. The Labs Study estimates that natural gas consumption will increase 14% to 191% above 2010 baseline consumption. However, the DRI 1.25 run in the IAT report indicates a 16% decline in natural gas consumption, while the SGM run resulted in only a negligible decline and Markal-Macro generated a 6% increase in natural gas consumption. These three models indicate that a non-integrated analysis that does not account for the price change in natural gas may overestimate natural gas substitution for coal and emissions reductions from fuel switching. 4 Hadley noted that a subsequent draft of the report should address at least some of the double counting. but he did not provide details. 7 08/12/97 TUE 12:39 FAX 202 6222633 AUG-11-1997 14:14 CEA Summary of Greenhouse Gas Emissions-Reduction Actions Million Metric Tons of Carbon Equivalent Source: U.S. Climate Action Report-1997 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date Residential & Commercial Sector Actions 26.9 10.3 - New Rebuild America 2.0 1.6 -- 1 and 2 Expanded Green Lights and Energy 3.6 3.3 1 Star Buildings 3 State Revolving Fund for Public 1.1 Terminated Buildings 4 Cost-Shared Demonstrations of Emerging Technologies 5 Operation and Maintenance Training 3.8 0.0 : for Commercial Building Facility Managers and Operators 6 Energy Star Products 5.0 4.3 I 7 Residential Appliance Standards 6.8 0.2 I S and 11 Energy Partnerships for Affordable Housing 9 Cool Communities 4.4 0.4 - 10 Update State Building Codes New Construction of Energy-Efficient 0.1 : Commercial and Industrial Buildings New Superwindow Collaborative 0.0 : New Expand Markets for Next-Generation 0.2 Lighting Products New Fuel Cells Initiative 0.0 I Industrial Sector Actions 19.0 4.8 I 12 Motor Challenge 8.8 1.8 -- 13 Industrial Golden Carrot Programs 2.9 Merged into Action 12 14 Accelerate the Adoption of Energy- Terminated Efficient Process Technologies 15 Industrial Assessment Centers 0.5 CCAP Component Terminated 16 Waste Minimization 4.2 2.1 -- 17 Improve Efficiency of Fertilizer 2.7 0.8 -- Nitrogen Use 18 Reduce the Use of Pesticides Terminated Transportation Sector Actions 8.1 5.3 - 19 Cash Value of Parking 20 Innovative Transportation Strategies 6.6 4.6 - 21 Telecommuting Program 22 Fuel Economy Labels for Tires 1.5 0.7 : Energy Supply Actions 10.8 1.3 -- 23 Increase Natural Gas Share of Energy Use Though Federal Regulatory 2.2 Terminated Reform 08/12/97 TUE 12:39 FAX 202 6222633 AUG-11-1997 14:15 CEA P.010/011 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date 24 Promote Seasonal Gas Use for Control 2.8 0.5 - of Nitrogen Oxides 25 High-Efficiency Gas Technologies 0.6 Terminated 26 Renewable-Energy Commercialization 0.8 0.3 - 27 Expand Utility Integrated Resource 1.4 Terminated Planning 28 Profitable Hydroelectric Efficiency 2.0 0.0 -- Upgrades 29 Energy-Efficient Distribution Transformer Standards 0.8 0.5 - 30 Energy Star Distribution Transformers 31 Transmission Pricing Reform 0.8 Terminated New Green Power Network Not included 0.0 : Land-Use Change & Forestry Actions 10.0 2.4 - 43 Private Depletion of Nonindustrial 4.0 Terminated Private Forests 44 Accelerate Tree Planting in 0.5 0.4 - Nonindustrial Private Forests 16 Waste Minimization 4.2 2.0 -- 9 Expand Cool Communities 0.5 To be determined Methane Actions 16.3 15.5 - 32 Expand Natural Gas STAR 3.0 3.4 -- 33 Increase Stringency of Landfill Rule 4.2 6.3 -- 34 Landfill Methane Outreach Program 1.1 1.9 -- 35 Coalbed Methane Outreach Program 2.2 2.6 - 36 RD&D for Coal Mine Methane 1.5 Terminated 37 RD&D for Landfill Methane 1.0 Terminated 38 AgSTAR Program 1.5 0.3 -- 39 Ruminant Livestock Efficiency 1.3 1.0 : Program Actions to Address Other Greenbouse Gases 16.3 25.4 - 17 Improved Fertilizer Management 4.5 5.3 - 40 Significant New Alternatives Program 5.0 6.4 -- 41 HFC-23 Partnerships 5.0 5.0 - 42 Voluntary Aluminum Partnership 1.8 2.2 -- New Environmental Stewardship Initiative Not included 6.5 - Foundation Actions 11.3 - Climate Wise Not estimated 1.8 : Climate Challenge Not estimated 7.6 -- State and Local Outreach Programs Not estimated 1.9 - Total GHG Emission Reductions From CCAP 108.6 76.0 14.0 Data is not readily available for cumulative emissions reductions for many CCAP programs. Emissions reductions of about 5 MMTCE can be attributed to DOE's CCAP programs. EPA's Office of Air and Radiation is responsible for emissions reductions of about 9 MMTCE through their CCAP programs. Costs of Reductions ($/ton) AUG-11-1997 14:15 ,000 TC = $45.2 billion; Annuity = $1.1billion 800 TC = $16.1 billion; Annuity = $375 million 08/12/97 TUE 12:40 FAX 202 6222633 CEA 600 TC = $6.4 billion; Annuity = $148 million 400 200 0 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 130-140 140-150 150-160 160-170 170-180 Carbon Reductions (MMTC) MC = AC of CCAP MC Increases 10% Every 10 MMTC P.011/01 MC Increases 20% Every 10 MMTC TTOM DOELABS3.MEM Page 1 MEMORANDUM TO: Jay Shogren FROM: Joe Aldy and Quindi Franco DATE: August 11, 1997 RE: Review of DOE Labs Study, June 10 Draft We provide an assessment of U.S. Carbon Reductions by 2010 and Beyond: The Potential Impact of Energy-Efficient and Low-Carbon Technologies (DOE Labs Study, June 10 draft). The report identifies carbon reductions and their associated costs in the buildings, industry, transportation, and electric power sectors under two hypothetical climate policy scenarios. The study finds that, with unspecified "very aggressive policies" and a $50/ton permit price, carbon emissions can be stabilized at 1990 levels by 2010. We believe several limitations of the study should be highlighted; some of these are serious. First, the authors do not estimate the costs or even describe the programs necessary to stimulate the adoption of technology necessary to reduce emissions of 180 million metric tons of carbon equivalent (MMTCE) by 2010. These reductions correspond to about 1/3 of the reductions needed to attain 1990 levels by 2010. In the absence of cost estimates it is impossible to evaluate the merit of the technology options. --Based on the actual history of CCAP, we estimate the costs of government programs to achieve these reductions could total $45 billion (1995$), depending on the form of the marginal cost curve. We annuitized this figure and found that government programs could cost $1.1 billion annually into perpetuity. These are costs in addition to existing program costs to improve energy efficiency technology development and adoption. Second, the penetration rate of energy efficient technologies may be overestimated because the separate sectoral analyses do not account for declines in fuel prices that would result from adoption. Third, the industry sector analysis ignores the costs of accelerating capital retirement. Fourth, the transportation sector requires "luck" in terms of technological innovation to achieve the projected emissions reductions. Fifth, the utilities sector analysis resulted in double counting of emissions reductions. Upon receiving the final version of the report and documentation of the analyses, we will update this review. DOELABS3.MEM Page 2 Overview of Report The study provides emissions reductions for two scenarios: 1) "efficiency" and 2)"high efficiency/low carbon", over the period 1998-2010. The study describes neither the policies necessary to achieve emissions reductions nor the costs of federal programs for these scenarios, only their outcomes: increased adoption rates and better technologies through technology development and adoption and fuel switching. The Study employs the Energy Information Administration's 1997 baseline (AEO97) for the buildings and industry sectors. For the transportation sector, the authors changed the AEO97 assumption of increasing fuel efficiency in automobiles to constant fuel efficiency. For the utility sector, the baseline was modified to reflect a fully competitive bulk-power market in the year 2010. Discharge, shutdown, repowering, and new construction decisions were optimized to select the amount of capacity that minimized the cost of the power-supply system plus the cost of unserved energy. Compared to AEO97, the final baseline has slightly lower energy prices, larger electricity sales, and a greater share of gas generation. Efficiency Scenario This case assumes that all technologies adopted are at zero or negative net cost to the user given more aggressive federal policies to stimulate development and diffusion of energy efficient technologies. The Study estimates that 120 MMTCE will be reduced at negative net cost in this scenario. No fuel switching by utilities or industry occurs in this scenario. High Efficiency/Low Carbon Scenario This case assumes a "greater commitment" to reduce emissions through federal policies, in concert with state and private activities and a $50/ton permit price that stimulates low carbon technologies (read: primarily fuel switching). The policy is announced in 2000 and the policy's restrictions are phased-in through 2010. There is an assumed announcement effect (although an autonomous energy efficiency index is not specified), increased domestic and international R&D in low-carbon technologies, and a "change in psychology". The Study estimates reductions of 180 MMTCE from the baseline through energy efficient technology adoption at negative net cost (see high efficiency without low carbon technology in chart below). The carbon permits will yield an additional 150-200 MMTCE of reductions through low-carbon technology adoption. Given that these occur at a permit price of $50/ton, the Study estimates the upper limits of these costs to be $10 billion per year in 2010. Carbon Emissions (MMTCE): 1990 - 2010 1990 2010 2010 2010 High Efficiency 2010 High Baseline Efficiency (w/o Low Carbon Tech) Efficiency/Low Carbon 1340 1720 1600 1540 1340 - 1390 DOELABS3.MEM Page 3 Review The Lab Study employs several important assumptions in the analyses. We discuss several of the most significant assumptions and discussions of their impact below. The Study Does Not Specify Policies to Achieve Carbon Reductions "Cost effectiveness is improved because R&D, in combination with increased deployment efforts, result in declining capital costs. We do not specify the policies, economic conditions, or exogenous events that could precipitate such changes" (p. 1.4). The entire study is premised on two sets of undefined policies: an "aggressive" or "invigorated" public sector effort to stimulate energy efficient technology adoption (efficiency case) and a "very aggressive" public sector effort to stimulate technology development and adoption (high efficiency/low carbon case). The only specific policy instrument mentioned in the report is a $50/ton tradeable permit. The study merely asserts that these unidentified policies would result in: 1) "better technology" (p. 1.5); 2) "higher penetration rates" (p. 1.6); 3) "changing the capital recovery factor [in industry sector] from 33% to 15%" (p. 4.8); and 4) "technological breakthroughs" (p. 5.3). These policy effects would result in reductions of 180 MMTCE for "free": firms and individuals receive energy cost-savings in excess of technology costs. The report specifically states that the efficiency case reduces, but does not eliminate, market barriers (p. x). Market barriers can occur in well-functioning private markets. Even in the presence of such barriers, resource allocation is efficient. Policy intervention to eliminate a market barrier cannot make society better off and usually makes society worse off, regardless of the implementation mechanism. 1 A set of climate change policies to stimulate technology adoption could increase efficiency, but only because they also decrease the effect of a market failure (carbon emissions). Social welfare is improved because the net benefits of indirectly fixing the market failure exceed the costs of removing the barriers. If carbon emissions are appropriately priced (assuming in this case that $50/ton is the correct carbon price), a policy to remove a barrier (e.g., lower the capital recovery factor from 33% to 15%), will generate negative net benefits. High adoption rates will be realized once prices rise enough for some technologies to clear the barriers. The economics literature has identified several barriers that may help explain the slow rate of technology adoption. Energy efficiency is not the only thing that matters to consumers. Other qualities of products appeal to consumers' preferences. For example, some technology consumers may purchase a product with a lower efficiency than another product because the former is more reliable than the efficiency-superior product. New technologies may be costly to integrate with existing technologies. Consumers of new technologies need to expend effort to understand how to match the two sets of DOELABS3.MEM Page 4 technologies. Until all components of a complex system are fully depreciated, it may make sense to replace an energy-inefficient component with another, likewise inefficient component. The social discount rates to evaluate new technologies may not fully capture the uncertainty and risks faced by individuals and firms. These private actors may use a much higher discount rate reflecting the uncertainty of future energy prices, the reliability of the technology, the irreversibility of a technology investment, and the constraints on their ability to borrow capital. Just because a technology is cost-effective on average, it does not mean that it is cost-effective for all individuals. Those who use the device less intensively may find it optimal to purchase a less efficient model. These barriers clearly demonstrate the difference between the technological maximum adoption (which only assumes technology costs) and the economic maximum adoption (which assumes total costs). The Labs Study does not provide an estimate of the cost to government of lowering market barriers. We estimated the costs of emissions reduction based on the government's experience with the Climate Change Action Plan (CCAP). CCAP promotes carbon reductions through a broad array of voluntary programs that stimulate "cost-effective" technology adoption by private firms.2 Firms participating in a CCAP program benefit financially in return for reducing emissions. CCAP received appropriations totaling $494 million (1995$) during the FY95 - FY97 period. During this period, the Department of Energy and the Environmental Protection Agency can account for reductions of 14 MMTCE (see attached table).3 Each ton of carbon reduced in CCAP cost the federal government, on average, $35.29 (1995$) over this three year period. Assuming a flat marginal cost curve (MC = AC of CCAP) to reduce 180 MMTCE beyond the baseline, the government costs of the unidentified policies in the Labs Study would exceed $6.3 billion. To account for the opportunity cost of capital through time, we annuitized this value and found that government costs would run about $150 million annually into perpetuity.4 However, the marginal costs of technology adoption in the efficiency and high efficiency/low carbon scenarios are definitely higher than the average cost of 14 million metric tons reduced in the baseline scenario, especially as more and more efforts to reduce emissions are undertaken. Assuming that costs increase 10% for every 10 million metric tons of carbon reduced, then the government cost would total $16.1 billion (annuity of $375 million). This is a slowly increasing marginal cost curve given baseline assumptions of gains in energy efficiency. Assuming that costs increase 20% for every 10 million metric tons of carbon reduced, then the cost to government would exceed $45.2 billion (annuity of $1.1 billion) (see attached chart). Price Declines Resulting from Reductions in Energy Demand Are Not Taken into Account "The model runs for each of the three end-use sectors were not integrated and therefore DOELABS3.MEM Page 5 may overstate the effects of technology penetration. In an integrated modeling effort, fuel prices might fall as consumption declines, resulting in less penetration of energy-conserving technologies" (p. 1.1). The analyses of the various sectors assume that the reduction of 180 MMTCE occurs through energy efficient technology adoption without any energy price effect. This implies that, with government programs, private agents will adopt technologies because the benefits (energy cost-savings) exceed the costs of adoption. The authors claim that these reductions occur at zero or negative net costs. However, if the energy prices fall due to the decline in energy demand, some of those adoptions would no longer generate positive net benefits for private consumers. The Study indicates that energy consumption will decline from the baseline by 11.6% under the high efficiency/low carbon case (p. xi). This decrease in demand should result in a decrease in the price of energy, and cause some energy efficient technologies to become unattractive to private agents. The efficiency scenario should be modified to reflect the downward effect on prices resulting from decreased energy consumption. The high efficiency scenario should reflect two counteracting effects on prices: the decline, in energy consumption and the $50/ton permit fee. The Study ignores the former effect and insufficiently incorporates the latter effect. For example, in the buildings chapter, the penetration rate is assumed to be 65% instead of 60% of the maximum cost-effective technical potential in this scenario because of the $50/ton permit price.5 However, the energy-cost savings calculations for buildings technologies assume the same energy price as in the business as usual scenario. Since the study assumes that exogenous, undefined policy influences drive the penetration rates, the penetration rate should be set at 60% and the energy cost-savings should be recalculated with the appropriate energy price. In the industry chapter, the high efficiency penetration rate is assumed to be double the rate used in the efficiency scenario, which is the undefined "normal" rate. It is impossible to determine how much of this doubling of penetration results from undefined, aggressive policy efforts and how much results from the permit price. The Industry Sector Analysis Ignores the Costs of Accelerated Retirement "When the economic losses of accelerated retirement are accounted for this means that, at the margin, all investments are not likely to be cost effective at our assumed 15% CRF [capital recovery factor]. Since we do not have a model to account for this potential early retirement and the economic losses, we must caveat our estimates of investment and net costs. the investment cost may be understated by the amount of loss due to any early retirement that may occur" (p. 4.15). The omission of costs related to the early retirement of capital implies that either the projected emissions reduction should be revised downward, because less technology will be adopted; or the cost of reducing emissions should be revised upward; or DOELABS3.MEM Page 6 the cost of government programs to somehow force down the CRF from 33% to 15% would increase. The Transportation Sector Requires "Luck" to Achieve Carbon Reductions "[B]ecause the [transportation sector] outcomes postulated in the high efficiency/low carbon scenario require technological breakthroughs, they require a certain degree of luck to be achieved by 2010. There are no credible methods to accurately gauge the probability of such breakthroughs -- we believe they stand a decent chance of occurring with an intensification of research efforts, but we stop short of claiming they are a likely outcome of such an intensification" (p. 5.3). Note that this statement includes two qualifications: luck on top of intensified research efforts. These intensified research efforts may require two to ten times existing funding on transportation (p. 5.3). For perspective on what this might mean, it is worth keeping in mind that the Partnership for a New Generation of Vehicles (PNGV) alone is funded at the federal level at $263 million, and there are substantial additional private sector costs. The "luck" apparently pertains to a number of tenuous technological assumptions: commercial development of the fuel cell for passenger cars, commercial availability of cellulosic ethanol (and the elimination of the ethanol excise tax exemption), and an apparently arbitrary 30% reduction in costs for "certain key technologies" (p. 5.24). These technologies are not identified, and the report does not provide documentation for the NEMS model runs conducted for the transportation sector. Interactions Ignored, Double Counting "Likely" in Utilities Sector Analysis "Chapter 7 considers other electricity supply technology options, including: converting coal-based power plants to natural gas; cofiring coal with biomass; efficiency improvements in generation and transmission and distribution systems; extending the life of existing nuclear plants; increasing generation and capacity of existing hydropower plants; and constructing new powerplants using advanced coal technologies. Each of these options is assessed independently. Thus, interactions between the options are not taken into account, and the possibility of double counting is therefore likely" (p. 7.1). The emissions reductions in the electricity sector come from two sources. First, utilities undertake fuel switching at the $50/ton permit price through carbon-ordered dispatching. Utilities would change their fuel mix by dispatching more electricity from their lower carbon-emitting facilities (e.g., nuclear, hydropower, and natural gas) and less from coal plants. Second, utilities modify their capital stock through several technology options such as those listed in the above quotation. As electricity production moves away from coal through modified dispatching, fewer and fewer plants (and therefore potential emissions reductions) will be available for conversion to natural gas or cofiring with biomass. The second set of analyses were "static" and did not optimize unit/plant production cost, dispatch, or system load. Conversations DOELABS3.MEM Page 7 with one of the co-authors (Stanton Hadley of Oak Ridge National Laboratory) indicate that 10 GW of coal-to-gas conversions (approximately 9 million tons of carbon reductions, or 7% of reductions for dispatching and conversions) were double-counted.5 AUG-11-1997 14:11 CEH P.002/011 Wilcox MEMORANDUM TO: Jay Shogren FROM: Joe Aldy and Quindi Franco DATE: August 11, 1997 RE: Review of DOE Labs Study, June 10 Draft We provide an assessment of U.S. Carbon Reductions by 2010 and Beyond: The Potential Impact of Energy-Efficient and Low-Carbon Technologies (DOE Labs Study, June 10 draft). The report identifies carbon reductions and their associated costs in the buildings, industry, transportation, and electric power sectors under two hypothetical climate policy scenarios. The study finds that, with unspecified "very aggressive policies" and a $50/ton permit price, carbon beneficial to highlight several limitations of the study would should First, the authors do not provide describe emissions can be stabilized at 1990 levels by 2010. In reviewing LE the Wishted we believe it is descriptions of the government programs necessary that to stimulate the necessary technology 180 ton adoption to reduce emissions of 180 million metric tons of carbon equivalent (MMTCE). Based Itind on the actual history of CCAP, we estimate the costs of government programs to achieve these reductions could total $45 billion (1995$), depending on the form of the marginal cost curve. 1/3 14 of We annuitized this figure and found that government programs could cost $1.1 billion annually small into perpetuity. These are costs in addition to existing program costs to improve energy the way efficiency technology development and adoption. Second, the analyses across sectors are not integrated. Therefore, the penetration rate of energy efficient technologies may be overestimated because the separate sectoral analyses do not account for declines in fuel prices. Third, the steeply industry sector analysis ignores the costs of accelerating capital retirement, resulting in either an mereasi overestimate of reductions or an underestimate of costs. Fourth, the transportation sector requires "luck" in terms of technological innovation to achieve the projected emissions marging reductions. Fifth, the utilities sector analysis resulted in double counting of emissions reductions. cost These examples of the qualifications can provide additional insight into the projections of emissions reductions and costs. Upon receiving the final version of the report and documentation of the analyses, we will update this review. Doesn't the analysis also ignora Overview of Report implementation costs? The Labs Study estimated the carbon emissions reduction opportunities available through technology development and adoption and fuel switching. The Study employs the Energy Information Administration's 1997 baseline (AEO97) for the buildings and industry sectors. For the transportation sector, the authors changed the AEO97 assumption of increasing fuel efficiency in automobiles to constant fuel efficiency. For the utility sector, the baseline was modified to reflect a fully competitive bulk-power market in the year 2010. Discharge, 1 AUG-11-1997 LCH shutdown, repowering, and new construction decisions were optimized to select the amount of capacity that minimized the cost of the power-supply system plus the cost of unserved energy ? Compared to AEO97, the final baseline has slightly lower energy prices, larger electricity sales, and a greater share of gas generation. The study provides emissions reductions from this baseline for two scenarios: 1) "efficiency" and 2) 'high efficiency/low carbon" over the period 1998- 2010. The study describes neither the policies necessary to achieve emissions reductions nor the costs of federal programs for these scenarios, only their outcomes: increased adoption rates and better technologies. with more electricity and more 995 generate Efficiency Scenario and more gasoline conjumption, aren't we consuming a lot more energy in This case assumes that all technologies adopted are at zero or negative net cost to the user given the baselli more aggressive federal policies to stimulate development and diffusion of energy efficient Does This technologies. The Study estimates that 120 MMTCE will be reduced at negative net cost in this make sens scenario. None of these reductions occur in the utility sector as a function of fuel switching or wir what is technology adoption to increase combustion efficiency. lower this High saying! Efficiency/Low Carbon Scenario prices? sentence This case assumes a "greater commitment" to reduce emissions through federal policies, in concert with state and private activities and a $50/ton permit price that stimulates low carbon technologies (read: primarily fuel switching). The policy is announced in 2000 and the policy's restrictions are phased-in through 2010. There is an assumed announcement effect (although an AEE is-not specified), increased domestic and international R&D in low-carbon technologies, and a "change in psychology". The Study estimates reductions of 180 MMTCE from the baseline through energy efficient technology adoption at negative net cost (see high efficiency without low carbon technology in chart below). The carbon permits will yield an additional 150- 200 MMTCE of reductions through low-carbon technology adoption. Given that these occur at a permit price of $50/ton, the Study estimates the upper limits of these costs to be $10 billion per year in 2010. Carbon Emissions (MMTCE): 1990 2010 1990 2010 2010 2010 High Efficiency 2010 High Baseline Efficiency (w/o Low Carbon Tech) Efficiency/Low Carbon 1340 1720 1600 1540 1340 1390 Review The Lab Study employs several important assumptions in the analyses. Loarning about these qualifications can enhance one understanding of the estimates regarding cost effectivenes and 2 discuss emissions reductions in this study. We present several of the most significant assumptions and- discussions of their impact below. The Study Does Not Specify Policies 10 Achieve Carbon Reductions "Cost effectiveness is improved because R&D, in combination with increased deployment efforts, result in declining capital costs. We do not specify the policies, economic conditions, or exogenous events that could precipitate such changes" (p. 1.4). The entire study is premised on two sets of undefined policies: an "aggressive" or "invigorated" public sector effort to stimulate energy efficient technology adoption (efficiency case) and a "very aggressive" public sector effort to stimulate technology development and adoption (high efficiency/low carbon case). The only specific policy instrument mentioned in the report is a $50/ton tradeable permit. The study makes vague references to policies and only provides details. morely assertsthate on what these unidentified policies would result in: 1) "better technology" (p. 1.5); 2) "higher penetration rates" (p. 1.6); 3) "changing the capital recovery factor [in industry sector] from 33% to 15%" (p. 4.8); and 4) "technological breakthroughs" (p. 5.3). These policy effects would result in reductions of 180 MMTCE for "free": firms and individuals receive energy cost-savings in excess of technology adoption and implementation costs. I thought these were ignored? The report specifically states that the efficiency case reduces, but does not eliminate, market barriers (p. x), and that implementation costs are ignored (p. xvi). First, we will discuss the implications of reducing market barriers, and then we will address the cost issue. Market barriers" can occur even in well-functioning, but occasionally complex, private markets. Information is fully This is transmitted among all those in the market, although some information may seem incomplete to so text- non-participants (read: policy or market analysts) A person addresses all real costs, but these EVEN in The costs can go undetected by analysts, hence the term "hidden costs". For market barriers, private Lookish. costs match social costs, and resource allocation is efficient. Policy intervention to eliminate a of berners market barrier cannot make society better off and usually makes society worse off, regardless of could # ial the implementation mechanism.¹ here from next pace. may help mater The economics literature has identified several barriers that explain the slow rate of technology adoption:- 1) qualitative attributes differ across technologies; 2) private attributes of information; 3) private discount rate diverges from social rate) and 4) erogeneous usage levels. Energy efficiency is not the only thing that ma Hers to consumers. Indent Qualitative attributes of technologies can affect adoption Consumers of technology prefer technologies because of a set of characteristics, not just energy efficiency. For example, some technology consumers may purchase a product with a lower efficiency than another product because the former is more reliable than the efficiency-superior product. 1 In contrast, policy intervention to eliminate a market failure (such as carbon emissions) may make society better off, depending on the costs of the intervention. 3 New technologies may be costly to imtegrate with existing ones. until Private attributes of information include the tasks of learning how a new technology works with all compan one's existing suite of technologies and identifying technology suppliers. Transaction costs exist ofa with these tasks. If opportunities for economic gains in supplying this information exist, one comptex would expect arbitrage to occur where middlemen consultants would assess the needs of systemare potential adopters and match them with the appropriate suppliers. it may make sense fully depreciate replace a encrgy-inetticient component with another likenise. The social discount rates to evaluate new technologies may not fully capture the uncertainty and inetticer N° risks faced by individuals and firms. These private actors may use a much higher discount rate companer reflecting the uncertainty of future energy prices, the reliability of the technology, the irreversibility of a technology investment, and the constraints on their ability to borrow capital. If innest ment is inreversible, secial = private, doesn't it? Heterogeneous use of energy Just intensive because technology can affect who benefits from &does efficiency not mean that if is investments and who does not. If a technology is cost-effective on average HE may not be cost- effective for individuals' who do not use much energy. These who may use the device less utensmily may find it optimal to purchase a less efficient model. These four types of market barriers illustrate the economic behavior that explains the apparent under adoption of energy efficiency technology (the "efficiency gap"). These barriers clearly demonstrate the difference between the technological maximum adoption (which only assumes technology costs) and the economic maximum adoption (which assumes total costs). A set of climate change policies to stimulate technology adoption could increase efficiency, but only because they also decrease the effect of a market failure (caibon emissions). Social welfare is improved because the net benefits of indirectly fixing the market failure exceed the costs of (or removing the barriers If carbon emissions are appropriately priced (assuming in this case that $50/ton is the correct carbon price), a policy to remove a barrier (e.g., lower the capital recovery of 0 this factor from 33% to 15%), will generate negative net benefits. High adoption rates will be page. to prevous realized once prices rise enough for some technologies to elear the barriers. From the literature on market barriers, we know that the costs of reducing barriers are greater than their benefits. In this case the authors of the Labs Study do not provide an estimate of the of lowerng costs to government to lower market barriers to achieve these substantial reductions (and they only provide benefits in terms of projected emissions reductions). We calculated estimates of the costs of emissions reductions programs based on the government's experience with the Climate Change Action Plan (CCAP). CCAP promotes carbon reductions through a broad array of voluntary programs that stimulate in "cost-effective" technology adoption by private firms. benefit financially in reducing Participation emissions, by a firm in a CCAP program implies that carbon IS reduced and the firm gains financially. CCAP received appropriations totaling $494 million (1995$) during the FY95 FY97 period. During this period, the Department of Energy and the Environmental Protection Agency can account for reductions of 14 MMTCE (see attached table). Each ton of carbon reduced in CCAP cost the federal government, on average, $35.29 (1995$) over this three year period. Assuming a flat marginal cost curve (MC = AC of CCAP) to reduce 180 MMTCE beyond the baseline, the government costs of the unidentified policies in the 1 abs Study would 2 Need footnote I think Note that CCAP efforts are incorporated in the AE09 baseline. stockflows anced 4 questroning why any govt action is required to attain truly cost attaining the reduction 180 MMTCE BY year ^ exceed $6.3 billion. To account for the opportunity COSE of capital through time, we annuftized this value and found that government costs would sume run about the $150 million annually perpetuity: However, the marginal as 11 costs almos of technology adoption in the efficiency and high efficiency/low carbon scenarios are definitely higher than the average could cost of to 14 million much metric tons reduced in the baseline scenario, especially aggregast as more and 1075 more efforts to reduce emissions are higher. undertaken. Assuming that costs increase 10% for every 10 million metric tons of carbon reduced, then the government cost would total $16.1 billion per (annuffy of $375 million). This is a slowly increasing marginal cost curve given baseline assumptions of gains in energy efficiency. Assuming that costs increase 20% for every 10 million metric tons of carbon reduced, then the cost to government would exceed $45.2 billion (annuity of $1.1 officen) (see attached chart). Price declines resul Hing from reductions in energy demant The Are price increases The Sectoral Analys Not Integrated are not taken into recount. "The model runs for each of the three end-use sectors were not integrated and induced the permits tavien modeling effort, fuel prices might fall as consumption declines, resulting in less by therefore may overstate the effects of technology penetration. In an integrated schime into account? penetration of energy-conserving technologies" (p. IS 1.1). this right? why the with the $50/ton The analyses of the various sectors assume that the reduction of 180 MMTCE occurs through fee? energy efficient technology adoption without any price effect This implies that, with Add FN government programs, private agents will adopt technologies because the benefits (energy cost- savings) exceed the costs of adoption. The authors the enery claim that these fall reductions due Qccur to the at zero decliner or negative net costs. By failing However, to account If for the effectof prices decreased energy demand due to energy efficiency technology adoption, some of those adoption decisions that occur on the margin would no longer generate positive net benefits for private consumers. The Labs Study indicates that energy consumption will decline from the baseline by 11.6% under the high efficiency/low carbon case (p. xi). This decrease in demand should result in a decrease in the price of energy, and cause some energy efficient technologies to become unattractive to private agents. In addition. the absence of an integrated analysis precludes an assessment of the economy-wide effects of a tradeable permit system. As previous studies have indicated, the nature of the permit allocation (e.g., grandfathering, auction, or a hybrid) and the characteristics of the revenue recycling (e.g., various adjustments to existing taxes) can affecteconomy-wide investment. Understanding the effects on investment is instrumental in assessing the likelihood of success of these scenarios, given that/they rely on substantial R&D and technology adoption. The Industry Sector Analysis Ignores the Costs of Accelerated Retirement "When the economic losses of accelerated retirement are accounted for this means that, at the margin, all investments are not likely to be cost effective at our assumed 15% CRF [capital recovery factor]. Since we do not have a model to 3 We employed a 7% discount rate to calculate the annuities in this paper. I'd like to see this neallys hammer 2 points: (1) Firms will do what is profitable negardless of you if programs; (2) Assumed hurdle rate for industry is exagenously reduced from account for this potential early retirement and the economic losses, we must caveat our estimates of investment and net costs. the investment cost may be understated by the amount of loss due to any early retirement that may occur" (p. 4.15). ? The mission of costs repled to early reternement of capital implies that This assumption implies thatcone of three results should be accounted for in the analysis First. will either be the projected emissions reduction should be revised downward, because the less assessment technology of the adepted industry sector overestimates technology adoption by ignoring these costs. Second, the cost of because with some technology adoption adopted decisions occurring at positive net costs' Third, if the authors reducing emissions will the reduced be carbon from this sector remains constant, should increase, benevised up maintain the same costs and emissions estimates, then the costs of government programs to ? This somehow force down the CRF from 33% to 15% would increase. locsn't flow The Transportation Sector Requires "Luck" to Achieve Carbon Reductions logically "[B]ecause the [transportation sector] outcomes postulated in the high efficiency/low carbon scenario require technological breakthroughs, they require a certain degree of luck to be achieved by 2010. There are no credible methods to accurately gauge the probability of such breakthroughs - we believe they stand a decent chance of occurring with an intensification of research efforts, but we stop short of claiming they are a likely outcome of such an intensification" (p. 5.3). Note that this statement includes two qualifications: luck on top of intensified research efforts. To achieve just the efficiency level of emission reductions for this sector (73 MMTCE), these intensified research efforts may require two to ten times existing funding on transportation (p. For 5,3). To gain a sense of the magnitude of what such an increase in funding might be, epnsider perspes what this might worth keep that the Partnership for and a New There Generation are of substatial Vehicles (PNGV) alone is funded at the federal level at $263 million, Private sector costs, of participating in the PNGV are in addition to this It The "luck" large our Further, the estimates of costless (on net) carbon reductions relies on a series of tenuous assumptions: commercial development of the fuel cell for passenger cars, commercial apparently availability of cellulosic ethanol (and the elimination of the ethanol excise tax exemption), and pertains an apparently arbitrary 30% reduction in costs for "certain key technologies" (p. 5.24). These to a technologies are not identified, and the report does not provide documentation for the NEMS jumber model runs conducted for the transportation sector. of tenvous technologicals Interactions Ignored, Double Counting "Likely" in Utilities Sector Analysis "Chapter 7 considers other electricity supply technology options, including: converting coal-based power plants to natural gas; cofiring coal with biomass; efficiency improvements in generation and transmission and distribution systems; extending the life of existing nuclear plants; increasing generation and capacity of existing hydropower plants; and constructing new powerplants using advanced coal technologies. Each of these options is assessed independently. Thus, 6 interactions between the options are not taken into account, and the possibility of double counting is therefore likely" (p. 7.1). The emissions reductions in the electricity sector come from two sources. First, utilities undertake fuel switching at the $50/ton permit price through dispatching Second, utilities modify their capital stock through several technology options such as those listed in the above quotation. As electricity production moves away from coal in the first analysis, fewer and fewer plants (and therefore potential emissions reductions) will be available what for conversion to natural gas or cofiring with biomass. The second set of analyses were "static" ufirst" and did not optimize unit/plant production cost, dispatch, or system load. Conversations with analysis? one of the co-authors (Stanley Hadley of Oak Ridge National Laboratory) indicate that 10 GW of coal-to-gas conversions (approximately 9 million tons of carbon reductions) were double- counted.⁴ express as a % ofsomething Furthermore, the analysis of specific options did not examine the effect of increasing fuel prices (from gas demand due to coal-to-gas repowering). Given the range of assumptions considered in the study, the authors actually estimated that repowering coal plants for natural gas could result in carbon reductions between 5 million and 269 million tons (p. 7.2), depending on the gas/coal price differential, the cost of carbon, and the costs of sulfur dioxide and nitrous oxide emissions. This is quite a substantial range. The Labs Study estimates that natural gas consumption will increase 14% to 191% above 2010 baseline consumption. However, the DRI 1.25 run in the IAT report indicates a 16% decline in natural gas consumption, while the SGM run resulted in-eady a negligible decline and Markal-Macro generated a 6% increase in natural gas consumption. These three models indicate that a non-integrated analysis that does not account for the price change in natural gas may overestimate natural gas substitution for coal and emissions reductions from fuel switching. message is a bit murky. what point? 4 Hadley noted that a subsequent draft of the report should address at least some of the double counting, but he did not provide details. 7 Auu-11-1yy 14.14 LCH P.009/011 Summary of Greenhouse Gas Emissions-Reduction Actions Million Metric Tons of Carbon Equivalent Source: U.S. Climate Action Report-1997 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date Residential & Commercial Sector Actions 26.9 10.3 - New Rebuild America 2.0 1.6 - I and 2 Expanded Green Lights and Energy 3.6 3.3 - Star Buildings 3 State Revolving Fund for Public 1.1 Terminated Buildings 4 Cost-Shared Demonstrations of Emerging Technologies 5 Operation and Maintenance Training 3.8 0.0 -- for Commercial Building Facility Managers and Operators 6 Energy Star Products 5.0 4.3 - 7 Residential Appliance Standards 6.8 0.2 - s and 11 Energy Partnerships for Affordable Housing 9 Cool Communities 4.4 0.4 - 10 Update State Building Codes New Construction of Energy-Efficient 0.1 - Commercial and Industrial Buildings New Superwindow Collaborative 0.0 -- New Expand Markets for Next-Generation 0.2 - Lighting Products New Fuel Cells Initiative 0.0 : Industrial Sector Actions 19.0 4.8 - 12 Motor Challenge 8.8 1.8 - 13 Industrial Golden Carrot Programs 2.9 Merged into Action 12 14 Accelerate the Adoption of Energy- Terminated Efficient Process Technologies 15 Industrial Assessment Centers 0.5 CCAP Component Terminated 16 Waste Minimization. 42 2.1 - 17 Improve Efficiency of Fertilizer 2.7 0.8 -- Nitrogen Use 18 Reduce the Use of Pesticides Terminated Transportation Sector Actions 8.1 5.3 - 19 Cash Value of Parking 20 Innovative Transportation Strategies 6.6 4.6 -- 21 Telecommuting Program 22 Fuel Economy Labels for Tires 1.5 0.7 -- Energy Supply Actions 10.8 1.3 - 23 Increase Natural Gas Share of Energy Use Though Federal Regulatory 2.2 Terminated Reform HOG-11-1997 14:15 LEH P.010/011 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date 24 Promote Seasonal Gas Use for Control 2.8 0.5 - of Nitrogen Oxides 25 High-Efficiency Gas Technologies 0.6 Terminated 26 Renewable-Energy Commercialization 0.8 0.3 - 27 Expand Utility Integrated Resource 1.4 Terminated Planning 28 Profitable Hydroelectric Efficiency 2.0 0.0 -- Upgrades 29 Energy-Efficient Distribution Transformer Standards 0.8 0.5 I 30 Energy Star Distribution Transformers 31 Transmission Pricing Reform 0.8 Terminated New Green Power Network Not included 0.0 : Land-Use Change & Forestry Actions 10.0 2.4 - 43 Private Depletion of Nonindustrial 4.0 Terminated Private Forests 44 Accelerate Tree Planting in 0.5 0.4 - Nonindustrial Private Forests 16 Waste Minimization 4.2 2.0 -- 9 Expand Cool Communities 0.5 To be determined Methane Actions 16.3 15.5 - 32 Expand Natural Gas STAR 3.0 3.4 -- 33 Increase Stringency of Landfill Rule 4.2 6.3 -- 34 Landfill Methane Outreach Program 1.1 1.9 -- 35 Coalbed Methane Outreach Program 2.2 2.6 - 36 RD&D for Coal Mine Methane 1.5 Terminated 37 RD&D for Landfill Methane 1.0 Terminated 38 AgSTAR Program 1.5 0.3 -- 39 Ruminant Livestock Efficiency 1.8 1.0 -- Program Actions to Address Other Greenhouse Gases 16.3 25.4 - 17 Improved Fertilizer Management 4.5 53 -- 40 Significant New Alternatives Program 5.0 6.4 - 41 HFC-23 Partnerships 5.0 5.0 - 42 Voluntary Aluminum Partnership 1.8 22 -- New Environmental Stewardship Initiative Not included 6.5 -- Foundation Actions 11.3 - Climate Wise Not estimated 1.8 -- Climate Challenge Not estimated 7.6 -- State and Local Outreach Programs Not estimated 1.9 - Total GHG Emission Reductions From CCAP 108.6 76.0 14.0 Data is not readily available for cumulative emissions reductions for many CCAP programs. Emissions reductions of about 5 MMTCE can be attributed to DOE's CCAP programs. EPA's Office of Air and Radiation is responsible for emissions reductions of about 9 MMTCE through their CCAP programs. Costs of Reductions ($/ton) HOG-11-1997 14:15 TC = $45.2 billion; Annuity = $1.1billion 1,000 800 TC = $16.1 billion; Annuity = $375 million CEA 600 TC = $6.4 billion; Annuity = $148 million 400 200 0 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 130-140 140-150 150-160 160-170 170-180 Carbon Reductions (MMTC) MC = AC of CCAP MC Increases 10% Every 10 MMTC IIA/IIA'H MC Increases 20% Every 10 MMTC MEMORANDUM TO: Jay Shogren FROM: Joe Aldy and Quindi Franco DATE: August 7, 1997 RE: Review of DOE Labs Study, June 10 Draft We provide an assessment of U.S. Carbon Reductions by 2010 and Beyond: The Potential Impact of Energy-Efficient and Low-Carbon Technologies (DOE Labs Study, June 10 draft). The report identifies carbon reductions and their associated costs in the buildings, industry, transportation, and electric power sectors under two hypothetical climate policy scenarios. The study finds that, with "very aggressive policies" and a $50/ton permit price, carbon emissions can be stabilized at 1990 levels by 2010. In reviewing the report we believe it is beneficial to highlight several of the caveats the authors explicitly state. First the authors do not provide enfitront descriptions of the government programs necessary to stimulate the necessary technology adoption to reduce 180 million metric tons of carbon emissions. We devised an estimate of the costs of 3 government programs to achieve these reductions based on the nation's experience with CCAP, and found that program costs could total $45 billion (1995$). Second, the analyses across sectors are not integrated. Therefore, the penetration rate of energy efficient technologies may be overestimated because the separate sectoral analyses do not account for cross-sector interactions (e.g., in energy prices). Third, the industry sector analysis ignores the costs of accelerating capital retirement, resulting in either an overestimate of reductions or an underestimate of costs. Fourth, the transportation sector requires "luck" in terms of technological innovation to achieve the projected emissions reductions. Fifth, the utilities sector analysis resulted in double counting of emissions reductions. These examples of the qualifications can provide additional insight into the projections of emissions reductions and costs. Upon receiving the final version of the report and documentation of the analyses, we will conduct a more thorough review. Overview of Report The Labs Study estimated the carbon emissions reduction opportunities available through technology development and adoption and fuel switching. The Study employs the Energy Information Administration's 1997 baseline (AEO97) for the buildings and industry sectors. For the transportation sector, the authors changed the AEO97 assumption of increasing fuel efficiency in automobiles to constant fuel efficiency. For the utility sector, the baseline was modified to reflect a fully competitive bulk-power market in the year 2010. Discharge, shutdown, repowering, and new construction decisions were optimized to select the amount of capacity that minimized the cost of the power-supply system plus the cost of unserved energy. Compared to AEO97, the final baseline has slightly lower energy prices, larger electricity sales, 1 neither and a greater share of gas generation. The study provides emissions reductions from this baseline for two scenarios: 1) "efficiency" and 2)"high efficiency/low carbon". The study does not describe the policies necessary to achieve emissions reductions nor the costs of federal programs for these scenarios, only their outcomes: increased adoption rates and better technologies. Efficiency This case assumes that all technologies adopted are cost-effective given more aggressive federal policies to stimulate development and diffusion of energy efficient technologies. Since these technologies are cost-effective, they all have a net cost less than or equal to zero. The Study estimates that 120 will be reduced at negative net cost in this scenario. None of these reductions occur in the utility sector as a function of fuel switching or technology adoption to increase combustion efficiency. High Efficiency/Low Carbon reductions of of This case assumes a "greater commitment" to reduce emissions through federal policies, in concert with state and private activities and a $50/ton permit price that stimulates low carbon technologies (read: primarily fuel switching). The policy is announced in 2000 and the policy's restrictions are phased-in through 2010. There is an assumed announcement effect (although an AEEI is not specified), increased domestic and international R&D in low-carbon technologies, and a "change in psychology". The Study estimates/ 180 hithics of carbon emissions reductions from the baseline through energy efficient technology adoption at negative net cost (see high efficiency without low carbon technology in chart below). The carbon permits will yield an additional 150-200 MATGE of reductions through low-carbon technology adoption. Given that these occur at a permit price of $50/ton, the Study estimates the upper limits of these costs to be $10 billion per year in 2010. Carbon Emissions (hintse): MMD 1990 - 2010 1990 2010 2010 2010 High Efficiency 2010 High Baseline Efficiency (w/o Low Carbon Tech) Efficiency/Low Carbon 1340 1720 1600 1540 1340 - 1390 (20 180 Review The authors of the Lab Study are forthright about many of the assumptions underlying the analysis. Learning about these qualifications can enhance one's understanding of the estimates regarding cost-effectiveness and emissions reductions in this study. We present several of the most significant assumptions and discussions of their impact below. The Study Does Not Specify Policies to Achieve Carbon Reductions 2 "Cost effectiveness is improved because R&D, in combination with increased deployment efforts, result in declining capital costs. We do not specify the policies, economic conditions, or exogenous events that could precipitate such changes" (p. 1.4). The entire study is premised on two sets of undefined policies: an "aggressive" or "invigorated" public sector effort to stimulate energy efficient technology adoption (efficiency case) and a "very aggressive" public sector effort to stimulate technology development and adoption (high efficiency/low carbon case). The only specific policy instrument mentioned in the report is a $50/ton tradeable permit. The study makes vague references to policies and only provides details on what these unidentified policies would result in: 1) "better technology" (p. 1.5); 2) "higher penetration rates" (p. 1.6); 3) "changing the capital recovery factor [in industry sector] from 33% to 15%" (p. 4.8); and 4) "technological breakthroughs" (p. 5.3). The authors assume that policies are implemented that significantly lower the barriers to technology adoption without any discussion of their costs. In effect, this assumes the magical wand of government intervention is waved to lower technology marginal costs enough to get 180 million tons of carbon reductions for "free": firms and individuals receive energy cost-savings in excess of the technology adoption and implementation costs. CEA AlternotiveCostErmates Since the authors do not provide an estimate of the costs to government to lower market barriers to achieve these substantial reductions, we calculated estimates of the costs of emissions reductions programs based on the government's experience with the Climate Change Action Plan (CCAP). CCAP promotes carbon reductions through a broad array of voluntary programs that stimulate "cost-effective" technology adoption by private firms. Participation by a firm in a CCAP program implies that carbon is reduced and the firm gains financially.¹ CCAP received appropriations totaling $494 million (1995$) during the FY95 - FY97 period. During this period, the Department of Energy and the Environmental Protection Agency can account for 14 million of metric tons of reduced earbon (see attached table). Each ton of carbon reduced in CCAP cost the 14 MMICE federal MUTTE government $35.29 (1995$). Assuming a flat marginal cost curve (MC=AC) to reduce can he 180 million metric tons of earbon beyond the baseline, the government costs of the unidentified expected policies in the I abs Study would exceed $6.3 billion. However, the marginal costs of technology prequied to EMISSAS requestions adoption are definitely higher than the average cost of the first 14 million metric tons, especially to be as more and more efforts to reduce emissions are undertaken. Assuming that costs increase 10% at for every 10 million metric tons of carbon reduced, then the government cost would come to least as $16.1 billion. This is slowly increasing marginal cost curve given baseline assumptions of gains as in energy efficiency. Assuming that costs increase 20% for every 10 million metric tons of high carbon reduced, then the cost to government would exceed $45.2 billion. The Sectoral Analyses Are Not Integrated 1 Note that CCAP efforts are incorporated in AEO97. 3 "The model runs for each of the three end-use sectors were not integrated and therefore may overstate the effects of technology penetration. In an integrated modeling effort, fuel prices might fall as consumption declines, resulting in less penetration of energy-conserving technologies" (p. 1.1). The analyses of the various sectors assume that 180 million tons of carbon reduction occurs through energy efficient technology adoption without the price effect of a $50/ton permit. This implies that, with government programs, private agents will adopt technologies because the benefits (energy cost-savings) exceed the costs of adoption. The authors claim that these reductions occur at negative net costs. By failing to account for the effect of decreased energy demand due to energy efficiency technology adoption, some of those adoption decisions that occur on the margin would no longer generate positive net benefits for private consumers. The Labs Study indicates that energy consumption will decline from the baseline by 11.6% under the high efficiency/low carbon case (p. xi). This decrease in demand should result in a decrease in the price of energy, and cause some energy-intensive technologies to become unattractive to private agents. In addition, the absence of an integrated analysis precludes an assessment of the economy-wide effects of a tradeable permit system. As previous studies have indicated, the nature of the permit allocation (e.g., grandfathering, auction, or a hybrid) and the characteristics of the revenue recycling (e.g., various adjustments to existing taxes) can affect economy-wide investment. Understanding the effects on investment is instrumental in assessing the likelihood of success of these scenarios, given that they rely on substantial R&D and technology adoption. The Industry Sector Analysis Ignores the Costs of Accelerated Retirement "When the economic losses of accelerated retirement are accounted for this means that, at the margin, all investments are not likely to be cost effective at our assumed 15% CRF [capital recovery factor]. Since we do not have a model to account for this potential early retirement and the economic losses, we must caveat our estimates of investment and net costs. the investment cost may be understated by the amount of loss due to any early retirement that may occur" (p. 4.15). This assumption implies that one of three results should be accounted for in the analysis. First, the projected emissions reduction should be revised downward, because the assessment of the industry sector overestimates technology adoption by ignoring these costs. Second, the cost of reducing emissions, if the reduced carbon from this sector remains constant, should increase, with some technology adoption decisions occurring at positive net costs. Third, if the authors maintain the same costs and emissions estimates, then the costs of government programs to somehow force down the CRF from 33% to 15% would increase. The Transportation Sector Requires "Luck" to Achieve Carbon Reductions 4 "[B]ecause the [transportation sector] outcomes postulated in the high efficiency/low carbon scenario require technological breakthroughs, they require a certain degree of luck to be achieved by 2010. There are no credible methods to accurately gauge the probability of such breakthroughs -- we believe they stand a decent chance of occurring with an intensification of research efforts, but we stop short of claiming they are a likely outcome of such an intensification" (p. 5.3). Note that this statement includes two qualifications: luck on top of intensified research efforts. To achieve just the efficiency level of emissions reductions for this sector (73 million MMTCO tons), these intensified research efforts may require two to ten times existing funding on transportation (p. 5.3). To gain a sense of the magnitude of what such an increase in funding might be, consider that the Partnership for a New Generation of Vehicles alone is funded at the federal level at $263 million. Further, the estimates of costless (on net) carbon reductions relies on a series of tenuous assumptions: commercial development of the fuel cell for passenger cars, commercial availability of cellulosic ethanol (and the elimination of the ethanol excise tax exemption), and an apparently arbitrary 30% reduction in costs for "certain key technologies" (p. 5.24). These technologies are not identified, and the report does not provide documentation for the NEMS model runs conducted for the transportation sector. Interactions Ignored, Double Counting "Likely" in Utilities Sector Analysis "Chapter 7 considers other electricity supply technology options, including: converting coal-based power plants to natural gas; cofiring coal with biomass; efficiency improvements in generation and transmission and distribution systems; extending the life of existing nuclear plants; increasing generation and capacity of existing hydropower plants; and constructing new powerplants using advanced coal technologies. Each of these options is assessed independently. Thus, interactions between the options are not taken into account, and the possibility of double counting is therefore likely" (p. 7.1). The emissions reductions in the electricity sector come from two sources. First, utilities undertake fuel switching at the $50/ton permit price through carbon-ordered dispatching. Second, utilities modify their capital stock through several technology options such as those listed in the above quotation. As electricity production moves away from coal in the first analysis, fewer and fewer plants (and therefore potential emissions reductions) will be available for conversion to natural gas or cofiring with biomass. The second set of analyses were "static" and did not optimize unit/plant production cost, dispatch, or system load. Conversations with one of the co-authors (Stanley Hadley of Oak Ridge National Laboratory) indicates that 10 GW of coal-to-gas conversions (approximately 9 million tons of carbon reductions) were double- counted.² 2 Hadley noted that a subsequent draft of the report should address at least some of the double counting, but he did not provide details. 5 Furthermore, the analysis of specific options did not examine the effect of increasing fuel prices (from gas demand due to coal-to-gas repowering). Given the range of assumptions considered in the study, the authors actually estimated that repowering coal plants for natural gas could result in carbon reductions between 5 million and 269 million tons (p. 7.2), depending on the gas/coal price differential, the cost of carbon, and the costs of sulfur dioxide and nitrous oxide emissions. This is quite a substantial range. The Labs Study estimates that natural gas consumption will increase 14% to 191% above 2010 baseline consumption. However, the DRI 1.25 run in the IAT report indicates a 16% decline in natural gas consumption, while the SGM run resulted in only a negligible decline and Markal-Macro generated a 6% increase in natural gas consumption. These three models indicate that a non-integrated analysis that does not account for the price change in natural gas may overestimate natural gas substitution for coal. 6 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date 24 Promote Seasonal Gas Use for Control 2.8 0.5 -- of Nitrogen Oxides 25 High-Efficiency Gas Technologies 0.6 Terminated 26 Renewable-Energy Commercialization 0.8 0.3 - 27 Expand Utility Integrated Resource 1.4 Terminated Planning 28 Profitable Hydroelectric Efficiency 2.0 0.0 -- Upgrades 29 Energy-Efficient Distribution Transformer Standards 0.8 0.5 -- 30 Energy Star Distribution Transformers 31 Transmission Pricing Reform 0.8 Terminated New Green Power Network Not included 0.0 -- Land-Use Change & Forestry Actions 10.0 2.4 - 43 Private Depletion of Nonindustrial 4.0 Terminated Private Forests 44 Accelerate Tree Planting in 0.5 0.4 -- Nonindustrial Private Forests 16 Waste Minimization 4.2 2.0 -- 9 Expand Cool Communities 0.5 To be determined Methane Actions 16.3 15.5 - 32 Expand Natural Gas STAR 3.0 3.4 -- 33 Increase Stringency of Landfill Rule 4.2 6.3 -- 34 Landfill Methane Outreach Program 1.1 1.9 -- 35 Coalbed Methane Outreach Program 2.2 2.6 -- 36 RD&D for Coal Mine Methane 1.5 Terminated 37 RD&D for Landfill Methane 1.0 Terminated 38 AgSTAR Program 1.5 0.3 -- 39 Ruminant Livestock Efficiency 1.8 1.0 -- Program Actions to Address Other Greenhouse Gases 16.3 25.4 - 17 Improved Fertilizer Management 4.5 5.3 -- 40 Significant New Alternatives Program 5.0 6.4 -- 41 HFC-23 Partnerships 5.0 5.0 -- 42 Voluntary Aluminum Partnership 1.8 2.2 -- New Environmental Stewardship Initiative Not included 6.5 -- Foundation Actions 11.3 -- Climate Wise Not estimated 1.8 -- Climate Challenge Not estimated 7.6 -- State and Local Outreach Programs Not estimated 1.9 -- Total GHG Emission Reductions From CCAP 108.6 76.0 14.0 Data is not readily available for cumulative emissions reductions for many CCAP programs. Emissions reductions of about 5 MMTCE can be attributed to DOE's CCAP programs. EPA's Office of Air and Radiation is responsible for emissions reductions of about 9 MMTCE through their CCAP programs. Summary of Greenhouse Gas Emissions-Reduction Actions Million Metric Tons of Carbon Equivalent Source: U.S. Climate Action Report-1997 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date Residential & Commercial Sector Actions 26.9 10.3 - New Rebuild America 2.0 1.6 : 1 and 2 Expanded Green Lights and Energy 3.6 3.3 -- Star Buildings 3 State Revolving Fund for Public 1.1 Terminated Buildings 4 Cost-Shared Demonstrations of Emerging Technologies 5 Operation and Maintenance Training 3.8 0.0 -- for Commercial Building Facility Managers and Operators 6 Energy Star Products 5.0 4.3 -- 7 Residential Appliance Standards 6.8 0.2 -- 8 and 11 Energy Partnerships for Affordable Housing 9 Cool Communities 4.4 0.4 -- 10 Update State Building Codes New Construction of Energy-Efficient 0.1 - Commercial and Industrial Buildings New Superwindow Collaborative 0.0 -- New Expand Markets for Next-Generation 0.2 -- Lighting Products New Fuel Cells Initiative 0.0 -- Industrial Sector Actions 19.0 4.8 - 12 Motor Challenge 8.8 1.8 -- 13 Industrial Golden Carrot Programs 2.9 Merged into Action 12 14 Accelerate the Adoption of Energy- Terminated Efficient Process Technologies 15 Industrial Assessment Centers 0.5 CCAP Component Terminated 16 Waste Minimization 4.2 2.1 -- 17 Improve Efficiency of Fertilizer 2.7 0.8 -- Nitrogen Use 18 Reduce the Use of Pesticides Terminated Transportation Sector Actions 8.1 5.3 - 19 Cash Value of Parking 20 Innovative Transportation Strategies 6.6 4.6 -- 21 Telecommuting Program 22 Fuel Economy Labels for Tires 1.5 0.7 -- Energy Supply Actions 10.8 1.3 - 23 Increase Natural Gas Share of Energy Use Though Federal Regulatory 2.2 Terminated Reform for mont in effizing cests list ignored /madmawledged costs policies implementation MEMORANDUM mht acceptance TO: Jay Shogren FROM: Joe Aldy and Quindi Franco 3 07 DATE: August 7, 1997 RE: Review of DOE Labs Study, June 10 Draft We provide an assessment of U.S. Carbon Reductions by 2010 and Beyond: The Potential Impact of Energy-Efficient and Low-Carbon Technologies (DOE Labs Study, June 10 draft). The report identifies carbon reductions and their associated costs in the buildings, industry, transportation, and electric power sectors under two hypothetical climate policy scenarios. The be more study finds that, with "very aggressive policies" and a $50/ton permit price, carbon emissions can explait be stabilized at 1990 levels by 2010. In reviewing the report, we believe it is beneficial to e what highlight several of the caveats the authors explicitly state. First, the authors do not provide they us? say in descriptions of the government programs necessary to stimulate the necessary technology BALST, release adoption to reduce 180 million tons of carbon emissions. We devised an estimate of the costs of government programs to achieve these reductions based on the nation's experience with CCAP, and found that program costs could total $45 billion (1995$). Second, the analyses across sectors are not integrated. Therefore, the penetration rate of energy efficient technologies may be overestimated because the separate sectoral analyses do not account for cross-sector interactions (e.g., in energy prices). Third, the industry sector analysis ignores the costs of accelerating capital retirement, resulting in either an overestimate of reductions or an underestimate of costs. Fourth, the transportation sector requires "luck" in terms of technological innovation to achieve the projected emissions reductions. Fifth, the utilities sector analysis resulted in double counting of emissions reductions. These examples of the qualifications can provide additional insight into the projections of emissions reductions and costs. Upon receiving the final version of the report and documentation of the analyses, we will conduct a more thorough review. Overview of Report The Labs Study estimated the carbon emissions reduction opportunities available through technology development and adoption and fuel switching. The Study employs the Energy Information Administration's 1997 baseline (AEO97) for the buildings and industry sectors. For the transportation sector, the authors changed the AEO97 assumption of increasing fuel efficiency in automobiles to constant fuel efficiency. For the utility sector, the baseline was modified to reflect a fully competitive bulk-power market in the year 2010. Discharge, shutdown, repowering, and new construction decisions were optimized to select the amount of capacity that minimized the cost of the power-supply system plus the cost of unserved energy. Compared to AEO97, the final baseline has slightly lower energy prices, larger electricity sales, 1 and a greater share of gas generation. The study provides emissions reductions from this baseline for two scenarios: 1) "efficiency" and 2) "high efficiency/low carbon". The study does not describe the policies necessary to achieve emissions reductions nor the costs of federal programs for these scenarios, only their outcomes: increased adoption rates and better technologies. Efficiency This case assumes that all technologies adopted are cost-effective given more aggressive federal policies to stimulate development and diffusion of energy efficient technologies. Since these technologies are cost-effective, they all have a net cost less than or equal to zero. The Study estimates that 120 mmtce will be reduced at negative net cost in this scenario. None of these reductions occur in the utility sector as a function of fuel switching or technology adoption to increase combustion efficiency. High Efficiency/Low Carbon This case assumes a "greater commitment" to reduce emissions through federal policies, in concert with state and private activities and a $50/ton permit price that stimulates low carbon technologies (read: primarily fuel switching). The policy is announced in 2000 and the policy's restrictions are phased-in through 2010. There is an assumed announcement effect (although an AEEI is not specified), increased domestic and international R&D in low-carbon technologies, and a "change in psychology". The Study estimates 180 mmtce of carbon emissions reductions from the baseline through energy efficient technology adoption at negative net cost (see high efficiency without low carbon technology in chart below). The carbon permits will yield an additional 150-200 mmtce of reductions through low-carbon technology adoption. Given that these occur at a permit price of $50/ton, the Study estimates the upper limits of these costs to be $10 billion per year in 2010. Carbon Emissions (mmtce): 1990 - 2010 1990 2010 2010 2010 High Efficiency 2010 High Baseline Efficiency (w/o Low Carbon Tech) Efficiency/Low Carbon 1340 1720 1600 1540 1340 - 1390 Review The authors of the Lab Study are forthright about many of the assumptions underlying the analysis. Learning about these qualifications can enhance one's understanding of the estimates regarding cost-effectiveness and emissions reductions in this study. We present several of the most significant assumptions and discussions of their impact below. The Study Does Not Specify Policies to Achieve Carbon Reductions 2 "Cost effectiveness is improved because R&D, in combination with increased deployment efforts, result in declining capital costs. We do not specify the policies, economic conditions, or exogenous events that could precipitate such changes" (p. 1.4). The entire study is premised on two sets of undefined policies: an "aggressive" or "invigorated" public sector effort to stimulate energy efficient technology adoption (efficiency case) and a "very aggressive" public sector effort to stimulate technology development and adoption (high efficiency/low carbon case). The only specific policy instrument mentioned in the report is a $50/ton tradeable permit. The study makes vague references to policies and only provides details on what these unidentified policies would result in: 1) "better technology" (p. 1.5); 2) "higher penetration rates" (p. 1.6); 3) "changing the capital recovery factor [in industry sector] from 33% to 15%" (p. 4.8); and 4) "technological breakthroughs" (p. 5.3). The authors assume that policies are implemented that significantly lower the barriers to technology adoption without any discussion of their costs. In effect, this assumes the magical wand of government intervention is waved to lower technology marginal costs enough to get 180 million tons of carbon reductions for "free": firms and individuals receive energy cost-savings in excess of the technology adoption and implementation costs. Since the authors do not provide an estimate of the costs to government to lower market barriers to achieve these substantial reductions, we calculated estimates of the costs of emissions reductions programs based on the government's experience with the Climate Change Action Plan (CCAP). CCAP promotes carbon reductions through a broad array of voluntary programs that stimulate "cost-effective" technology adoption by private firms. Participation by a firm in a CCAP program implies that carbon is reduced and the firm gains financially.¹ CCAP received appropriations totaling $494 million (1995$) during the FY95 - FY97 period. During this period, the Department of Energy and the Environmental Protection Agency can account for 14 million metric tons of reduced carbon (see attached table). Each ton of carbon reduced in CCAP cost the federal government $35.29 (1995$). Assuming a flat marginal cost curve (MC=AC) to reduce 180 million metric tons of carbon beyond the baseline, the government costs of the unidentified policies in the Labs Study would exceed $6.3 billion. However, the marginal costs of technology adoption are definitely higher than the average cost of the first 14 million metric tons, especially as more and more efforts to reduce emissions are undertaken. Assuming that costs increase 10% for every 10 million metric tons of carbon reduced, then the government cost would come to $16.1 billion. This is slowly increasing marginal cost curve given baseline assumptions of gains in energy efficiency. Assuming that costs increase 20% for every 10 million metric tons of carbon reduced, then the cost to government would exceed $45.2 billion. The Sectoral Analyses Are Not Integrated 1 Note that CCAP efforts are incorporated in AEO97. 3 "The model runs for each of the three end-use sectors were not integrated and therefore may overstate the effects of technology penetration. In an integrated modeling effort, fuel prices might fall as consumption declines, resulting in less penetration of energy-conserving technologies" (p. 1.1). The analyses of the various sectors assume that 180 million tons of carbon reduction occurs through energy efficient technology adoption without the price effect of a $50/ton permit. This implies that, with government programs, private agents will adopt technologies because the benefits (energy cost-savings) exceed the costs of adoption. The authors claim that these reductions occur at negative net costs. By failing to account for the effect of decreased energy demand due to energy efficiency technology adoption, some of those adoption decisions that occur on the margin would no longer generate positive net benefits for private consumers. The Labs Study indicates that energy consumption will decline from the baseline by 11.6% under the high efficiency/low carbon case (p. xi). This decrease in demand should result in a decrease in the price of energy, and cause some energy-intensive technologies to become unattractive to private agents. In addition, the absence of an integrated analysis precludes an assessment of the economy-wide effects of a tradeable permit system. As previous studies have indicated, the nature of the permit allocation (e.g., grandfathering, auction, or a hybrid) and the characteristics of the revenue recycling (e.g., various adjustments to existing taxes) can affect economy-wide investment. Understanding the effects on investment is instrumental in assessing the likelihood of success of these scenarios, given that they rely on substantial R&D and technology adoption. The Industry Sector Analysis Ignores the Costs of Accelerated Retirement "When the economic losses of accelerated retirement are accounted for this means that, at the margin, all investments are not likely to be cost effective at our assumed 15% CRF [capital recovery factor]. Since we do not have a model to account for this potential early retirement and the economic losses, we must caveat our estimates of investment and net costs. the investment cost may be understated by the amount of loss due to any early retirement that may occur" (p. 4.15). This assumption implies that one of three results should be accounted for in the analysis. First, the projected emissions reduction should be revised downward, because the assessment of the industry sector overestimates technology adoption by ignoring these costs. Second, the cost of reducing emissions, if the reduced carbon from this sector remains constant, should increase, with some technology adoption decisions occurring at positive net costs. Third, if the authors maintain the same costs and emissions estimates, then the costs of government programs to somehow force down the CRF from 33% to 15% would increase. The Transportation Sector Requires "Luck" to Achieve Carbon Reductions 4 "[B]ecause the [transportation sector] outcomes postulated in the high efficiency/low carbon scenario require technological breakthroughs, they require a certain degree of luck to be achieved by 2010. There are no credible methods to accurately gauge the probability of such breakthroughs -- we believe they stand a decent chance of occurring with an intensification of research efforts, but we stop short of claiming they are a likely outcome of such an intensification" (p. 5.3). Note that this statement includes two qualifications: luck on top of intensified research efforts. To achieve just the efficiency level of emissions reductions for this sector (73 million tons), these intensified research efforts may require two to ten times existing funding on transportation (p. 5.3). To gain a sense of the magnitude of what such an increase in funding might be, consider that the Partnership for a New Generation of Vehicles alone is funded at the federal level at $263 million. Further, the estimates of costless (on net) carbon reductions relies on a series of tenuous assumptions: commercial development of the fuel cell for passenger cars, commercial availability of cellulosic ethanol (and the elimination of the ethanol excise tax exemption), and an apparently arbitrary 30% reduction in costs for "certain key technologies" (p. 5.24). These technologies are not identified, and the report does not provide documentation for the NEMS model runs conducted for the transportation sector. Interactions Ignored, Double Counting "Likely" in Utilities Sector Analysis "Chapter 7 considers other electricity supply technology options, including: converting coal-based power plants to natural gas; cofiring coal with biomass; efficiency improvements in generation and transmission and distribution systems; extending the life of existing nuclear plants; increasing generation and capacity of existing hydropower plants; and constructing new powerplants using advanced coal technologies. Each of these options is assessed independently. Thus, interactions between the options are not taken into account, and the possibility of double counting is therefore likely" (p. 7.1). The emissions reductions in the electricity sector come from two sources. First, utilities undertake fuel switching at the $50/ton permit price through carbon-ordered dispatching. Second, utilities modify their capital stock through several technology options such as those listed in the above quotation. As electricity production moves away from coal in the first analysis, fewer and fewer plants (and therefore potential emissions reductions) will be available for conversion to natural gas or cofiring with biomass. The second set of analyses were "static" and did not optimize unit/plant production cost, dispatch, or system load. Conversations with one of the co-authors (Stanley Hadley of Oak Ridge National Laboratory) indicates that 10 GW of coal-to-gas conversions (approximately 9 million tons of carbon reductions) were double- counted.² 2 Hadley noted that a subsequent draft of the report should address at least some of the double counting, but he did not provide details. 5 Furthermore, the analysis of specific options did not examine the effect of increasing fuel prices (from gas demand due to coal-to-gas repowering). Given the range of assumptions considered in the study, the authors actually estimated that repowering coal plants for natural gas could result in carbon reductions between 5 million and 269 million tons (p. 7.2), depending on the gas/coal price differential, the cost of carbon, and the costs of sulfur dioxide and nitrous oxide emissions. This is quite a substantial range. The Labs Study estimates that natural gas consumption will increase 14% to 191% above 2010 baseline consumption. However, the DRI 1.25 run in the IAT report indicates a 16% decline in natural gas consumption, while the SGM run resulted in only a negligible decline and Markal-Macro generated a 6% increase in natural gas consumption. These three models indicate that a non-integrated analysis that does not account for the price change in natural gas may overestimate natural gas substitution for coal. 6 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date 24 Promote Seasonal Gas Use for Control 2.8 0.5 -- of Nitrogen Oxides 25 High-Efficiency Gas Technologies 0.6 Terminated 26 Renewable-Energy Commercialization 0.8 0.3 I 27 Expand Utility Integrated Resource 1.4 Terminated Planning 28 Profitable Hydroelectric Efficiency 2.0 0.0 -- Upgrades 29 Energy-Efficient Distribution Transformer Standards 0.8 0.5 -- 30 Energy Star Distribution Transformers 31 Transmission Pricing Reform 0.8 Terminated New Green Power Network Not included 0.0 : Land-Use Change & Forestry Actions 10.0 2.4 - 43 Private Depletion of Nonindustrial 4.0 Terminated Private Forests 44 Accelerate Tree Planting in 0.5 0.4 -- Nonindustrial Private Forests 16 Waste Minimization 4.2 2.0 -- 9 Expand Cool Communities 0.5 To be determined Methane Actions 16.3 15.5 - 32 Expand Natural Gas STAR 3.0 3.4 : 33 Increase Stringency of Landfill Rule 4.2 6.3 -- 34 Landfill Methane Outreach Program 1.1 1.9 -- 35 Coalbed Methane Outreach Program 2.2 2.6 -- 36 RD&D for Coal Mine Methane 1.5 Terminated 37 RD&D for Landfill Methane 1.0 Terminated 38 AgSTAR Program 1.5 0.3 -- 39 Ruminant Livestock Efficiency 1.8 1.0 -- Program Actions to Address Other Greenhouse Gases 16.3 25.4 - 17 Improved Fertilizer Management 4.5 5.3 : 40 Significant New Alternatives Program 5.0 6.4 -- 41 HFC-23 Partnerships 5.0 5.0 I 42 Voluntary Aluminum Partnership 1.8 2.2 -- New Environmental Stewardship Initiative Not included 6.5 I Foundation Actions 11.3 - Climate Wise Not estimated 1.8 : Climate Challenge Not estimated 7.6 -- State and Local Outreach Programs Not estimated 1.9 : Total GHG Emission Reductions From CCAP 108.6 76.0 14.0 Data is not readily available for cumulative emissions reductions for many CCAP programs. Emissions reductions of about 5 MMTCE can be attributed to DOE's CCAP programs. EPA's Office of Air and Radiation is responsible for emissions reductions of about 9 MMTCE through their CCAP programs. Summary of Greenhouse Gas Emissions-Reduction Actions Million Metric Tons of Carbon Equivalent Source: U.S. Climate Action Report-1997 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date Residential & Commercial Sector Actions 26.9 10.3 -- New Rebuild America 2.0 1.6 -- 1 and 2 Expanded Green Lights and Energy 3.6 3.3 -- Star Buildings 3 State Revolving Fund for Public 1.1 Terminated Buildings 4 Cost-Shared Demonstrations of Emerging Technologies 5 Operation and Maintenance Training 3.8 0.0 -- for Commercial Building Facility Managers and Operators 6 Energy Star Products 5.0 4.3 -- 7 Residential Appliance Standards 6.8 0.2 : 8 and 11 Energy Partnerships for Affordable Housing 9 Cool Communities 4.4 0.4 -- 10 Update State Building Codes New Construction of Energy-Efficient 0.1 -- Commercial and Industrial Buildings New Superwindow Collaborative 0.0 : New Expand Markets for Next-Generation 0.2 -- Lighting Products New Fuel Cells Initiative 0.0 : Industrial Sector Actions 19.0 4.8 - 12 Motor Challenge 8.8 1.8 -- 13 Industrial Golden Carrot Programs 2.9 Merged into Action 12 14 Accelerate the Adoption of Energy- Terminated Efficient Process Technologies 15 Industrial Assessment Centers 0.5 CCAP Component Terminated 16 Waste Minimization 4.2 2.1 -- 17 Improve Efficiency of Fertilizer 2.7 0.8 -- Nitrogen Use 18 Reduce the Use of Pesticides Terminated Transportation Sector Actions 8.1 5.3 - 19 Cash Value of Parking 20 Innovative Transportation Strategies 6.6 4.6 -- 21 Telecommuting Program 22 Fuel Economy Labels for Tires 1.5 0.7 -- Energy Supply Actions 10.8 1.3 -- 23 Increase Natural Gas Share of Energy Use Though Federal Regulatory 2.2 Terminated Reform BAsed an the actual history of MEMORANDUM CCAP, we estimate TO: Jay Shogren FROM: Joe Aldy and Quindi Franco RANDI'S DATE: August 7, 1997 RE: Review of DOE Labs Study, June 10 Draft evil United eye We provide an assessment of U.S. Carbon Reductions by 2010 and Beyond: The Potential Impact of Energy-Efficient and Low-Carbon Technologies (DOE Labs Study, June 10 draft). The report identifies carbon reductions and their associated costs in the buildings, industry, transportation, and electric power sectors under two hypotherical climate policy scenarios. The study finds that, with "very aggressive policies" and a $50/ton permit price, carbon emissions can be stabilized at 1990 levels by 2010. In reviewing the report, we believe it is beneficial to highlight several of the caveats the authors explicitly state. First the authors do not provide descriptions of the government programs necessary to stimulate the necessary technology should. adoption to reduce 180 million tons of carbon emissions, We devised an estimate of the costs of government programs to achieve these reductions based on the nation's experience with CCAP, The be. Add and found that program costs could total $45 billion (1995$). Second, the analyses across sectors we d to are not integrated. Therefore, the penetration rate of energy efficient technologies may be fiel pice $ton Tech, overestimated because the separate sectoral analyses do not account for cross sector interactions (e.g., in energy prices). Third, the industry sector analysis ignores the costs of accelerating maigins costs capital retirement, resulting in either an overestimate of reductions or an underestimate of costs. ravg. Fourth, the transportation sector requires "luck" in terms of technological innovation to achieve the projected emissions reductions. Fifth, the utilities sector analysis resulted in double counting of emissions reductions. These examples of the qualifications can provide additional insight into source the projections of emissions reductions and costs. Upon receiving the final version of the report and documentation of the analyses, we will conduct a more thorough review. update Overview of Report The Labs Study estimated the carbon emissions reduction opportunities available through technology development and adoption and fuel switching. The Study employs the Energy Information Administration's 1997 baseline (AEO97) for the buildings and industry sectors. For the transportation sector, the authors changed the AEO97 assumption of increasing fuel efficiency in automobiles to constant fuel efficiency. For the utility sector, the baseline was modified to reflect a fully competitive bulk-power market in the year 2010. Discharge, shutdown, repowering, and new construction decisions were optimized to select the amount of capacity that minimized the cost of the power-supply system plus the cost of unserved energy. Compared to AEO97, the final baseline has slightly lower energy prices, larger electricity sales, With changing flot mig 30% Town bigger sport recently vehicley bee and a greater share of gas generation. The study provides emissions reductions from this baseline for two scenarios: 1) "efficiency" and 2) 'high efficiency/low carbon". The study does not describe the policies necessary to achieve emissions reductions nor the costs of federal programs for these scenarios, only their outcomes: increased adoption rates and better technologies. Efficiency Seanario #25 ? zero net cest to the user This case assumes that all technologies adopted are cost given more aggressive federal policies to stimulate development and diffusion of energy efficient technologies. Since these technologies are cost effective, they all have a net cost less than or equal to zero. The Study estimates that 120 mmtce will be reduced at negative net cost in this scenario. None of these reductions occur in the utility sector as a function of fuel switching or technology adoption to increase combustion efficiency. & High Efficiency/Low Carbon @ $50/t This case assumes a "greater commitment to reduce emissions through federal policies, in concert with state and private activities and a $50/ton permit price that stimulates low carbon technologies (read: primarily fuel switching). The policy is announced in 2000 and the policy's restrictions are phased-in through 2010. There is an assumed announcement effect (although an AEEI is not specified), increased domestic and international at R&D in low-carbon technologies, and a "change in psychology The Study estimates 180 mmtce of carbon emissions reductions from the baseline through energy efficient technology adoption at negative net cost (see high efficiency without low carbon technology in chart below). The carbon permits will yield an additional 150-200 mmtce of reductions through low-carbon technology adoption. Given that these occur at a permit price of $50/ton, the Study estimates the upper limits of these costs to be $10 billion per year in 2010. Carbon Emissions (mmtce): 1990 - 2010 1990 2010 2010 2010 High Efficiency 2010 High Baseline Efficiency (w/o Low Carbon Tech) Efficiency/Low Carbon 1340 1720 1600 1540 1340 - 1390 Review The authors of the Lab Study are forthright about many of the assumptions underlying the analysis. Learning about these qualifications can enhance one's understanding of the estimates regarding cost-effectiveness and emissions reductions in this study. We present several of the most significant assumptions and discussions of their impact below. 2 1) Dear In The Study Does Not Specify Policies to Achieve Carbon Reductions "Cost effectiveness is improved because R&D, in combination withhncreased deployment efforts, result in declining capital costs. We do not specify the policies, economic conditions, or exogenous events that could precipitate such changes" (p. 1.4). calculation? The entire study is premised on two sets of undefined policies: an "aggressive" or "invigorated" public sector effort to stimulate energy efficient technology adoption (efficiency case) and a "very aggressive" public sector effort to stimulate technology development and adoption (high efficiency/low carbon case). The only specific policy instrument mentioned in the report is a $50/ton tradeable permit. The study makes vague references to policies and only provides details on what these unidentified policies would result in: 1) "better technology" (p. 1.5); 2) "higher penetration rates" (p. 1.6); 3) "changing the capital recovery factor [in industry sector] from 33% to 15%" (p. 4.8); and 4) "technological breakthroughs" (p. 5.3). The authors assume that policies are implemented that significantly lower the barriers to technology adoption without any discussion of their costs. In effect, this assumes the magical wand of government intervention is waved to lower technology marginal costs enough to get 180 million tons of carbon reductions for "free": firms and individuals receive energy cost-savings in excess of the technology adoption and implementation costs. Since the authors do not provide an estimate of the costs to government to lower market barriers to achieve these substantial reductions, we calculated estimates of the costs of emissions reductions programs based on the government's experience with the Climate Change Action Plan (CCAP). CCAP promotes carbon reductions through a broad array of voluntary programs that stimulate "cost-effective" technology adoption by private firms. Participation by a firm in a CCAP program implies that carbon is reduced and the firm gains financially.¹ CCAP received appropriations totaling $494 million (1995$) during the FY95 - FY97 period. During this period, the Department of Energy and the Environmental Protection Agency can account for 14 million metric tons of reduced carbon (see attached table). Each ton of carbon reduced in CCAP cost the federal government $35.29 (1995$) Assuming a flat marginal cost curve (MC=AC) to reduce 180 million metric tons of carbon beyond the baseline, the government costs of the unidentified policies in the Labs Study would exceed $6.3 billion. However, the marginal costs of technology adoption are definitely higher than the average cost of the first 14 million metric tons, especially as more and more efforts to reduce emissions are undertaken. Assuming that costs increase 10% for every 10 million metric tons of carbon reduced, then the government cost would come to $16.1 billion. This is slowly increasing marginal cost curve given baseline assumptions of gains good to in energy efficiency. Assuming that costs increase 20% for every 10 million metric tons of have carbon reduced, then the cost to government would exceed $45.2 billion. other 3/ evidence 1 Note that CCAP efforts are incorporated in AEO97. on how fast MC V 3 redise risue The Sectoral Analyses Are Not Integrated "The model runs for each of the three end-use sectors were not integrated and therefore may overstate the effects of technology penetration. In an integrated modeling effort, fuel prices might fall as consumption declines, resulting in less penetration of Frew energy conserving technologies" (p. 1.1). or The analyses of the various sectors assume that 180 million tons of carbon reduction occurs through energy efficient technology adoption without the price effect of a $50/ton permit. This implies that, with government programs, private agents will adopt technologies because the benefits (energy cost-savings) exceed the costs of adoption. The authors claim that these reductions occur at negative net costs. By failing to account for the effect of decreased energy demand due to energy efficiency technology adoption, some of those adoption decisions that occur on the margin would no longer generate positive net benefits for private consumers. The Labs Study indicates that energy consumption will decline from the baseline by 11.6% under the /high efficiency/low carbon case (p. xi). This decrease in demand should result in a decrease in the price of energy, and cause some energy-intensive technologies to become unattractive to private agents. Powehang ante off how yruch they In addition, the absence of an integrated analysis precludes an assessment of the economy-wide effects of a tradeable permit system. As previous studies have indicated, the nature of the permit allocation (e.g., grandfathering, auction, or a hybrid) and the characteristics of the revenue wer- predict recycling (e.g., various adjustments to existing taxes) can affect economy-wide investment. , J Understanding the effects on investment is instrumental in assessing the likelihood of success of these scenarios, given that they rely on substantial R&D and technology adoption. The Industry Sector Analysis Ignores the Costs of Accelerated Retirement "When the economic losses of accelerated retirement are accounted for this means that, at the margin, all investments are not likely to be cost effective at our assumed 15% CRF [capital recovery factor]. Since we do not have a model to account for this potential early retirement and the economic losses, we must caveat our estimates of investment and net costs. the investment cost may be understated by the amount of loss due to any early retirement that may occur" (p. 4.15). This assumption implies that one of three results should be accounted for in the analysis. First, the projected emissions reduction should be revised downward, because the assessment of the industry sector overestimates technology adoption by ignoring these costs. Second, the cost of reducing emissions, if the reduced carbon from this sector remains constant, should increase, with some technology adoption decisions occurring at positive net costs. Third, if the authors maintain the same costs and emissions estimates, then the costs of government programs to somehow force down the CRF from 33% to 15% would increase. 4 The Transportation Sector Requires "Luck" to Achieve Carbon Reductions are there "[B]ecause the [transportation sector] outcomes postulated in the high with efficiency/low carbon scenario require technological breakthroughs, they require a included certain degree of luck to be achieved by 2010. There are no credible methods to accurately gauge the probability of such breakthroughs -- we believe they stand a an your decent chance of occurring with an intensification of research efforts, but we stop short of claiming they are a likely outcome of such an intensification" (p. 5.3). 45 B ? Note that this statement includes two qualifications: luck on top of intensified research efforts. To achieve just the efficiency level of emissions reductions for this sector (73 million tons), these intensified research efforts may require two to ten times existing funding on transportation (p. 5.3). To gain a sense of the magnitude of what such an increase in funding might be, consider that the Partnership for a New Generation of Vehicles alone is funded at the federal level at $263 million. Further, the estimates of costless (on net) carbon reductions relies on a series of tenuous assumptions: commercial development of the fuel cell for passenger cars, commercial availability of cellulosic ethanol (and the elimination of the ethanol excise tax exemption), and an apparently privacy arbitrary 30% reduction in costs for "certain key technologies" (p. 5.24). These technologies are not identified, and the report does not provide documentation for the NEMS model runs conducted for the transportation sector. are in redation Interactions Ignored, Double Counting "Likely" in Utilities Sector Analysis "Chapter 7 considers other electricity supply technology options, including: converting coal-based power plants to natural gas; cofiring coal with biomass; efficiency improvements in generation and transmission and distribution systems; extending the life of existing nuclear plants; increasing generation and capacity of existing hydropower plants; and constructing new powerplants using advanced coal technologies. Each of these options is assessed independently. Thus, interactions between the options are not taken into account, and the possibility of double counting is therefore likely" (p. 7.1). The emissions reductions in the electricity sector come from two sources. First, utilities undertake fuel switching at the $50/ton permit price through carbon-ordered dispatching. Second, utilities modify their capital stock through several technology options such as those listed in the above quotation. As electricity production moves away from coal in the first analysis, fewer and fewer plants (and therefore potential emissions reductions) will be available for conversion to natural gas or cofiring with biomass. The second set of analyses were "static" and did not optimize unit/plant production cost, dispatch, or system load. Conversations with one of the co-authors (Stanley Hadley of Oak Ridge National Laboratory) indicates that 10 GW 5 of coal-to-gas conversions (approximately 9 million tons of carbon reductions) were double- counted.² Furthermore, the analysis of specific options did not examine the effect of increasing fuel prices (from gas demand due to coal-to-gas. repowering). Given the range of assumptions considered in the study, the authors actually estimated that repowering coal plants for natural gas could result in carbon reductions between 5 million and 269 million tons (p. 7.2), depending on the gas/coal price differential, the cost of carbon, and the costs of sulfur dioxide and nitrous oxide emissions. This is quite a substantial range. The Labs Study estimates that natural gas consumption will increase 14% to 191% above 2010 baseline consumption. However, the DRI 1.25 run in the IAT report indicates a 16% decline in natural gas consumption, while the SGM run resulted in only a negligible decline and Markal-Macro generated a 6% increase in natural gas consumption. These three models indicate that a non-integrated analysis that does not account for the price change in natural gas may overestimate natural gas substitution for coal. 2 Hadley noted that a subsequent draft of the report should address at least some of the double counting, but he did not provide details. 6 Summary of Greenhouse Gas Emissions-Reduction Actions Million Metric Tons of Carbon Equivalent Source: U.S. Climate Action Report-1997 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date Residential & Commercial Sector Actions 26.9 10.3 -- New Rebuild America 2.0 1.6 -- 1 and 2 Expanded Green Lights and Energy 3.6 3.3 -- Star Buildings 3 State Revolving Fund for Public 1.1 Terminated Buildings 4 Cost-Shared Demonstrations of Emerging Technologies 5 Operation and Maintenance Training 3.8 0.0 -- for Commercial Building Facility Managers and Operators 6 Energy Star Products 5.0 4.3 -- 7 Residential Appliance Standards 6.8 0.2 -- 8 and 11 Energy Partnerships for Affordable Housing 9 Cool Communities 4.4 0.4 - 10 Update State Building Codes New Construction of Energy-Efficient 0.1 -- Commercial and Industrial Buildings New Superwindow Collaborative 0.0 -- New Expand Markets for Next-Generation 0.2 Lighting Products New Fuel Cells Initiative 0.0 -- Industrial Sector Actions 19.0 4.8 - 12 Motor Challenge 8.8 1.8 -- 13 Industrial Golden Carrot Programs 2.9 Merged into Action 12 14 Accelerate the Adoption of Energy- Terminated Efficient Process Technologies 15 Industrial Assessment Centers 0.5 CCAP Component Terminated 16 Waste Minimization 4.2 2.1 -- 17 Improve Efficiency of Fertilizer 2.7 0.8 -- Nitrogen Use 18 Reduce the Use of Pesticides Terminated Transportation Sector Actions 8.1 5.3 -- 19 Cash Value of Parking 20 Innovative Transportation Strategies 6.6 4.6 -- 21 Telecommuting Program 22 Fuel Economy Labels for Tires 1.5 0.7 -- Energy Supply Actions 10.8 1.3 -- 23 Increase Natural Gas Share of Energy Use Though Federal Regulatory 2.2 Terminated Reform Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date 24 Promote Seasonal Gas Use for Control 2.8 0.5 -- of Nitrogen Oxides 25 High-Efficiency Gas Technologies 0.6 Terminated 26 Renewable-Energy Commercialization 0.8 0.3 I 27 Expand Utility Integrated Resource 1.4 Terminated Planning 28 Profitable Hydroelectric Efficiency 2.0 0.0 : Upgrades 29 Energy-Efficient Distribution Transformer Standards 0.8 0.5 -- 30 Energy Star Distribution Transformers 31 Transmission Pricing Reform 0.8 Terminated New Green Power Network Not included 0.0 -- Land-Use Change & Forestry Actions 10.0 2.4 - 43 Private Depletion of Nonindustrial 4.0 Terminated Private Forests 44 Accelerate Tree Planting in 0.5 0.4 : Nonindustrial Private Forests 16 Waste Minimization 4.2 2.0 -- 9 Expand Cool Communities 0.5 To be determined Methane Actions 16.3 15.5 - 32 Expand Natural Gas STAR 3.0 3.4 -- 33 Increase Stringency of Landfill Rule 4.2 6.3 -- 34 Landfill Methane Outreach Program 1.1 1.9 -- 35 Coalbed Methane Outreach Program 2.2 2.6 -- 36 RD&D for Coal Mine Methane 1.5 Terminated 37 RD&D for Landfill Methane 1.0 Terminated 38 AgSTAR Program 1.5 0.3 -- 39 Ruminant Livestock Efficiency 1.8 1.0 -- Program Actions to Address Other Greenhouse Gases 16.3 25.4 - 17 Improved Fertilizer Management 4.5 5.3 : 40 Significant New Alternatives Program 5.0 6.4 -- 41 HFC-23 Partnerships 5.0 5.0 -- 42 Voluntary Aluminum Partnership 1.8 2.2 : New Environmental Stewardship Initiative Not included 6.5 -- Foundation Actions 11.3 - Climate Wise Not estimated 1.8 -- Climate Challenge Not estimated 7.6 : State and Local Outreach Programs Not estimated 1.9 -- Total GHG Emission Reductions From CCAP 108.6 76.0 14.0 Data is not readily available for cumulative emissions reductions for many CCAP programs. Emissions reductions of about 5 MMTCE can be attributed to DOE's CCAP programs. EPA's Office of Air and Radiation is responsible for emissions reductions of about 9 MMTCE through their CCAP programs. 12/30/97 TUE 10:34 FAX 202 6222633 001 Office of Economic Policy Department of the Treasury Washington, D.C. 20220 FAX Date: Number of pages including cover sheet: 5 Dec.30, 1997 Name Fax Number Phone Number To: Jff Franked 395-6958 395-5046 From: Jon Gruber 202-622-2633 REMARKS: Urgent For your review Reply ASAP Please comment Hold Close 12/30/97 TUE 10:35 FAX 202 6222633 4. 002 Jeff- - Larry Summers suggested that I talk, on deep background, to a couple of outside experts to see what they would view as reasonable for our economic analysis. I attach (for your eyes only, obviously) some discussion of their views. I also attach a revised set of runs that I think Joe should do, reflecting these views. I haven't passed these on to Joe yet. My only question is: have I been too conservative on the possibilities for JI/LDC participation? Maybe we should do some runs where JI/LDCs give us more benefits than the 20% that I assume? I don't know enough about the structure of the model to assess this. Give me a call when you get in and lets discuss. Jon 12/30/97 TUE 10:35 FAX 202 6222633 003 Expert Opinion on Economic Analysis Contacts: Rob Stavins, Mike Toman, Dick Schmalansee Starting Point: Roughly $100/ton from SGM model (this is with domestic only, and 1% AEEI) All experts seem comfortable with this model and have great respect for Edmunds and Batelle. They seem to think that this is a sensible starting point. Stavins did note that this assumes that we equate marginal costs and pursue only least cost efficient paths to reduction. He points out that in reality this is unlikely. Schmalansee highlighted that timing is key - but the runs that CEA is using are calibrated for reductions to start in 2005, when our permit system kicks in. Annex I Trading: Pure Annex I trading can roughly halve costs to $50; the umbrella might get the price down to $30 There was general comfort with the numbers, but all highlighted that this was conditional on perfect trading being in place, and there was also general skepticism that this could emerge. Stavins once again highlighted that this assumes we find global least cost reductions, which seems even more suspect since it requires other countries to be efficient as well. On the other hand, all recognize the huge value of the Russian tons. There was some skepticism that the EU would, at the end of the day, let themselves get shut out of the Russian tons. Technology: The IAT used an AEEI of 1.25, and it was mildly criticized as a bit optomistic - but this was before the $6 billion in the budget, and without any price signal. 5 Labs implies an AEEI of 1.71. An AEEI of 1.25 would bring price down to roughly $80. I think that there would be support for 1.25 AEEI, but not for much more - certainly nothing approaching 5 labs. Schmalansce pointed out that, even without an explicit price signal, Kyoto highlights that "the train is moving" and should promote the growth of the energy efficient technologies. Joint Implementation and LDCs: Models tell us effect of Annex I trading, and worldwide trading. Question is: how much of the gap between these two is closed by JI? Would be little support for claiming anything close to the majority of this gap. But there would, I think, be support for claiming a small share, say 10-20% of the gap. Schamalansee said <10%, Toman said up to 20%. There was no real view on the prospects of additional LDC participation, but pretty strong skepticism that it would amount to much. ini, Electriticity Restructuring: We can claim economic benefits from restructuring. Can we claim some enviro benefits as well? Gruenspecht says that we can get electricity savings of 10% from restructuring. This is $20 billion now and ??? in 2010. Joskow says 8-10%, but will defer to Howard as well. By 2010, stranded costs are all paid, so that we get full benefits of lower prices. 12/30/97 TUE 10:35 FAX 202 6222633 4. 004 There was generally strong skepticism among this group about claims of environmental benefits. Toman even pointed out that Howard Gruenspecht had actually changed his position on this over the past year. I think that the best we can do here is claim environmental neutrality, and maybe sqres some qualitative statements about potential gains. putting numbers on this is going to get us in trouble. Other Benefits: There may be gains from reducing costs of meeting NAAQS. or health benefits (but not both). There was support for some benefits here. Toman thought that most of the action would be health benefits, since he didn't think that NAAQS would be effective by 2010. Ray notes that administration policy is that NAAQS will be effective by then, so that the benefits will bc in dollars, not health. Bottom Line: There seems to be professional support for claiming some technology gains, some small JI/LDC benefit, full Annex I trading gains, and perhaps for some gains from umbrella. But we have to recognize that this analysis will be criticized by some folks on the grounds that it is too idealistic on trading, and by others on the grounds that it implies that we buy our way out of the majority of our reductions. Both these criticisms will rise in proportion to the gains that we claim from umbrella. It seems like an analysis with 1.25 AEEI, Annex I trading including EU (no umbrella), and JI/LDC that gets 20% of WW trading benefits would be pretty well received. Moving to the umbrella would bring some criticism, but not an avalanche, Going beyond that on trading or technology would be steeply criticized. I remain unclear on how much we can justify on JI/LDC 12/30/97 TUE 10:36 FAX 202 6222633 4 005 Given these comments, here is a revised set of runs that I think we should do: Model Permutations to Consider: 1) Full Annex I trading 2) Umbrella - trading block that excludes EU and Eastern Europe 3) AEEI of 1.25 4) JI/LDC participation that gets us 20% of the gains of worldwide trading (beyond Annex I trading). Question: Are interactions linear? Worth trying some combined and separate cstimates to get a feel for this, although we don't want to get out of control. I would suggest the following set of runs: 1-3) Do the first three factors individually 4) Annex I trading with JI/LDC 5) Annex I trading, with umbrella, and JI/LDC 6) Higher AEEI, with Annex I trading 7) Higher AEEI, with Annex I trading and umbrella 8) Higher AEEI, with Annex I trading and JI/LDC 9) Higher AEEI, with Annex I trading, umbrella, and JI/LDC For each of these cases, we should show what happens to permit price, and what share of our reductions are purchased abroad. 12/30/97 TUE 10:34 FAI 202 6222633 J. 001 Office of Economic Policy Department of the Treasury Washington, D.C. 20220 FAX Date: Dec. Dec. 30, 1997 1997 Number of pages including cover sheet: 5 Name Fax Number Phone Number To: Jeff Frankee 395-6958 395-5046 From: Jon Gruber 202-622-2633 REMARKS: Urgent For your review Reply ASAP Please comment Hold Close 12/30/97 TUE 10:35 FAI 202 6222633 002 Jeff Larry Summers suggested that I talk, on deep background, to a couple of outside experts to see what they would view as reasonable for our economic analysis. I attach (for your eyes only, obviously) some discussion of their views. I also attach a revised set of runs that I think Joe should do, reflecting these views. I haven't passed these on to Joe yet. My only question is: have I been too conservative on the possibilities for JI/LDC participation? Maybe we should do some runs where JI/LDCs give us more benefits than the 20% that I assume? I don't know enough about the structure of the model to assess this. Give me a call when you get in and lets discuss. Jon 12/30/97 TUE 10:35 FAI 202 6222633 003 Expert Opinion on Economic Analysis Contacts: Rob Stavins, Mike Toman, Dick Schmalansee Starting Point: Roughly $100/ton from SGM model (this is with domestic only, and 1% AEEI) All experts seem comfortable with this model and have great respect for Edmunds and Batelle. They seem to think that this is a sensible starting point. Stavins did note that this assumes that we equate marginal costs and pursue only least cost efficient paths to reduction. He points out that in reality this is unlikely. Schmalansee highlighted that timing is key - but the runs that CEA is using are calibrated for reductions to start in 2005, when our permit system kicks in. Annex I Trading: Pure Annex I trading can roughly halve costs to $50; the umbrella might get the price down to $30 There was general comfort with the numbers, but all highlighted that this was conditional on perfect trading being in place, and there was also general skepticism that this could emerge. Stavins once again highlighted that this assumes we find global least cost reductions, which seems even more suspect since it requires other countries to be efficient as well. On the other hand, all recognize the huge value of the Russian tons. There was some skepticism that the EU would, at the end of the day, let themselves get shut out of the Russian tons. Technology: The IAT used an AEEI of 1.25, and it was mildly criticized as a bit optomistic - but this was before the $6 billion in the budget, and without any price signal. 5 Labs implies an AEEI of 1.71. An AEEI of 1.25 would bring price down to roughly $80. I think that there would be support for 1.25 AEEI, but not for much more - certainly nothing approaching 5 labs. Schmalansce pointed out that, even without an explicit price signal, Kyoto highlights that "the train is moving" and should promote the growth of the energy efficient technologies. Joint Implementation and LDCs: Models tell us effect of Annex I trading, and worldwide trading. Question is: how much of the gap between these two is closed by JI? Would be little support for claiming anything close to the majority of this gap. But there would, I think, be support for claiming a small share, say 10-20% of the gap. Schamalansee said <10%, Toman said up to 20%. There was no real view on the prospects of additional LDC participation, but pretty strong skepticism that it would amount to much. Electriticity Restructuring: We can claim economic benefits from restructuring. Can we claim some enviro benefits as well? Gruenspecht says that we can get electricity savings of 10% from restructuring. This is $20 billion now and ??? in 2010. Joskow says 8-10%, but will defer to Howard as well. By 2010, stranded costs are all paid, so that we get full benefits of lower prices. 12/30/97 TUE 10:35 FAI 202 6222633 001 There was generally strong skepticism among this group about claims of environmental benefits. Toman even pointed out that Howard Gruenspecht had actually changed his position on this over the past year. I think that the best we can do here is claim environmental neutrality, and maybe : some qualitative statements about potential gains . putting numbers on this is going to get us in trouble. Other Benefits: There may be gains from reducing costs of meeting NAAQS, or health benefits (but not both). There was support for some benefits here. Toman thought that most of the action would be health benefits, since he didn't think that NAAQS would be effective by 2010. Ray notes that administration policy is that NAAQS will be effective by then, so that the benefits will bc in dollars, not health. Bottom Line: There seems to be professional support for claiming some technology gains, some small JI/LDC benefit, full Annex I trading gains, and perhaps for some gains from umbrella. But we have to recognize that this analysis will be criticized by some folks on the grounds that it is too idealistic on trading, and by others on the grounds that it implies that we buy our way out of the majority of our reductions. Both these criticisms will rise in proportion to the gains that WC claim from umbrella. It seems like an analysis with 1.25 AEEI, Annex I trading including EU (no umbrella), and JI/LDC that gets 20% of WW trading benefits would be pretty well received. Moving to the umbrella would bring some criticism, but not an avalanche, Going beyond that on trading or technology would be steeply criticized. I remain unclear on how much we can justify 00 JI/LDC 12/30/97 TUE 10:36 FAI 202 6222633 005 Given these comments, here is a revised set of runs that I think we should do: Model Permutations to Consider: 1) Full Annex I trading 2) Umbrella - trading block that excludes EU and Eastern Europe 3) AEEI of 1.25 4) JI/LDC participation that gets us 20% of the gains of worldwide trading (beyond Annex I trading). Question: Are interactions linear? Worth trying some combined and separate estimates to get a feel for this, although we don't want to get out of control. I would suggest the following set of runs: 1-3) Do the first three factors individually 4) Annex 1 trading with JI/LDC 5) Annex I trading, with umbrella, and JI/LDC 6) Higher AEEI, with Annex I trading 7) Higher AEEI, with Annex I trading and umbrella 8) Higher AEEI, with Annex I trading and JI/LDC 9) Higher AEEI, with Annex I trading, umbrella, and JI/LDC For each of these cases, we should show what happens to permit price, and what share of our reductions are purchased abroad. phonecon.m13 Page 1 Talking Points for Calls to Potential Economist Validators Wanted to talk to you about global climate change and the Kyoto agreement. Know you have been following issue, and wanted to provide our perspective on the economics involved: As the letter from 2,500 economists [which you signed] on this issue indicates, "The most efficient approach to slowing climate change is through market based policies. In order for the world to achieve its climate objectives at minimum cost, a cooperative approach among nations is required -- such as an international emissions trading program." The treaty we agreed to in Kyoto, and the Administration's overall approach to climate change, emphasize this market-based approach. The Kyoto agreement includes provisions for international trading, for a Clean Development Mechanism, and for other sources of flexibility. We prevailed on these and other key points against very broad opposition. At same time, the treaty did not bring developing countries on board and that is the next great hurdle. Without it, treaty will never get ratified and the global problem won't get solved. I think it is extremely important that economists start pointing out that if the provisions of the Kyoto agreement are fully implemented, the costs are likely to be quite modest. In particular, international trading and other aspects of the treaty (e.g., six gases, sinks) dramatically reduce the costs involved, and many outside studies are neglecting to reflect the institutional details of what was actually agreed. The opposition, as you may have seen, is already on the field. [WEFA] What I would like to do is start to engage thoughtful leaders like yourself on this issue. I would like to get your best thoughts on how we should be proceeding in analyzing the issue and making our case with the public. And I would hope to enlist your moral support for our efforts to deal with this issue in a sound and sober manner. We will also need your help in thinking through many of the important questions as we move forward. For example, if we ratify the Kyoto agreement, designing an effective cap-and-trade system for carbon emissions will be one of our most important public policy challenges over the next decade, and we will need the help of people like you to make sure we get it right. We are trying to organize a dinner with the Vice President, Deputy Secretary Summers, and Chair Yellen in the near future to focus on the economic aspects of climate change, and I hope you'll be available to join us. JF 10/16/97 DRAFT TALKING POINTS, TO DEFEND AN AGGRESSIVE GCC PROGRAM We will protect the environment, and at the same time protect the economy. Vigorous economic growth and environmental protection must go hand in hand. The President has spent five years growing the economy, and laying the foundations to sustain that growth into the future. He is not going to endanger that record now. "I have worked far too hard to revitalize the American economy to jeopardize our progress now." [Oct. 1, 1997, WH meeting with weather forecasters.] The U.S. will not go along with an agreement that threatens sustained economic growth. While the President is deeply committed to reducing carbon emisssion, no one could reasonably expect that he -- or future adminstrations or Congresses -- would allow the costs of this effort to become extreme. That's why we have built in regular economic reviews, to monitor our progress and re-assess how the economy is responding. Economic models are always subject to a great deal of uncertainty, and forecasts are all the more so. In this case we are talking about unexplored territory. Fifteen years ago, we never thought we souls see an unemployment rate under 5 percent, a core inflation rate of under 2.5 percent, and a Federal budget deficit well under $50 billion. But that's what has happened. So a degree of modesty with regard to our forecasting abilities is appropriate. The economics of technological progress, for example, is a big unknown. The processes of invention, innovation, diffusion, and learning, fall under the sway of other Muses at least as much as under the dictates of the goddess of Economics. The President has faith in the American people, in their ingenuity and determination to overcome obstacles. In 194_ President Roosevelt called the leaders of industry to the White House, and asked how much, with an all-out effort, they could increase their production of war- related materials and equipment. He then announced to the American people goals that were more ambitious by an order of %. In the end, those goals were attained. The key was national mobilization in the fact of a shared global threat. Climate change is a shared global threat. We might not be able to replicate the achievement of 1941. The American people are not all familiar with the new threat. But the President believes that we will never know what we can accomplish unless we try. "If we would change our habits tomorrow, just some of our habits, we could with no extra charge, no cost at all on society, get rid of 20 percent of the GHG with presently available technology' [8/4/97 press conference]. It is impossible to know to what extent American will heed the call. We consumers are fond of our incandescent light bulbs, our high-flow showers and our sport utility vehicles. Industry is fond of its old ways of doing things as well. But the adoption of carbon- saving technology has to be a large part of the solution, and there is no knowing how far it can take us. Even more unknown than future technological progress is the extent to which the cost of addressing climate change can be substantially reduced by so-called "where flexibility." There are many other countries that could reduce emissions relative to their "business-as-usual" paths at lower costs than can industrialized countries such as the U.S.. We are insisting at Kyoto on a number of treaty provisions that would take advantage of this to reduce the costs to the American economy. These provisions include international trading, joint implementation, and -- most significantly - participation by developing countries. Many countries disagree with one or more of these proposals. Furthermore, even if we somehow were able to win agreement with all our flexibility proposals, it is difficult to know whether they could be implemented efficiently and enforced fully. But if these uncertainties were resolved in a positive manner, there would be scope for bringing down the costs very substantially. DRART ≈: DO YOU BELIEVE THAT IT IS POSSIBLE TO MEET THE BINDING TARGET OF 1990 EMISSIONS LEVELS BY 2010 WITHOUT INFLICTING HEAVY COSTS ON THE ECONOMY? A: I believe that the President's package of R&D initiatives, tax cuts for energy efficiency, Federal government energy initiatives, and the market-based permit trading system will be a solid first step in addressing the critical problem of global climate change. While the President is deeply committed to reducing carbon emissions, no one could reasonably expect that he - or future administrations or Congresses would allow the costs of this effort to become extreme. That's why we have built in regular economic reviews, to monitor our progress and re-assess how the economy is responding. [SAFETY VALVE ALSO.] After all, making predictions 10 or 15 years into the future is very difficult. We can not know with any precision how factors such as technological breakthroughs or successful international trading could lower the costs of addressing climate change. ä BUT DON'T MODELS SHOW THAT REACHING 1990 LEVELS BY 2010 WILL REQUIRE $100 OR $200 PER TON INCREASES IN CARBON PRICES? A: There is no definitive answer as to how much it will cost to reach 1990 levels by 2010. Different economic models come up with different numbers based on different assumptions. Many published estimates are in the range that you mention. But as just one example of the uncertainties involved, full worldwide permit trading -- which admittedly may be difficult to implement -- has the potential to reduce costs substantially, to perhaps $10 or $20 per ton. If such a worldwide system of trading were to develop, the costs will thus be substantially lower. In sum, there is substantial uncertainty surrounding how the economy will behave in 10 or 15 years. I must admit that 15 years ago, I never thought I would see an unemployment rate under 5 percent, a core inflation rate of under 2.5 percent, and a Federal budget deficit well under $50 billion. But that's what's happened. So I would urge a similar degree of modesty here. The important point is that the President's plan is a solid first step in reducing carbon emissions without inflicting excessive costs on the economy. Q: BUT DOESN'T YOUR OWN ANALYSIS SHOW THAT REACHING 1990 BY 2010 WOULD INVOLVE SUBSTANTIAL COSTS? A: It seems inappropriate for me to reveal the analysis and views that I have shared with the President. I can say that I have informed him of the results from conventional economic models available in the public domain, and discussed both the benefits and limitations of such models. Given the importance I attach to frank internal discussions, I would prefer not to elaborate beyond that. Q: HOW DO YOU DEFINE "EXCESSIVE" COSTS? IS $50 PER TON EXCESSIVE? A: [ANSWER DEPENDS ON SAFETY VALVE DECISION.] ≈: BUT WON'T THIS EFFORT CAUSE JOB LOSSES OF OVER 1.5 MILLION JOBS? A: President Clinton's top priority, since his first days in office, has been revitalizing the economy, creating jobs and investing in people and technology to enhance long-term growth. And we have made tremendous progress: growth is strong, unemployment is down to 4.9 percent, and over 13 million jobs have been created since January 1993. He is not going to jeopardize that progress. As for the specific impact of reducing carbon emissions to 1990 levels by 2010, I would emphasize once again that different economic models come up with different numbers based on different assumptions, and that there is substantial uncertainty involved in the estimates. Furthermore, no one could reasonably expect that this President -- or future administrations or Congresses - would allow the costs of the effort to become extreme. That's why we have built in regular economic reviews, to monitor our progress and re-assess how the economy is responding. [SAFETY VALVE ALSO.] ≈: IF YOU DON'T KNOW HOW MUCH IT WILL COST TO REACH 1990 LEVELS BY 2010, HOW CAN YOU JUSTIFY GOING FORWARD WITH SUCH A POLICY? As I have said, it is difficult to determine the exact cost of policies designed to reach 1990 emissions levels by 2010. Global climate change involves economic and physical processes that operate globally and over decades, if not centuries. But the science is clear and compelling: we humans are changing the global climate. Over time, climate changes will disrupt agriculture, cause droughts and floods and the spred of infectious diseases. We must take steps now to reduce greenhouse gas emissions, and after carefully weighing the advantages and disadvantages of different approaches, the President has decided that reducing emissions to 1990 levels by 2010 is the best way forward. The President's approach -- which starts with increased R&D spending, tax incentives for energy efficiency, and an aggressive effort on the Federal government's own energy efficiency -- recognizes the uncertainties involved in reducing carbon emissions. No one could reasonably expect that this President -- or future administrations or Congresses -- would allow the costs of the effort to become extreme. That's why we have built in regular economic reviews, to monitor our progress and re-assess how the economy is responding. [SAFETY VALVE ALSO.] DRAFT Q: DO YOU BELIEVE THAT IT IS POSSIBLE TO MEET THE BINDING TARGET OF 1990 EMISSIONS LEVELS BY 2010 WITHOUT INFLICTING HEAVY COSTS ON THE ECONOMY? A: I believe that the President's package of R&D initiatives, tax cuts for energy efficiency, Federal government energy initiatives, and the market-based permit trading system will be a solid first step in addressing the critical problem of global climate change. While the President is deeply committed to reducing carbon emissions, no one could reasonably expect that he -- or future administrations or Congresses -- would allow the costs of this effort to become extreme. That's why we have built in regular economic reviews, to monitor our progress and re-assess how the economy is responding. [SAFETY VALVE ALSO.] After all, making predictions 10 or 15 years into the future is very difficult. We can not know with any precision how factors such as technological breakthroughs or successful international trading could lower the costs of addressing climate change. a: BUT DON'T MODELS SHOW THAT REACHING 1990 LEVELS BY 2010 WILL REQUIRE $100 OR $200 PER TON INCREASES IN CARBON PRICES? A: There is no definitive answer as to how much it will cost to reach 1990 levels by 2010. Different economic models come up with different numbers based on different assumptions. Many published estimates are in the range that you mention. But as just one example of the uncertainties involved, full worldwide permit trading -- which admittedly may be difficult to implement -- has the potential to reduce costs substantially, to perhaps $10 or $20 per ton. If such a worldwide system of trading were to develop, the costs will thus be substantially lower. In sum, there is substantial uncertainty surrounding how the economy will behave in 10 or 15 years. I must admit that 15 years ago, I never thought I would see an unemployment rate under 5 percent, a core inflation rate of under 2.5 percent, and a Federal budget deficit well under $50 billion. But that's what's happened. So I would. urge a similar degree of modesty here. The important point is that the President's plan is a solid first step in reducing carbon emissions without inflicting excessive costs on the economy. Q: BUT DOESN'T YOUR OWN ANALYSIS SHOW THAT REACHING 1990 BY 2010 WOULD INVOLVE SUBSTANTIAL COSTS? A: It seems inappropriate for me to reveal the analysis and views that I have shared with the President. I can say that I have informed him of the results from conventional economic models available in the public domain, and discussed both the benefits and limitations of such models. Given the importance I attach to frank internal discussions, I would prefer not to elaborate beyond that. ≈: HOW DO YOU DEFINE "EXCESSIVE" COSTS? IS $50 PER TON EXCESSIVE? A: [ANSWER DEPENDS ON SAFETY VALVE DECISION.] Q: BUT WON'T THIS EFFORT CAUSE JOB LOSSES OF OVER 1.5 MILLION JOBS? A: President Clinton's top priority, since his first days in office, has been revitalizing the economy, creating jobs and investing in people and technology to enhance long-term growth. And we have made tremendous progress: growth is strong, unemployment is down to 4.9 percent, and over 13 million jobs have been created since January 1993. He is not going to jeopardize that progress. As for the specific impact of reducing carbon emissions to 1990 levels by 2010, I would emphasize once again that different economic models come up with different numbers based on different assumptions, and that there is substantial uncertainty involved in the estimates. Furthermore, no one could reasonably expect that this President -- or future administrations or Congresses -- would allow the costs of the effort to become extreme. That's why we have built in regular economic reviews, to monitor our progress and re-assess how the economy is responding. [SAFETY VALVE ALSO.] Q: IF YOU DON'T KNOW HOW MUCH IT WILL COST TO REACH 1990 LEVELS BY 2010, HOW CAN YOU JUSTIFY GOING FORWARD WITH SUCH A POLICY? As I have said, it is difficult to determine the exact cost of policies designed to reach 1990 emissions levels by 2010. Global climate change involves economic and physical processes that operate globally and over decades, if not centuries. But the science is clear and compelling: we humans are changing the global climate. Over time, climate changes will disrupt agriculture, cause droughts and floods and the spred of infectious diseases. We must take steps now to reduce greenhouse gas emissions, and after carefully weighing the advantages and disadvantages of different approaches, the President has decided that reducing emissions to 1990 levels by 2010 is the best way forward. The President's approach -- which starts with increased R&D spending, tax incentives for energy efficiency, and an aggressive effort on the Federal government's own energy efficiency -- recognizes the uncertainties involved in reducing carbon emissions. No one could reasonably expect that this President -- or future administrations or Congresses -- would allow the costs of the effort to become extreme. That's why we have built in regular economic reviews, to monitor our progress and re-assess how the economy is responding. [SAFETY VALVE ALSO.] Some analysts have proposed using revenues from the sale of emissions permits (designed to slow global warming) to plug the projected entitlements financing gap. This memo argues that earmarking is a bad idea, because there is no inherent connection between the magnitude of the two streams, and because the size of both streams (particularly the emission permit revenues) is extremely uncertain. Comparison of Anticipated Magnitudes The first three columns of the attached table show the Social Security Actuary's low, intermediate, and high estimates of the annual financing shortfall in the Social Security system as a fraction of GDP (excluding interest payments or receipts on trust fund balances).¹ In the intermediate projections, the shortfall is monotonically increasing over time, although about 80 percent of the ultimate deterioration has occurred by 2030, when the shortfall is projected to be about 1.7 percent of GDP. The next three columns show projected emissions permit revenues as a fraction of GDP under the "Peak in 2015" plan (the least aggressive option considered), under three assumptions about the nature of the market for emissions permits: no international trading, trading among Annex I countries, and global trading.2 Revenue estimates in 2030 vary by about a factor of ten depending on the assumption about trading. In the "no trading" regime, revenues in 2030 are roughly the same size as the Social Security shortfall, but under the other two options revenues fall far short. By 2050, revenues under "no trading" are more than twice the Social Security shortfall, but under the Annex I trading regime revenues are about 3/4 of the shortfall. The remaining columns of the table show projections of revenues under Annex I trading for the main alternatives to the "Peak in 2015" plan. Revenues for either the "-10% in 2010" or the "1995 in 2010" plans roughly match the intermediate projection of the Social Security financing gap by 2050, but in the early decades revenues substantially exceed the financing gap. Revenues from the "1990 in 2010" plan exceed the financing gap early on but are about 2/3 of the financing gap by 2050. Of course, the emission-permit revenue scenarios that best match the intermediate projections of the Social Security shortfall do not even remotely match the low or high projections of the Social Security shortfall. Furthermore, the degree of uncertainty about the emission permit revenues is enormous. As an example, the current market price of sodium dioxide emissions permits in the US is about 1/10 of the price that economists had anticipated before the opening of the market. Public Finance Considerations Standard public finance theory indicates that earmarked taxes are generally a bad idea, because an optimizing social planner should be free to choose the overall tax schedule to minimize ¹This memo ignores Medicare because the cost estimates for Medicare are so extremely uncertain at this point. ²The permit price streams, emissions paths, and GDP for the climate policy simulations are outputs of SGM. Permit revenues assume an annual auction of all permits allocated to the U.S. goyernment. efficiency losses, and free to allocate incoming revenues to the places where the social marginal utility of expenditure is highest. From this perspective, the circumstances in which earmarking is least damaging are those in which the tax revenues bear some relationship to expenditure needs. The best example is the dedication of gasoline tax revenues to a trust fund for highway construction and maintenance. On this test, earmarking emissions permit revenues to financing Social Security is a terrible idea. Many of the kinds of shocks that would affect emission permit revenues (especially improvement of emissions technologies) are completely unrelated to shocks to the need for Social Security revenues, while shocks to Social Security needs are probably negatively correlated with emission permit revenues. For example, if fertility increases, Social Security financing needs will decline (at least over the projection period), but the extra population will produce more pollution and therefore more emission permit revenues. The most persuasive arguments for earmarked taxes are generally public choice or political economy arguments, which begin either from the premise that the policymaker's objective function differs from that of society, or that the politics of the specific earmarked tax and expenditure have some important interaction that justifies the inefficiencies of earmarking. The best argument that can be made along such lines in this context is that the emission permit is "good" in a sense not likely to be appreciated by the political system at large (presumably because of global externalities), and that therefore it is important to tie the selfish interests of some powerful constituency to the fate of the emissions permits revenues. Specifically, the argument is that if emission permits are perceived as funding Social Security, they may survive political battles because the AARP would defend them (I use the AARP here as a stand-in for all self-proclaimed guardians of Social Security). This argument is not persuasive. The enormous uncertainty about the magnitude of emission permit revenues makes it seem very unlikely that the AARP would be keen to tie the fate of Social Security to an emissions-permit horse in the first place. Even if the Social Security and emissions permits could be linked initially, it seems likely that any shocks to emissions permit revenues would only make the AARP more eager to switch to a more reliable source of revenues. By contrast, the payroll tax is very well-suited as a source of Social Security revenues, because benefits are a function of wages and wage growth, so long-term productivity shocks affect both revenues and expenditures in similar ways. Conclusion A less explicit link between entitlements needs and emission permit revenues is more palatable than earmarking. In essentially any forseeable future, the retirement of the baby boom generation will expand the Federal spending beyond any automatic increase in revenues, putting pressure on the Federal deficit. Extra general revenues from emissions permits would likely be used to some extent to plug that deficit hole, although it also seems likely that part of the revenues would be absorbed by extra spending. One could then use standard public finance arguments to say that, if emission permits are desirable from a social welfare perspective, they should be phased in gradually over time, with a time path chosen to deliver revenues just when the budget deficit is under the greatest pressure. Emissions Trading Revenues Social Security Financing Gap (Annual, Percent of GDP) (Annual, Percent of GDP) Peak in 2015 1990 in 2010 -10% in 2010 1995 in 2010 Year Low Intermediate High US Only Annex I Global Annex I Annex I Annex I 2000 0.6 0.4 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2005 0.7 0.3 -0.2 0.0 0.0 2010 0.7 0.1 -0.5 0.0 0.0 0.0 0.6 1.2 1.2 2015 0.3 -0.3 -1.0 0.0 0.0 0.0 0.8 1.4 1.4 0.1 0.4 0.1 0.9 1.5 1.5 2020 -0.2 -0.9 -1.6 2025 -0.5 -1.4 -2.3 1.2 0.7 0.2 1.0 1.5 1.6 2030 -0.7 -1.7 -2.7 1.7 0.9 0.2 1.0 1.5 1.6 -1.8 -3.0 2.3 1.1 0.2 1.1 1.6 1.6 2035 -0.7 2040 -0.5 -1.7 -3.2 3.0 1.2 0.3 1.2 1.6 1.7 2045 -0.4 -1.7 -3.4 3.7 1.3 0.3 1.2 1.7 1.8 2050 -0.3 -1.7 -3.6 4.7 1.3 0.3 1.3 1.8 1.9 Social Security & Carb Tax Revenues Some analysts have proposed using revenues from the sale of emissions permits (designed to slow global warming) to plug the projected entitlements financing gap. This memo argues that earmarking is a bad idea, because there is no inherent connection between the magnitude of the two streams, and because the size of both streams (particularly the emission permit revenues) is extremely uncertain. Comparison of Anticipated Magnitudes The first three columns of the attached table show the Social Security Actuary's low, intermediate, and high estimates of the annual financing shortfall in the Social Security system as a fraction of GDP (excluding interest payments or receipts on trust fund balances). In the intermediate projections, the shortfall is monotonically increasing over time, although about 80 percent of the ultimate deterioration has occurred by 2030, when the shortfall is projected to be about 1.7 percent of GDP. The next three columns show projected emissions permit revenues as a fraction of GDP under the "Peak in 2015" plan (the least aggressive option considered), under three assumptions about the nature of the market for emissions permits: no international trading, trading among Annex I countries, and global trading.² Revenue estimates in 2030 vary by about a factor of ten depending on the assumption about trading. In the "no trading" regime, revenues in 2030 are roughly the same size as the Social Security shortfall, but under the other two options revenues fall far short. By 2050, revenues under "no trading" are more than twice the Social Security shortfall, but under the Annex I trading regime revenues are about 3/4 of the shortfall. The remaining columns of the table show projections of revenues under Annex I trading for the main alternatives to the "Peak in 2015" plan. Revenues for either the "-10% in 2010" or the "1995 in 2010" plans roughly match the intermediate projection of the Social Security financing gap by 2050, but in the early decades revenues substantially exceed the financing gap. Revenues from the "1990 in 2010" plan exceed the financing gap early on but are about 2/3 of the financing gap by 2050. Of course, the emission-permit revenue scenarios that best match the intermediate projections of the Social Security shortfall do not even remotely match the low or high projections of the Social Security shortfall. Furthermore, the degree of uncertainty about the emission permit revenues is enormous. As an example, the current market price of sodium dioxide emissions permits in the US is about 1/10 of the price that economists had anticipated before the opening of the market. Public Finance Considerations Standard public finance theory indicates that earmarked taxes are generally a bad idea, because an optimizing social planner should be free to choose the overall tax schedule to minimize ¹This memo ignores Medicare because the cost estimates for Medicare are so extremely uncertain at this point. ²The permit price streams, emissions paths, and GDP for the climate policy simulations are outputs of SGM. Permit revenues assume an annual auction of all permits allocated to the U.S. government. efficiency losses, and free to allocate incoming revenues to the places where the social marginal utility of expenditure is highest. From this perspective, the circumstances in which earmarking is least damaging are those in which the tax revenues bear some relationship to expenditure needs. The best example is the dedication of gasoline tax revenues to a trust fund for highway construction and maintenance. On this test, earmarking emissions permit revenues to financing Social Security is a terrible idea. Many of the kinds of shocks that would affect emission permit revenues (especially improvement of emissions technologies) are completely unrelated to shocks to the need for Social Security revenues, while shocks to Social Security needs are probably negatively correlated with emission permit revenues. For example, if fertility increases, Social Security financing needs will decline (at least over the projection period), but the extra population will produce more pollution and therefore more emission permit revenues. The most persuasive arguments for earmarked taxes are generally public choice or political economy arguments, which begin either from the premise that the policymaker's objective function differs from that of society, or that the politics of the specific earmarked tax and expenditure have some important interaction that justifies the inefficiencies of earmarking. The best argument that can be made along such lines in this context is that the emission permit is "good" in a sense not likely to be appreciated by the political system at large (presumably because of global externalities), and that therefore it is important to tie the selfish interests of some powerful constituency to the fate of the emissions permits revenues. Specifically, the argument is that if emission permits are perceived as funding Social Security, they may survive political battles because the AARP would defend them (I use the AARP here as a stand-in for all self-proclaimed guardians of Social Security). This argument is not persuasive. The enormous uncertainty about the magnitude of emission permit revenues makes it seem very unlikely that the AARP would be keen to tie the fate of Social Security to an emissions-permit horse in the first place. Even if the Social Security and emissions permits could be linked initially, it seems likely that any shocks to emissions permit revenues would only make the AARP more eager to switch to a more reliable source of revenues. By contrast, the payroll tax is very well-suited as a source of Social Security revenues, because benefits are a function of wages and wage growth, so long-term productivity shocks affect both revenues and expenditures in similar ways. Conclusion A less explicit link between entitlements needs and emission permit revenues is more palatable than earmarking. In essentially any forseeable future, the retirement of the baby boom generation will expand the Federal spending beyond any automatic increase in revenues, putting pressure on the Federal deficit. Extra general revenues from emissions permits would likely be used to some extent to plug that deficit hole, although it also seems likely that part of the revenues would be absorbed by extra spending. One could then use standard public finance arguments to say that, if emission permits are desirable from a social welfare perspective, they should be phased in gradually over time, with a time path chosen to deliver revenues just when the budget deficit is under the greatest pressure. Emissions Trading Revenues Social Security Financing Gap (Annual, Percent of GDP) (Annual, Percent of GDP) Peak in 2015 1990 in 2010 -10% in 2010 1995 in 2010 Year Low Intermediate High US Only Annex I Global Annex I Annex I Annex I 2000 0.6 0.4 0.2 0.0 0.0 0.0 0.0 0.0 0.0 2005 0.7 0.3 -0.2 0.0 0.0 0.0 0.0 0.0 0.0 2010 0.7 0.1 -0.5 0.0 0.0 0.0 0.6 1.2 1.2 2015 0.3 -0.3 -1.0 0.0 0.0 0.0 0.8 1.4 1.4 2020 -0.2 -0.9 -1.6 0.1 0.4 0.1 0.9 1.5 1.5 2025 -0.5 -1.4 -2.3 1.2 0.7 0.2 1.0 1.5 1.6 2030 -0.7 -1.7 -2.7 1.7 0.9 0.2 1.0 1.5 1.6 2035 -0.7 -1.8 -3.0 2.3 1.1 0.2 1.1 1.6 1.6 2040 -0.5 -1.7 -3.2 3.0 1.2 0.3 1.2 1.6 1.7 2045 -0.4 -1.7 -3.4 3.7 1.3 0.3 1.2 1.7 1.8 2050 -0.3 -1.7 -3.6 4.7 1.3 0.3 1.3 1.8 1.9 09/11/97 THU 16:31 FAX 202 6222633 001 Office of Economic Policy Department of the Treasury Washington, D.C. 20220 FAX Date: 9/11/97 Number of pages including cover sheet: 11 Name Fax Number Phone Number To: Joe Aldy 395-6853 395-1455 From: Lara Thuldoon 202-622-2633 622-6773 REMARKS: Urgent For your review Reply ASAP Please comment 002 ranges Effect on Fuel Prices 4P if AP if AP if AP if APif AP if APif permit permit permit permit permit permit permit price = price = price = price = price = price = price = AP if permit 4P if permit $100 $50 $80 $90 $140 $30 $200 price = $5 price = $25 Natural gas $1.6000 $0.8000 $1.2800 $1.4400 $2.2400 $0.4800 $3.2000 $0.0800 0.4 $/thousand Cubic Feet Fuel Oil $0.3000 $0.1500 $0.2400 $0.2700 $0.4200 $0.0900 $0.6000 $0.0150 0.075 $/gallon Electricity $0.0150 $0.0075 $0.0120 $0.0135 $0.0210 $0.0045 $0.0300 $0.0008 0.00375 $/kWh LPG $0.3000 $0.1500 $0.2400 $0.2700 $0.4200 $0.0900 $0.6000 $0.0150 0.075 $/gallon Gasoline $0.2600 $0.1300 $0.2080 $0.2340 $0.3640 $0.0780 $0.5200 $0.0130 0.065 $/gallon Coal $70.00 $35.00 $56.00 $63.00 $98.00 $21.00 $140.00 $3.50 17.5 $/ton Oil $14.00 $7.00 $11.20 $12.60 $19.60 $4.20 $28.00 $0.70 3.5 $/barrel 09/11/97 THU 16:31 FAX 202 6222633 Par 1 003 Joe's Runs 1 Incidence of an Energy tax by Percent Share of Current Income Permit price = $5 Number % HH Q/hh of QAP Scaled of HH using each US down by A Expenditure - (millions) Units/HH this fuel HH P AP QAP 12% AQ/Q & = -0.4 Natural Gas 58.7 87.50 60.77% 53.17 $6.20 0.08 $4.25 $3.74 -0.01 $2.53 Units - thousand cubic feet $/gal Fuel Oil 10.8 684.00 11.18% 76.47 $0.90 0.02 $1.15 $1.01 -0.01 $0.68 Units - Gallons $/gal Electricit y 96.6 9,965.00 100.00% 9,965.00 $0.08 0.00 $7.47 $6.58 0.00 $4.46 Units - kWh $/kWh LPG 8.1 513.00 8.39% 43.02 $0.92 0.02 $0.65 $0.57 -0.01 $0.38 Units - Gallons $/gal Gasoline 84.9 1,067.00 87.89% 937.77 $1.16 0.01 $12.19 $10.73 0.00 $7.26 Units - Gallons $/gal $/gal Total Direct Costs per Household (1992 dollars) $25.71 $22.63 $15.31 Total Indirect Costs per Household (1992 dollars) $28.45 $25.04 $16.94 09/11/97 THU 16:32 FAX 202 6222633 Total Costs per Household (1992 dollars) $54.16 $47.66 $32.25 Median HH Income (1992 dollars) $30,543.18 $30,543.18 $30,543.18 increased direct expenditure as a percentage of median income 0.08% 0.07% 0.05% increased total expenditure as a percentage of median income 0.18% 0.16% 0.11% Par- 1 4 004 Joe's Runs Incidence of an Energy tax by Percent Share of Current Income Permit price = $25 > Number % HH Q/hh of QAP Scaled of HH using each US down by A Expenditure - (millions) Units/HH this fuel HH P AP QAP 12% AQ/Q E = -0.4 Natural Gas 58.7 87.50 60.77% 53.17 $6.20 0.40 $21.27 18.72 -0.03 $12.21 Units - thousand cubic feet S/gal Fuel Oil 10.8 684.00 11.18% 76.47 $0.90 0.08 $5.74 5.05 -0.03 $3.25 Units - Gallons $/gal Electricit y 96.6 9,965.00 100.00% 9,965.00 $0.08 0.00 $37.37 32.88 -0.02 $21.74 Units - kWh $/kWh LPG 8.1 513.00 8.39% 43.02 $0.92 0.08 $3.23 2.84 -0.03 $1.83 Units - Gallons Sigal Gasoline 84.9 1,067.00 87.89% 937.77 $1.16 0.07 $60.95 53.64 -0.02 $35.21 Units - Gallons Sigal $/gal Total Direct Costs per Household (1992 dollars) $128.55 $113.13 $74.24 Total Indirect Costs per Household (1992 dollars) $142.25 $125.18 $82.15 09/11/97 THU 16:32 FAX 202 6222633 Total Costs per Household (1992 dollars) $270.80 $238.31 $156.38 Median HH Income (1992 dollars) $30,543.18 $30,543.18 $30,543.18 increased direct expenditure as a percentage of median income 0.42% 0.37% 0.24% increased total expenditure as a percentage of median income 0.89% 0.78% 0.51% Par ? 005 Joe's Runs Incidence of an Energy tax by Percent Share of Current Income Permit price = $30 Number % HH Q/hh of QAP Scaled of HH using each US down by A Expenditure - (millions) Units/HH this fuel HH P AP QAP 12% AQ/Q E = -0.4 Natural Gas 58.7 87.50 60.77% 53.17 $6.20 $0.4800 $25.52 $22.46 -0.03 $14.52 Units - thousand cublc feet $/gal $/gal Fuel Oil 10.8 684.00 11.18% 76.47 $0.90 $0.0900 $6.88 $6.06 -0.04 $3.85 Units - Gallons $/gal $/gal Electricit y 96.6 9,965.00 100.00% 9,965.00 $0.08 $0.0045 $44.84 $39.46 -0.02 $25.92 Units - kWh $/kWh $/kWh LPG 8.1 513.00 8.39% 43.02 $0.92 $0.0900 $3.87 $3.41 -0.04 $2.17 Units - Gallons $/gal $/gal Gasoline 84.9 1,067.00 87.89% 937.77 $1.16 $0.0780 $73.15 $64.37 -0.03 $41.92 Units - Gallons $/gal $/gal Total Direct Costs per Household (1992 dollars) $154.26 $135.75 $88.39 Total Indirect Costs per Household (1992 dollars) $170.70 $150.22 09/11/97 THU 16:33 FAX 202 6222633 $97.81 Total Costs per Household (1992 dollars) $324.97 $285.97 $186.20 Median HH Income (1992 dollars) $30,543.18 $30,543.18 $30,543.18 increased direct expenditure as a percentage of median income 0.51% 0.44% 0.29% increased total expenditure as a percentage of median income 1.06% 0.94% 0.61% Par 006 1. ranges Incidence of an Energy tax by Percent Share of Current Income Permit price = $50 Number % HH Q/hh of of HH using each US A Expenditur QAP scaled (millions) Units/HH this fuel HH P AP AQ/Q e - E = -0.4 QAP down by 12% Natural Gas 58.7 87.50 60.77% 53.17 $6.20 0.80 -0.05 $23.33 $42.54 $37.43 Units - thousand cubic feet $/gal $/gal Fuel Oil 10.8 684.00 11.18% 76.47 $0.90 0.15 -0.07 $6.12 $11.47 $10.09 Units - Gallons $/gal $/gal Electricity 96.6 9,965.00 100.00% 9,965.00 $0.08 0.01 -0.04 $42.11 $74.74 $65.77 Units - kWh $/kVh $/kWh LPG 8.1 513.00 8.39% 43.02 $0.92 0.15 -0.07 $3.45 $6.45 $5.68 Units - Gallons $/gal $/gal Gasoline 84.9 1,067.00 87.89% 937.77 $1.16 0.13 -0.04 $67.68 $121.91 $107.28 Units - Gallons $/gal $/gal Total Direct Costs per Household (1992 dollars) $142.68 $257.11 $226.25 Total Indirect Costs per Household (1992 dollars) $157.89 $284.50 $250.36 Total Costs per Household (1992 dollars) $300.57 $541.61 $476.62 Median HH Income (1992 dollars) $30,543.18 $30,543.18 $30,543.18 increased direct expenditure as a percentage of median income 0.47% 0.84% 0.74% 09/11/97 THU 16:33 FAX 202 6222633 increased total expenditure as a percentage of median income 0.98% 1.77% 1.56% Par , 007 1. ranges Incidence of an Energy tax by Percent Share of Current Income Permit price = $80 Number % HH Q/hh of of HH using each US A Expenditur QAP scaled (millions) Units/HH this fuel HH P AP AQ/Q e - S = -0.4 QAP down by 12% Natural Gas 58.7 87.50 60.77% 53.17 $6.20 1.28 -0.08 $35.21 $68.06 $59.89 Units - thousand cubic feet $/gal $/gal Fuel Oil 10.8 684.00 11.18% 76.47 $0.90 0.24 -0.11 $9.05 $18.35 $16.15 Units - Gallons $/gal $/gal Electricity 96.6 9,965.00 100.00% 9,965.00 $0.08 0.01 -0.06 $64.75 $119.58 $105.23 Units - kWh $/kWh $/kWh LPG 8.1 513.00 8.39% 43.02 $0.92 0.24 -0.10 $5.12 $10.32 $9.08 Units - Gallons $/gal $/gal Gasoline 84.9 1,067.00 87.89% 937.77 $1.16 0.21 -0.07 $103.04 $195.06 $171.65 Units - Gallons $/gal $/gal Total Direct Costs per Household (1992 dollars) $217.18 $411.37 $362.01 Total Indirect Costs per Household (1992 dollars) $240.32 $455.21 $400.58 Total Costs per Household (1992 dollars) $457.50 $866.58 $762.59 Median HH Income (1992 dollars) $30,543.18 $30,543.18 $30,543.18 09/11/97 THU 16:33 FAX 202 6222633 increased direct expenditure as a percentage of median income 0.71% 1.35% 1.19% increased total expenditure as a percentage of median income 1.50% 2.84% 2.50% Par 008 ranges Incidence of an Energy tax by Percent Share of Current Income Permit price = $90 Number % HH Q/hh of of HH using each US A Expenditur QAP scaled (millions) Units/HH this fuel HH P AP AQ/Q e- 8 = -0.4 QAP down by 12% Natural Gas 58.7 87.50 60.77% 53.17 $6.20 1.44 -0.09 $38.83 $76.57 $67.38 Units - thousand cubic feet $/gal $/gal Fuel Oil 10.8 684.00 11.18% 76.47 $0.90 0.27 -0.12 $9.91 $20.65 $18.17 Units - Gallons $/gal $/gal Electricity 96.6 9,965.00 100.00% 9,965.00 $0.08 0.01 -0.07 $71.86 $134.53 $118.38 Units - kWh $/kWh $/kWh LPG 8.1 513.00 8.39% 43.02 $0.92 0.27 -0.12 $5.61 $11.61 $10.22 Units - Gallons $/gal $/gal Gasoline 84.9 1,067.00 87.89% 937.77 $1.16 0.23 -0.08 $113.96 $219.44 $193.10 Units - Gallons S/gal $/gal Total Direct Costs per Household (1992 dollars) $240.16 $462.79 $407.26 Total Indirect Costs per Household (1992 dollars) $265.75 $512.11 $450.65 Total Costs per Household (1992 dollars) $505.90 $974.90 $857.91 Median HH Income (1992 dollars) $30,543.18 $30,543.18 $30,543.18 09/11/97 THU 16:34 FAX 202 6222633 increased direct expenditure as a percentage of median income 0.79% 1.52% 1.33% increased total expenditure as a percentage of median income 1.66% 3.19% 2.81% Pa- 1 009 Incidence 4. Incidence of an Energy tax by Percent Share of Current Income For a $100/ton Permit x Q/hh of Number of HH % HH using each US A Expenditure - QAP (millions) Units/HH this fuel HH P ДР AQ/Q E = -0.4 unadjusted 12% reduc Natural Gas 58.7 87.50 60.77% 53.17 6.20 1.60 -0.10 $42.26 $85.07 74.86377 Units - thousand cubic feet S/gal $/gal Fuel Oil 10.8 684.00 11.18% 76.47 0.90 0.30 -0.13 $10.71 $22.94 20.18862 Units - Gallons S/gal $/gal Electricity 96.6 9,965.00 100.00% 9,965.00 0.08 0.02 -0.07 $78.75 $149.48 131.538 Units - kWh $/kWh $/gal LPG 8.1 513.00 8.39% 43.02 0.92 0.30 -0.13 $6.06 $12.90 11.3561 Units -- Gallons S/gal S/gal Gasoline 84.9 1,067.00 87.89% 937.77 1.16 0.26 -0.09 $124.43 $243.82 214.5611 Units - Gallons $/gal S/gal Total Direct Costs per Household (1992 dollars) $262.21 $514.21 452.5076 Total Indirect Costs per Household (1992 dollars) $290.15 $569.01 500.7256 Total Costs per Household (1992 dollars) $552.35 $1,083.22 953.2332 Median HH Income (1992 dollars) $30,543.18 $30,543.18 increased direct expenditure as a percentage of median income 0.86% 1.68% increased total expenditure as a percentage of median income 1.81% 3.55% Sources: Number of HH - Energy Information Administration/Household Energy Consumtion and Expenditures Survey 1993, adjusted to 1994 dollars Units/HH - EIA, RECS 1993 Average Consumption per HH (adjusted to 1994 dollars) 09/11/97 THU 16:34 FAX 202 6222633 % HH using this Fuel - (Number of hh)/(units/HH) Q/hh of each US HH - (Units/HH)*(%HH using the Fuel) Price - EIA AP - Buisness Daily, August 7, 1997, p.1 AQ/Q - based on price elaticity of demand for energy of -0.4 from Energy Economics and Policy, 1980, p. 232. A Expenditure - E = -0.4 -> new price*new quantity - old price*old quantity QAP - unadjusted Pa- 1 010 ranges Incidence of an Energy tax by Percent Share of Current Income Permit price = $140 Number % HH Q/hh of of HH using each US 1 Expenditur QAP scaled (millions) Units/HH this fuel HH P ДР AQ/Q e - E = -0.4 QAP down by 12% Natural Gas 58.7 87.50 60.77% 53.17 $6.20 2.24 -0.14 $54.25 $119.10 $104.81 Units - thousand cubic feet $/gal Fuel Oil 10.8 684.00 11.18% 76.47 $0.90 0.42 -0.19 $13.28 $32.12 $28.26 Units - Gallons $/gal Electricity 96.6 9,965.00 100.00% 9,965.00 $0.08 0.02 -0.10 $104.12 $209.27 $184.15 Units - kWh $/kWh LPG 8.1 513.00 8.39% 43.02 $0.92 0.42 -0.18 $7.54 $18.07 $15.90 Units - Gallons $/gal Gasoline 84.9 1,067.00 87.89% 937.77 $1.16 0.36 -0.13 $161.96 $341.35 $300.39 Units - Gallons $/gal $/gal Total Direct Costs per Household (1992 dollars) $341.15 $719.90 $633.51 Total Indirect Costs per Household (1992 dollars) $377.50 $796.61 $701.02 Total Costs per Household (1992 dollars) $718.65 $1,516.51 $1,334.53 Median HH Income (1992 dollars) $30,543.18 $30,543.18 $30,543.18 09/11/97 THU 16:35 FAX 202 6222633 increased direct expenditure as a percentage of median income 1.12% 2.36% 2.07% increased total expenditure as a percentage of median income 2.35% 4.97% 4.37% Par 011 Joe's Runs Incidence of an Energy tax by Percent Share of Current Income Permit price = $200 Number % HH Q/hh of QAP Scaled of HH using each US down by A Expenditure - (millions) Units/HH this fuel HH P ДР QAP 12% AQ/Q 6 = -0.4 Natural Gas 58.7 87.50 60.77% 53.17 $6.20 $3.2000 $170.14 $149.73 -0.21 $66.96 Units - thousand cubic feet $/gal $/gal Fuel Oil 10.8 684.00 11.18% 76.47 $0.90 $0.6000 $45.88 $40.38 -0.27 $15.29 Units - Gallons $/gal $/gal Electricit y 96.6 9,965.00 100.00% 9,965.00 $0.08 $0.0300 $298.95 $263.08 -0.15 $135.62 Units - kWh $/kWh $/kWh LPG 8.1 513.00 8.39% 43.02 $0.92 $0.6000 $25.81 $22.71 -0.26 $8.75 Units - Gallons $/gal $/gal Gasoline 84.9 1,067.00 87.89% 937.77 $1.16 $0.5200 $487.64 $429.12 -0.18 $205.14 Units - Gallons $/gal $/gal Total Direct Costs per Household (1992 dollars) $1,028.43 $905.02 $431.77 Total Indirect Costs per Household (1992 dollars) $1,138.01 $1,001.45 $477.78 09/11/97 THU 16:35 FAX 202 6222633 Total Costs per Household (1992 dollars) $2,166.44 $1,906.47 $909.56 Median HH Income (1992 dollars) $30,543.18 $30,543.18 $30,543.18 increased direct expenditure as a percentage of median income 3.37% 2.96% 1.41% increased total expenditure as a percentage of median income 7.09% 6.24% 2.98% Par- , pre-Kyoto Comments on $500 million/year Technology Proposal DOE's response does not appear to have answered the request of the Principals for a detailed description of the "technology option" in terms of specific policies (dollars spent on R&D, dollars spent on diffusion, imposition of standards, price measures, and moral suasion) and their best-guess estimated effects. At first DOE seems to say that the predicted effects are pure R&D, and then it turns out that there are hidden assumptions about public awareness changing or standards being changed. We continue to believe that much, if not all, of the R&D is probably worth doing, particularly along the lines of the PCAST report. But we also continue to believe in a pessimistic bottom line for the odds of incremental payoffs in terms of carbon reductions by 2010. As John Holdren said in his 9/28 presentation, the time lag for the payoff, even to applied R&D, is simply too long to expect that by 2010 we would get more than "diddly" from new spending. DOE stated that this proposal is based on the efficiency case in the 5 Labs Study. In a review of the proposal presented on 9/29, it appears that there are numerous divergences from the 5 Labs Study efficiency case. In light of the inconsistencies apparent in the proposal, and the substantial methodological flaws in the 5 Labs Study, it seems imprudent to base significant policy decisions on information received thus far. The point was raised at the 8/19/97 meeting on the 5 Labs Study that this study should not go to or be used by principals. To date, there appears to be insufficient justification to modify this view. General Comments The 5 Labs Study indicates that government programs, utility programs, and state programs cost 15% of technology costs (p. 1.13). Since DOE apportions all carbon savings in their proposal to the federal programs listed in the summary chart, we assume that all costs accrue to the federal government. In the "optimistic" assessment of the costs and benefits, DOE claims that government programs will likely only cost about 7% of technology costs, because half of emissions reductions occur through standards (see below). Assuming that this is true, and that the technology costs of the efficiency case range between $26 - $49 billion/year in the optimistic assessment (p. 1.14), then government programs should cost between $1.82 - $3.49 billion/year. However, using a 13.5% government cost stipulation (based on a 15% government program cost, minus 10% [instead of 50%] of these costs because of standards), the costs of government programs would range between $3.51 - $6.62 billion/year. The proposal presented on 9/29 costs 3 to 13 times less than projected by the 5 Labs Study, and yet generates greater carbon savings. The proposal employs standards to a much lesser extent than the 5 Labs Study. In the Study, "at least half of the efficiency occurs as a result of federal policies (e.g., standards and carbon permit charges)" (p. 1.13). Since the efficiency case does not include a carbon permit system, we assume that half of the carbon savings in the efficiency case result from standards. However, in this proposal, only about 10% of the emissions reductions occur due to standards (on the high ends of the ranges, 20 MMTCE out of 206 MMTCE). The projected carbon savings for 2010, while indicated as approximately 155 MMTCE, has a range of 127-206 MMTCE. Note that this large range does not overlap with the 5 Labs Study projection of 126 MMTCE (p. 1.12). The efficiency case is portrayed in the 5 Labs Study as carbon savings that occur because of already available and cost-effective technologies. In effect, government policies are only necessary to overcome market barriers. However, this proposal appears to be more reliant on R&D than on deployment and diffusion of existing technologies. Using the upper bounds of the ranges of carbon savings, we find that more than half of emissions reductions (104 of 206 MMTCE) result from R&D programs. Specific Comments Buildings Sector Two programs claim to achieve 17-29 MMTCE. Note that under the 5 Labs efficiency case, 25 MMTCE of potential reductions were identified (p. 1.12), with a different penetration assumption than the one provided in the summary chart. The Study employed a 35% penetration rate assumption (p. 3.3), while this summary chart claims to be based on a 15-25% assumption for equipment efficiency and a 25% assumption for 21st Century Housing. With the lower penetration rate assumed in this proposal, we would assume that the carbon savings would be less than those found in the 5 Labs Study -- instead, they are greater. Based on these assumptions and the information provided by the 5 Labs Study, we would expect carbon savings to range between 11-18 MMTCE. Equipment Efficiency: The 5 Labs Study indicates that a total of 25 MMTCE can be reduced in residential and commercial buildings under the efficiency case. The 9/26 memo describes this set of programs to include fuel cells, even though fuel cells are not in the 5 Labs efficiency case (they only come into play in high efficiency case, although the text is very unclear about whether fuel cells actually generate any benefits prior to 2020). 21st Century Housing: There is insufficient detail provided in the 9/26 memo to identify the appropriate carbon savings in the 5 Labs Study. Transportation Sector The 5 Labs Study projects 73 MMTCE reductions in emissions under the efficiency case (p. 1.12), while this proposal estimates carbon savings ranging between 65 - 87 MMTCE. Note that this upper bound is almost equivalent to the high efficiency case with a $25/ton permit fee in the Study (88 MMTCE). Two PNGV Categories: The 5 Labs Study indicates that annual PNGV funding is approximately $250 million/year (p. 5.2). The two categories of increased PNGV programs would sum to $140 million/year in FY99 and $260 million/year in FY03. This increase is much less than believed necessary to achieve the goals of the PNGV program. The authors of the 5 Labs Study note that for PNGV to be successful, "substantial additional funding for R&D will be required, perhaps two to ten times what is presently being spent" (p. 5.7). The proposal increases funding to just the lower end of this required spending level. Industry Sector The 5 Labs Study projects 28 MMTCE of carbon emissions reductions under its efficiency case (p. 1.12). This proposal estimates carbon savings under three categories of programs ranging between 30 - 65 MMTCE. Note that the upper end of this range is 11 MMTCE more than the high efficiency case with a $25/ton permit fee in the 5 Labs Study. Industries of the Future: In the 5 Labs Study, the list of heavy manufacturing industries in table 4.6 (p. 4.10) corresponds to the list of the Industries of the Future on the DOE webpage (http://www.oit.doe.gov/iof/industry.html) In this table, these industries account for only 9 MMTCE in the efficiency case and 21 MMTCE in the high efficiency case ($50/ton permit system). However, the proposal notes that Industries of the Future programs can achieve carbon reductions ranging between 10 - 20 MMTCE. Utilities Sector Three categories of programs claim to generate carbon savings of 15-25 MMTCE. However, the efficiency case in the 5 Labs Study does not project any reductions in carbon emissions in this sector (p. 1.12). Further, the carbon sequestration R&D category of programs is not described in the 9/26 memo. For additional review of the methodological flaws in the sectoral analyses, refer to earlier memos that provide comments on the 5 Labs Study. EXECUTIVE OFFICE OF THE PRESIDENT COUNCIL OF ECONOMIC ADVISERS WASHINGTON, D.C. 20500 SENIOR ECONOMIST MEMORANDUM TO: Joe Romm Acting Assistant Secretary for Energy Efficiency and Renewable Energy U.S. Department of the Energy FROM: Jason Shogren IS and Joe Aldy DATE: August 22, 1997 RE: Review of 5-Labs Report: Scenarios of U.S. Carbon Reductions Enclosed you will find our comments on the report Scenarios of U.S. Carbon Reductions. We appreciate the opportunity to review the report and we trust you will incorporate our comments. 4 Comments on "Scenarios of U.S. Carbon Reductions" (5 Labs Study) General comment The report, Scenarios of U.S. Carbon Reductions, presents a useful cataloging of the technological options that could play a key role in reducing greenhouse gas emissions. The report does not present the specific policies or the behavioral responses that will trigger the adoption or diffusion of these technologies. As such, the report cannot serve as the basis of decision-making by top policy-makers (as opposed to background detail that others may find useful). The report as written, however, leaves the impression that these technologies are policy, which they are not. The U.S. government does not have directly under its control the decisions of private firms and consumers whether to adopt these technologies. The report must be either redone to reflect how likely these technologies will be used given the behavioral responses to likely policies, or reframed to reflect what it is--a catalog of technologies that might be used to address global climate change. The latter path is more straightforward and could readily be accomplished. Among other changes, it is essential that DOE remove references to "net savings" from the report because this does not reflect the actual costs of implementing these policies. Our specific comments follow. Executive Summary 1) paragraph 3: The point regarding the opportunity to reduce emissions through "a vigorous national commitment" should be explicit that this commitment is implemented through federal policies. To state that 120 MMTCE can be reduced by "energy efficiency alone" seems to imply that these reductions are free. These reductions through energy efficiency are induced by aggressive policies, and the statement should reflect that. The subsequent discussion of reductions with carbon permits should also reflect that the level of reductions can occur only through a carbon pricing policy in addition to any non-price policies (e.g., standards). 2) Paragraph 4: Estimates of energy cost savings should not be provided unless and until an adequate estimation approach is used to calculate cost savings. 3) Paragraph 5: This paragraph should really concentrate on the point that this study identifies a vast array of technologies that may reduce carbon emissions through climate policy. This is the true strength of the study, and it should be emphasized more. 1 Chapter 1 4) The entire study is premised on two sets of undefined policies: an "aggressive" or "invigorated" public sector effort to stimulate energy efficient technology adoption (efficiency case) and a "very aggressive" public sector effort to stimulate technology development and adoption (high efficiency/low carbon case). The only specific policy instrument mentioned in the report is a tradeable permit at prices of $25/ton and $50/ton. The study merely asserts that these unidentified policies would result in: 1) "better technology" (p. 2.5); 2) "higher penetration rates" (p. 2.5); 3) "changing the capital recovery factor [in industry sector] from 33% to 15%" (p. 4.6); and 4) "technological breakthroughs" (p. 5.3). These policy effects would result in reductions of about 200 MMTCE for "free": firms and individuals receive energy cost-savings greater than or equal to technology adoption costs. The report specifically states that the efficiency case reduces, but does not eliminate, market barriers and that implementation costs are ignored. Policy intervention to eliminate a market barrier cannot make society better off and generally makes society worse off to the extent that it imposes costs. In particular, a set of climate change policies to stimulate technology adoption could increase efficiency, but only because they also decrease the effect of a market failure (carbon emissions). Social welfare is improved because the net benefits of indirectly fixing the market failure exceed the costs of removing the barriers. If carbon emissions are appropriately priced (assuming in this case that $50/ton is the correct carbon price), a policy to remove a barrier will generate negative net benefits. High adoption rates will be realized once prices rise enough for some technologies to clear the barriers.¹ The economics literature has identified several barriers that help explain the slow rate of technology adoption.² Qualitative attributes of technologies can affect adoption. Consumers of technology prefer technologies because of a set of characteristics, not just energy efficiency. For example, some consumers may purchase a product with a lower efficiency because the former is more reliable than the efficiency-superior product. 1 In contrast, policy intervention to address a market failure (such as unpriced carbon emissions) may well make society better off, depending on the stringency of the intervention. 2 For a review of these issues, see Gilbert Metcalf and Kevin Hassett, "Measuring the Energy Savings from Home Improvement Investments: Evidence from Monthly Billing Data," National Bureau of Economic Research Working Paper 6074, June 1997; Jeffrey A. Dubin and Daniel L. McFadden, "An Econometric Analysis of Residential Electric Appliance Holdings and Consumption," Econometrica, vol. 52, no. 2, 1984; Adam B. Jaffe and Robert N. Stavins, "Energy-Efficiency Investments and Public Policy", Energy Journal, vol. 15, no. 2, 1994; Albert L. Nichols, "Demand-Side Management: Overcoming Market Barriers or Obscuring Real Costs?" Energy Policy, vol. 22, no. 10, 1994; Jerry Hausman, "Individual Discount Rates and the Purchase and Utilization of Energy-Using Durables," Bell Journal of Economics, vol. 10, 1979. 2 New technologies may be costly to integrate with existing ones. Until all components of a complex system are fully depreciated it may make sense to replace energy inefficient components with another like-wise inefficient component. Uncertainty of future energy prices, or constraints on the ability to borrow capital may make households act as if they have implausibly high implicit discount rates. Just because a technology is cost-effective on average does not mean that it is cost- effective for all individuals. Those who may use the device less intensively may find it optimal to purchase a less efficient model. These barriers clearly demonstrate that investments to minimize energy costs are different from investments to minimize the total costs of production. The study provides a crude estimate of the costs to government of lowering market barriers to achieve the estimated emissions reductions. A sense of the likely costs can be derived by evaluating the experience of the Climate Change Action Plan (CCAP). CCAP promotes carbon reductions through a broad array of voluntary programs that stimulate "cost-effective" technology adoption by private firms. Participation by a firm in a CCAP program is supposed to reduce carbon emissions and to offer financial returns.³ We do not address an important threshold question of why cost-minimizing firms would ever need any need help from government programs to take actions that would lower their costs. CCAP received appropriations totaling $494 million (1995$) during the FY95 - FY97 period. We assume that these costs are distributed uniformly over the three years. During this period, the Department of Energy and the Environmental Protection Agency can account for reductions of 14 MMTCE (see attached table). Evaluating the cost effectiveness of these programs requires an assumption about emissions reductions in the outyears. Since the programs are persuading firms to adopt innovations that offer financial returns, we believe that the firms would soon have adopted them in the absence of the programs. Thus the emissions reductions might last for several years. To develop illustrative cost estimates we assume between 2 and 7 years of emissions reductions. We describe our estimates given the assumption of 2 years; the method for the assumption of 7 years is identical.4 Assuming constant effectiveness of program expenditures, the emissions reductions attributable to past expenditures would grow from 2.8 million tons in the first year to 5.6 million in the second year and third years, and then decline to 2.8 million tons in the 4th year, before falling to 0 in the 5th and subsequent years. The ratio of 3 Note that CCAP efforts are incorporated in the AEO97 baseline. 4 This analysis was conducted based on the June 10 draft that stated 180 MMTCE of reductions associated with government programs could be expected under a "very aggressive" set of policies. Since the August 1 draft notes that more non-price policy induced reductions would be expected, the cost estimates provided in this discussion underestimate the program costs associated with this greater number of reductions. 3 the present value of these costs and of these tons is $33/ton using a 7 percent discount rate. Thus taking into account the lag between program expenditures and emissions reductions, the average cost effectiveness is $33/ton (with a 7 year assumption the cost effectiveness is $13/ton). Assuming constant program effectiveness would imply Federal costs of $5.9 billion per year (with the 7 year assumption the total cost estimate would be half as much.) We believe that there are several good reasons to think that the cost of reducing emissions will rise, however, as these or similar programs are expanded. For example, if the potential adopters differed in terms of their cost savings (or willingness to adopt innovations) then the government program would encounter diminishing returns as it tried to reach the less receptive users of technology. Suppose that the (marginal) program cost effectiveness deteriorated by 10 percent every time the annual reductions in emissions grew by as much as the annual reductions implied by the current program. In this instance the marginal cost of achieving 180 million tons of emissions reductions would be about 21 times greater than the marginal cost of the existing program, and far in excess of reasonable values for a ton of carbon emissions averted. In this case the total cost of the program would be $45 billion per year. Similar assumptions for the 7 year case would lead to total costs of nearly $7 billion per year, and marginal costs of $80/ton. Thus we believe that the cost of a government program to ensure energy reductions of 180 million metric tons of carbon could easily run into many tens of billions of dollars per year. In addition, the marginal cost-effectiveness of these programs could substantially exceed reasonable estimates of the value of carbon emissions reductions. Of course the government might choose to achieve these reductions using command and control type regulations, such as national building codes, or increasingly stringent CAFE standards. These would substitute private sector costs for government administrative costs. 5) 1.1, paragraph 1: Sentence 3 should note that "the improved performance and increased penetration of efficient and low-carbon technologies" are program- and price-induced. 6) 1.1, section 1.1, purpose 1: The statement about a "vigorous national commitment" should be explicit that this means a commitment through government programs. 7) 1.4, Table 1.1: The study notes that it employs EIA's Annual Energy Outlook 1997 projections for the business-as-usual case in 2010 with a modification to the transportation sector. However, the numbers in this chart do not match up with the numbers in the AEO97 reference case. In fact, they are much closer to the AEO97 low economic growth case (for industry and buildings). We assume that the labs study is using the reference case from AEO97 since it uses the reference case carbon emissions for 2010. If this is the case, then the following should be the sectoral breakdown of energy consumption (AEO97, pp. 97-99, and pp. 125-127): 4 Buildings: 36.81 quads Residential: 20.83 quads Commercial: 15.98 quads Industrial: 39.69 quads Transportation: 31.39+ quads⁵ Total: 107.89+ quads If the labs study employs the low economic growth case, then it should state so explicitly in the report and explain why this case, as opposed to the reference or other cases, was selected. Otherwise, the AEO97 numbers should be used. 8) 1.6, Table 1.2: The 2010 BAU carbon emissions for buildings and industry seem not to match the AEO97 reference case emissions for these sectors. AEO97 projects carbon emissions to be 576.1 MMTCE for buildings (residential: 321.4; commercial: 254.7) and industry to be 548.5 MMTCE in 2010. This implies that, with the adjusted transportation emissions, BAU should total 1740 MMTCE, not 1720. 9) 1.8, Table 1.3: The study states that it employs the Annual Energy Outlook business-as-usual (BAU) forecast for the buildings and industry sectors, with modifications that are "not greatly different from the EIA case" (p. 1.2) for the transportation sector.⁶ The one modification stated in the study affects the fuel efficiency of the light duty vehicle fleet. While AEO97 assumes that fuel efficiency will increase in the future, the labs study assumes that fuel efficiency will remain constant. This change should result in a slower rate of energy efficiency improvement in the economy under business as usual conditions. The AEO97 BAU forecast assumes an improvement of energy efficiency of 0.9% per year through 2015 (annual E/GDP = -0.9%) (AEO97, p. 4). The study assumes energy efficiency improvement under BAU occurs at a rate of 0.77%. This difference appears to be somewhat significant. Inferring from the transportation chapter (table 5.1), this report states that energy use would increase 0.9 quads over the AEO97 reference case. CEA contacted Art Andersen at the Energy Information Administration to determine how much of an effect holding fuel efficiency constant would have on the E/GDP ratio.⁷ According to Andersen, constant fuel efficiency would "disappear in the rounding" in the E/GDP ratio. He stated that this assumption would 5 The 1997 AEO projects transportation energy consumption to be 31.39 quads in 2010. With the labs study assumption regarding constant fuel efficiency, the projected energy consumed in this sector should increase. Although the text never explicitly states the extent of this increase, we inferred that it is 0.9 quads. However, since Andersen informed us that the increase would be 0.6 quads, we assessed both in a subsequent comment. 6 The labs study also modified the electricity sector forecast, but that modification is not relevant to our discussion of the energy efficiency of the economy. 7 Art Andersen is the Director of the Energy Demand and Integration Division, EIA. 5 result in energy use increasing by 0.6 quads in 2010 over the reference case. CEA recalculated the E/GDP ratio used in the AEO97 BAU to account for this extra energy use (assuming GDP remains the same across these cases), and found that annual E/GDP would decrease 0.86% under the constant fuel efficiency assumption (0.6 quads case). With the inference from the transportation chapter, the annual E/GDP rate would be -0.85% (0.9 quads case). The labs study BAU assumes fewer energy efficiency improvements than should be expected under these adjusted BAU scenarios. Two related implications arise from this. First, since the study employs a modified version of the AEO97 energy use forecast, it should modify the AEO97 carbon emissions forecast. With a slower improvement in energy efficiency under the lab study assumptions, the 2010 BAU carbon emissions should be greater than the 1722.4 MMTCE in AEO97 (AEO97, p. 120). As we note in the previous comment, by incorporating the AEO97 reference case carbon emissions for buildings and industry, the BAU should be about 1740 MMTCE. This obviously implies that more reductions will be necessary to stabilize emissions at the 1990 level in 2010. Second, the carbon reductions calculated for the various scenarios in this study are based on this -0.77% E/GDP BAU. This implies that some of the carbon reductions claimed under the efficiency and high efficiency cases actually occur in the -0.86% BAU. This double-counting could be significant. The CEA-generated BAU (0.6 quads case) accounts for 17% of the E/GDP gains assumed in the efficiency case (see attached chart).⁸ The alternative CEA-generated BAU (0.9 quads case) accounts for 15% of the E/GDP gains assumed in the efficiency case. Assuming that E/GDP and carbon reductions are perfectly correlated (given that the efficiency case involves no fuel switching, this is reasonable), then the efficiency case overestimates emissions reductions by 18 to 20 MMTCE. The text should explain why the lab study E/GDP ratio differs from the AEO97 ratio since it does not seem to result exclusively from keeping fuel efficiency constant. Further, the 2010 BAU emissions reductions should reflect the E/GDP ratio (whether it is -0.77%, -0.85%, -0.86%, or some other rate) used in the study. 10) 1.11, Table 1.4: It would be valuable to understand the effect of non-price policies and permit prices on the carbon reductions. Could you break down the high efficiency cases into two categories: non-price policy induced and permit induced technology adoption? 11) 1.12, paragraph 1: The discussion on government programs costing 15% of investment costs should include references to the literature that have estimated this percentage. Does the literature describe the marginal costs for government programs, or does it just provide this average cost value? Consistent with the literature on technology adoption, the marginal costs should not be 8 The AEO97 BAU would capture 25% of the efficiency gains in the efficiency scenario. 6 assumed to be the same for all units of adoption. In fact, it is likely that the marginal costs increase for higher rates of adoption. 12) 1.12, paragraph 1: The text in this paragraph notes that government costs in the best estimate case are 15% of total investment costs, while figure A-1.1 on p. A-1.2 indicates that the costs are 7% of investment costs for the end-use sectors and 1% for the electricity sector. Which costs are correct? 13) 1.12, paragraph 1: The text notes that "one could argue that a social discount rate" of 3% or 7% should be used. Alternatively, one could argue that since these analyses assess private decisions on technology adoption, that private marginal rates of time preference should be used that reflect "current market behavior". This analysis is not attempting to conduct a social benefit-cost analysis. Rather, it is attempting to model the private individual or firm's benefits and costs associated with a technology adoption decision. Since this report projects technology adoption assuming a set of aggressive policies, the rates of time preference actually used by individuals and firms should be used in assessing the extent of adoption. Further, the report should not rely on government fiat to lower discount rates. It is very difficult to lower people's time preferences through policy. Government programs do not increase adoption rates by lowering time preferences, but by lowering the costs of adoption or increasing the benefits of adoption. For example, government programs cost-share the adoption of environmentally-benign agricultural production techniques through the Environmental Quality Incentives Program. Farmers still employ the same discount rate, but their stream of costs are lower because of the government subsidy, and so adoption increases. The report should not assume that government policies can lower private time preference rates, but rather that government policies can affect the stream of costs and benefits of an adoption decision. 14) 1.12, paragraph 5: Again, until the complete costs of investment and accurate cost-savings can be estimated, the text on net benefits should be deleted. 15) 1.13, Table 1.5: The estimates of carbon reductions under the alternative view do not appear to match with the carbon reductions in table A-1.2 on p. A-1.5. It seems implausible that discount rates could double and have no effects on any technology adoption decision in this report. Are the numbers in the appendix the right set? Is it true that every single energy efficient technology adoption decision is cost-effective under the "best estimate" discount rate and the "alternative view" discount rate? If these technologies are such big winners, then why aren't people and firms already adopting them? 16) 1.14: Unless and until the costs and cost-savings issues are adequately addressed, estimates of net savings should be omitted. 17) 1.15, paragraph 3: As stated in the executive summary comments, the text should not state "energy efficiency alone can take the nation 30 to 50% of the way to 1990 levels." If energy 7 efficiency could achieve these reductions "alone", there would be no need for an invigorated federal policy effort. It should be explicitly stated that policy instruments to achieve this goal are omitted from the analysis. Chapter 2 18) 2.6, transportation sector bullet: The text states that the transportation sector analysis uses a 5 year time horizon. However, the description of the best estimate and alternative view cost effectiveness calculations on p. A-1.10 indicate that a time horizon of 14 years was used. Further, the transportation chapter indicates that the cost effectiveness analyses were conducted using a 6% discount rate in combination with a decline in usage and depreciation to calculate fuel savings (p. 5.45). The footnote to table 5.9 implies that the analyses used a 14 year time horizon as well. The report should clarify these inconsistencies. Chapters 3-7 19) There does not appear to be any discussion of the $25/ton permit scenario in these chapters. Further, the appendices do not explicitly detail this case. The results from this scenario need to be documented in the text and in the appendices. For example, it is impossible to identify the penetration rate for this scenario in the buildings chapter (see p. 3.3). The industry chapter discusses three scenarios: BAU, efficiency, and high efficiency (p. 4.2). The descriptions of the high efficiency case in transportation on p. 5.3 and p. 5.31 do not even mention a permit price. The electricity sector appears to only consider a $50/ton permit (p. 6.9). 20) The results from the alternative view cost analysis should be incorporated in these chapters. Chapter 3 21) 3.3: In the context of fully incorporating the effects of climate policy on energy prices, the analyses on buildings technologies should be modified. Under the efficiency case, energy consumption declines. This scenario should be changed to reflect the downward effect on prices resulting from this decreased energy consumption. The result would be to reduce the net improvement in energy use resulting from technological advances. The high efficiency scenario should reflect two counteracting effects on prices: the decline in energy consumption and the $50/ton permit fee. The study ignores the former effect and insufficiently incorporates the latter effect. For example, in the buildings chapter, the penetration rate is assumed to be 65% instead of 60% of the maximum cost-effective technical potential in this scenario because of the $50/ton permit price. However, the energy-cost savings calculations for buildings technologies assume the same energy price as in the business as usual scenario. Since the study assumes that exogenous, undefined policy influences drive the penetration rates, the penetration rate should be set at 60% and the energy cost-savings should be recalculated with the appropriate energy price. In addition, the assumed energy use for these technologies should be based on consumers' responses to the lower operating costs. For example, if running an air conditioner becomes less 8 expensive with a new energy efficient technology, people will run the air conditioner more. This latter effect decreases the carbon reductions associated with each adoption decision. 22) 3.9 and Appendix C-1: The study uses the average cost of electricity to calculate the energy cost savings of these technologies. Many of the electricity-dependent technologies are only cost effective based on the assumption of average, not marginal electricity prices. Since marginal prices may be much lower than average prices, especially in cases where households use less electricity, this may overestimate the extent of cost effective technologies. The analysis should be modified using marginal electricity prices. 23) 3.23, box: The text in the box is not clear on the inclusion of fuel cell technology in the high efficiency scenarios. The first paragraph states that fuel cells were not included in the main building sector scenarios. However, emissions reductions of 3 MMTCE from fuel cells are listed under buildings in Table 1.4 on p. 1.11. Further, the discussion on fuel cells in the text outside of this box implies that fuel cell technology will not be available until 2020 (since it falls in the section "Potential for Advanced Technologies in 2020"). If the study relies on ADL's analysis of fuel cell adoption, then the assumptions of ADL's assessment should be provided in an appendix. Two issues should be addressed here. First, if fuel cells are projected to be adopted under the high efficiency scenarios, then this discussion should be placed in the appropriate section of this chapter (not the 2020 section), and the inconsistencies in the text should be remedied. Second, if fuel cells are adopted, then a discussion of the carbon emissions accounting should be provided to ensure that the study does not double count emissions reductions from energy efficient technology adoption and from switching energy sources to fuel cells. 24) 3.29, summary bullet two: This states that the high efficiency scenario yields 91 MMTCE of carbon reductions from the BAU emissions level. However, Table 1.4 indicates that only 44 MMTCE of reductions occur under the $25/ton permit price and only 62 MMTCE of reductions occur under the $50/ton permit price. Which carbon reduction estimate is correct? Chapter 4 25) 4.6, paragraph 4: The analysis assumes that the capital recovery factor falls by more than half. This appears to be an arbitrary reduction. Did industry behave as if it operated under a lower CRF during the 1970s oil shocks? There is no discussion of the economic behavior necessary to result in this outcome. 26) 4.6, paragraph 4: The high efficiency penetration rate is assumed to be double the rate used in the efficiency scenario, which is the undefined "normal" rate. It is impossible to determine how much of this doubling of penetration results from undefined, aggressive policy efforts and how much results from the permit price. It would be valuable to understand the relative impacts of both. 9 27) 4.13, paragraph 4: The text states that the high efficiency case causes an acceleration of capital retirement, but does not estimate these costs. Could you estimate the costs of early retirement? If not, can you provide some information on the extent of early retirement (e.g., percentage of capital retired early under this scenario)? Chapter 5 28) Does this sectoral analysis consider the impact of a permit price? Neither the $25/ton permit nor the $50/ton permit are mentioned in the entire chapter. Do all improvements in fuel efficiency and cellulosic ethanol result from non-price policy responses? As the report states, a degree of "luck" is required to achieve the necessary technological breakthroughs to result in the projected carbon reductions. However, without a price incentive, the probability of the nation becoming "lucky" appears more unlikely. If this analysis does not incorporate the permit prices, then this should be made explicit in the discussion of the high efficiency case. 29) The methodology for this sector is insufficiently transparent to grasp easily the impact of technology development and adoption on carbon reductions. Further, the appendices for transportation do not provide the details to remedy the shortcomings in the chapter. It is difficult to compute carbon reductions from increases in fuel efficiency. The text should take the reader through the process of piecing together all of these energy efficient technologies, illustrating the final effect on fuel economy, and then translating this into carbon reductions. 30) 5.4, Table 5.1: Why do the energy consumption numbers for this sector in 1997 vary across scenarios? We understand that the start year for the scenario analyses is 1998 (e.g., refer to Table 4.8 on p. 4.12 that implies 1998 as the start year for assessing cumulative incremental investment). 31) 5.21, paragraph 2: Does the assumption about reducing technology costs reflect any empirical analysis on transportation technologies? For example, the 1970s oil shocks and fuel efficient imports spurred domestic fuel efficiency R&D. Does this analysis incorporate information about accelerating technology development and lowering technology costs that may be available from this earlier period? 32) 5.23, paragraph 3: As we have noted before, it is difficult to modify consumers' preferences. Assuming that the demand for horsepower will decrease because consumers will become "more green" over the next decade appears tenuous. This statement is repeated on p. 5.32. Is there any evidence that concern about global warming has affected consumers' purchase decisions to date? 33) 5.25, paragraph 1: If the turbocompound diesel engine and the advanced LE-55 heat engine are not available in the reference case (which implies that the earliest these are available is 2016), then how are they made available by 2003? Is there any precedent for accelerating transportation technology from a time horizon of at least 18 years to a horizon of only 5 years? How do these 10 engines conform to present emissions standards for particulates and NOₓ and how will future standards under the new ozone and particulate matter rules affect them? 34) 5.25, paragraph 1: The efficiency case assumes that advanced drag reduction in heavy trucks has become available this year, but this is excluded from the reference case. First, has this actually occurred? Second, if it has already occurred, then why is it excluded from BAU? 35) 5.25, paragraph 4: Again, the assumption of reducing a technology price appears arbitrary. What is the basis for assuming that the trigger price for heavy truck technologies will fall? 36) 5.25, paragraph 5: The efficiency analysis assumes penetration rates of 100% for several of the heavy truck technologies over a 20 year period. This seems incongruous with the assumptions in the other sector analyses that imply much lower penetration rates. 100% penetration of a technology within 20 years, that is not expected to be available 18 years from now in the BAU, appears to be an extreme and unsubstantiated assumption. 37) 5.32, paragraph 1: The analysis assumes that fuel cells, because their costs are unknown, will be cost effective. This is an entirely arbitrary assumption, especially for a technology that is not even projected to become available in the BAU scenario. There is absolutely no economic basis for this assumption. This assumption is all the more extraordinary given that this analysis does not (apparently) incorporate the impacts of a carbon permit on fuel prices. 38) 5.36, Table 5.7: If the scenarios overestimate fuel economy, as stated in the footnote, then the numbers should be revised to remedy this error. 39) 5.38, paragraph 2: The modification of assumed improvement of fuel efficiency by multiplying the expected improvement by 0.7 to reflect offsetting performance effects, appears arbitrary. Why wouldn't the fraction be much smaller? The introduction to this chapter indicates that almost all improvements in fuel efficiency are offset by performance effects. Why would only 30% of fuel efficiency improvements in the future be offset? 40) 5.45: The chapter is not explicit about its accounting of fuel efficiency improvements in light duty vehicles and the change in the fuel mix reflecting an increase in cellulosic ethanol and the use of fuel cell technology. We cannot determine if carbon reductions are double counted. The study does not discuss how cellulosic ethanol and fuel cells are accounted for in the energy cost savings description on pp. 5.43-5.45. If the reductions from fuel efficiency and the reductions from cellulosic ethanol and fuel cells do not overlap, then this should be stated explicitly. Again, a more transparent presentation of the methodology would address the readers' uncertainty about this issue. 41) 5.45, Table 5.9: The cost effectiveness estimates assume a constant price of $1.20 per gallon through the life of the vehicles. The gasoline price should be adjusted down to reflect the impact of decreased gasoline consumption due to fuel efficiency. Further, vehicle miles traveled should 11 increase through time as cars become more fuel efficient, consistent with the trend to date, instead of held constant. Chapter 7 42) 7.1, paragraph 2: If the report has eliminated the double counting in this sector evident in the June 10 draft, then the sentence that states that double counting is "a likely possibility" should be deleted. A reference to appendix G-2 should be inserted. 43) 7.2, paragraph 3: Again, if the double counting has been remedied by integrating the dispatching and repowering analyses, then the following sentence should be modified: "The analytical approach was static in that the cost of repowering was computed for each candidate power plant but the analysis did not optimize unit/plant production cost, dispatch, or system load." 44) 7.6, paragraph 5: The reference to appendix G-2 appears to be wrong. This reference implies that appendix G-2 provides a methodology for SO₂ and NOₓ benefits. Nothing in the text of chapter 7 reads as if the double counting issue from the June 10 draft has been resolved. Appendices 45) A-1.9: The alternative view cost-effective estimates for buildings assumes an 18 year time horizon. However, the methodology discussion on p. 2.5 states that a "technology is defined as 'cost-effective' if it delivers a good or service at equal or lower life-cycle costs relative to the current practice". If cost-effectiveness is measured in terms of a technology's life cycle, then shouldn't the time horizon for calculating the stream of benefits be constrained by the life of the product? It is not possible for a technology to generate cost savings beyond its lifetime. In the buildings case, the end uses with the largest potential carbon reductions in the high efficiency case are other uses (10 MMTCE commercial, 6 MMTCE residential) and lighting (7 MMTCE commercial, 6 MMTCE residential). These account for 29 of the 62 MMTCE of reductions for this sector. However, the lifetimes for these end uses are much shorter than 18 years. For other uses, the lifetime ranges from 7 (commercial) to 10 years (residential). For lighting, the lifetime ranges from 1 (residential) to 12 years (commercial). The cost effectiveness estimates should be recalculated using specific end use lifetimes. 12 Summary of Greenhouse Gas Emissions-Reduction Actions Million Metric Tons of Carbon Equivalent Source: U.S. Climate Action Report-1997 Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date Residential & Commercial Sector Actions 26.9 10.3 I New Rebuild America 2.0 1.6 -- 1 and 2 Expanded Green Lights and Energy 3.6 3.3 -- Star Buildings 3 State Revolving Fund for Public 1.1 Terminated Buildings 4 Cost-Shared Demonstrations of Emerging Technologies 5 Operation and Maintenance Training 3.8 0.0 -- for Commercial Building Facility Managers and Operators 6 Energy Star Products 5.0 4.3 : 7 Residential Appliance Standards 6.8 0.2 : 8 and 11 Energy Partnerships for Affordable Housing 9 Cool Communities 4.4 0.4 -- 10 Update State Building Codes New Construction of Energy-Efficient 0.1 : Commercial and Industrial Buildings New Superwindow Collaborative 0.0 -- New Expand Markets for Next-Generation 0.2 : Lighting Products New Fuel Cells Initiative 0.0 -- Industrial Sector Actions 19.0 4.8 -- 12 Motor Challenge 8.8 1.8 : 13 Industrial Golden Carrot Programs 2.9 Merged into Action 12 14 Accelerate the Adoption of Energy- Terminated Efficient Process Technologies 15 Industrial Assessment Centers 0.5 CCAP Component Terminated 16 Waste Minimization 4.2 2.1 : 17 Improve Efficiency of Fertilizer 2.7 0.8 : Nitrogen Use 18 Reduce the Use of Pesticides Terminated Transportation Sector Actions 8.1 5.3 : 19 Cash Value of Parking 20 Innovative Transportation Strategies 6.6 4.6 : 21 Telecommuting Program 22 Fuel Economy Labels for Tires 1.5 0.7 -- Energy Supply Actions 10.8 1.3 - 23 Increase Natural Gas Share of Energy Use Though Federal Regulatory 2.2 Terminated Reform Action Action Title 1993 Action 1997 U.S. CAR Actual Number Plan Estimate Revised Estimate Reductions for 2000 for 2000 to Date 24 Promote Seasonal Gas Use for Control 2.8 0.5 : of Nitrogen Oxides 25 High-Efficiency Gas Technologies 0.6 Terminated 26 Renewable-Energy Commercialization 0.8 0.3 -- 27 Expand Utility Integrated Resource 1.4 Terminated Planning 28 Profitable Hydroelectric Efficiency 2.0 0.0 -- Upgrades 29 Energy-Efficient Distribution Transformer Standards 0.8 0.5 -- 30 Energy Star Distribution Transformers 31 Transmission Pricing Reform 0.8 Terminated New Green Power Network Not included 0.0 -- Land-Use Change & Forestry Actions 10.0 2.4 - 43 Private Depletion of Nonindustrial 4.0 Terminated Private Forests 44 Accelerate Tree Planting in 0.5 0.4 -- Nonindustrial Private Forests 16 Waste Minimization 4.2 2.0 -- 9 Expand Cool Communities 0.5 To be determined Methane Actions 16.3 15.5 -- 32 Expand Natural Gas STAR 3.0 3.4 -- 33 Increase Stringency of Landfill Rule 4.2 6.3 -- 34 Landfill Methane Outreach Program 1.1 1.9 -- 35 Coalbed Methane Outreach Program 2.2 2.6 -- 36 RD&D for Coal Mine Methane 1.5 Terminated 37 RD&D for Landfill Methane 1.0 Terminated 38 AgSTAR Program 1.5 0.3 -- 39 Ruminant Livestock Efficiency 1.8 1.0 -- Program Actions to Address Other Greenhouse Gases 16.3 25.4 I 17 Improved Fertilizer Management 4.5 5.3 -- 40 Significant New Alternatives Program 5.0 6.4 -- 41 HFC-23 Partnerships 5.0 5.0 -- 42 Voluntary Aluminum Partnership 1.8 2.2 : New Environmental Stewardship Initiative Not included 6.5 -- Foundation Actions 11.3 - Climate Wise Not estimated 1.8 -- Climate Challenge Not estimated 7.6 -- State and Local Outreach Programs Not estimated 1.9 -- Total GHG Emission Reductions From CCAP 108.6 76.0 14.0 Data is not readily available for cumulative emissions reductions for many CCAP programs. Emissions reductions of about 5 MMTCE can be attributed to DOE's CCAP programs. EPA's Office of Air and Radiation is responsible for emissions reductions of about 9 MMTCE through their CCAP programs. 1.05 Improvements in Efficiency, 1997-2010 Under Alternative No Efficiency Annual E/GDP Rates BAU-CEA (0.6) 1 * BAU-CEA (0.9) 0.95 BAU-DOE5 Efficiency Case 0.9 18-20 MMTCE 120 MMTCE 0.85 All efficiency values are shown relative to a 1997 E/GDP ratio indexed to 1. 0.8 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 08/20/97 WED 11:41 FAX 202 6222633 4. 001 Office of Economic Policy Department of the Treasury Washington, D.C. 20220 FAX Date: August 20, 1997 at 11:38 AM Number of pages including cover sheet: 8 Name FAX Number Phone Number To: JAY SHOGREN 395-6809 JOE ALDY From: Robert Gillingham 202-622-2633 202-622-2220 REMARKS: Urgent X For your review Reply ASAP Please comment See what you think. 08/20/97 WED 11:42 FAX 202 6222633 002 COMMENTS ON SCENARIOS OF U.S. CARBON REDUCTIONS Scenarios of U.S. Carbon Reductions, commonly known as the "5-labs study," examines the "potential for energy-efficient and low-carbon technologies to reduce carbon emissions in the United States." This study does a very nice job of cataloging the current state and likely future path of technological developments. What it does not do, however, is shed much light on what role technology is likely to play in mitigating the growth of carbon emissions, or on what government can or should do to enhance this role. As a tool for developing and implementing a technology policy, the report is hampered by the fact that it does not address any of the processes that determine the rate at which technology is developed and implemented. Rather, it catalogs technologies that currently exist, are likely to exist or could possibly exist by 2010, determines whether they are "cost-effective," and then makes somewhat arbitrary assumptions about the rate at which they are implemented. Although the analysis is couched in terms of the adoption of cost-effective technologies, cost-effectiveness is not rigorously defined; no evidence is presented on whether decision-makers are or are not acting in their best interest. It is important to recognize that the paper does not demonstrate an irrational unwillingness to invest in energy efficiency. The paper suggests that the adoption of standards and codes is necessary to ensure appropriate consumer behavior; again, no such demonstration is made. In other words, the study does not and cannot present evidence on the costs to either society or individual consumers and businesses of alternative technology paths. The study reaches three overarching conclusions: A vigorous national commitment to develop and deploy cost-effective energy- efficient and low-carbon technologies has the potential to substantially reduce energy consumption and carbon emissions. We would not argue with this conclusion. Rather we question (1) the level of government involvement-government R&D, tax incentives, efficiency standards, etc.-that would be necessary in the absence of energy-price increases and (2) whether the substantial increases in relative energy prices that would likely accompany constraints on carbon emissions would not be the most important component of any national commitment to reduce emissions. Carbon emissions can be reduced in ways that reduce energy costs more than they increase other societal costs. We question this assertion. The study makes only rudimentary cost-benefit calculations that ignore important considerations that reasonably effect decisions to invest in energy efficiency. It also substantially understates the costs of government policies to promote technology. The next generation of energy-efficient and low-carbon technologies promises to 08/20/97 WED 11:43 FAX 202 6222633 003 enable the continuation of an aggressive pace of cost-effective carbon reductions over the next quarter century. Again, absent a careful cost-benefit analysis, this conclusion is overly speculative. For instance, with respect to the transportation sector, the report concedes that future technological breakthroughs will be necessary to develop and implement the high-efficiency, low-carbon technologies. It is a stretch to assume that these breakthroughs will be forthcoming and that they will pass rigorous cost-benefit tests, especially in the absence of strong price signals. The methodology used in the report is to analyze sequentially the three major end-users of energy-the residential and commercial building, industry and transportation sectors-and separately analyze the electric utility sector. For each sector, the study develops three scenarios-business-as-usual, efficiency, and high- efficiency/low carbon. They are conceptually defined as follows: Business-as-usual (BAU). Best estimate of future energy use given current trends in service demand, stock turnover and natural progress in the efficiency of new equipment. Assumes no change in federal policies.¹ Efficiency (EFF). Likely market penetration of "cost-effective," energy-efficient technologies given "an invigorated effort to promote energy efficiency through enhanced public and private-sector R&D and market transformation activities." High-efficiency/low-carbon (HE/LC). Optimistic but feasible market penetration for energy efficient and low-carbon technologies given a "greater commitment to reduce carbon emissions resulting from actions that might include the creation of a market value for carbon of $25 and $50 per tonne." With the exception of the range of carbon emission charges, which seems out of place and is largely ignored in the study, these conceptual definitions are very vague. The operational definitions vary by sector and will be discussed in turn below. BUILDINGS For buildings, the business-as-usual scenario is the Annual Energy Outlook (AEO) reference case. The efficiency scenario assumes that 35 percent of the "cost- effective energy savings potential" relative to the BAU is achieved. The high- efficiency/low-carbon scenario assumes that 65 percent of these savings are achieved. We have a number of problems with this analysis. The cost-effective energy saving potential is badly defined. The study makes two critical assumptions in defining cost effectiveness. First, the analysis takes as given 1 The report provides relatively little analysis of past trends and their determinants. The past has certainly informed our view of the future, and a more detailed analysis of the price and quantity paths, both past and in the BAU scenario, would be very useful. 2 08/20/97 WED 11:43 FAX 202 6222633 004 that engineering estimates of energy use for alternative technologies are known with certainty and independent of the age of the capital good in which the technology is embodied. There is evidence that adopting these estimates is likely to overestimate savings (e.g., attic insulation may not be perfectly installed; consumers may increase usage rates for energy-efficient appliances or consumers may add more windows when their energy costs are reduced). Second, the analysis ignores the trends and uncertainty around these trends in the future prices of both capital goods and energy as well as the potential irreversibility of investment decisions (proper treatment of these issues goes a long way toward explaining high measured internal rates of return, which are often confused with high discount rates). As a result of these two assumptions, the study almost certainly exaggerates the "cost- effectiveness" of investments in energy efficiency. The assumed take-up rates of 35 percent and 65 percent are arbitrary. The study makes no attempt to model behavior; rather it makes arbitrary assumptions about the rate at which energy-saving investments are made. With no idea what determines the rate at which new technologies are adopted, we have no basis for informed policy-making. The HE/LC case includes substantial price increases in energy, yet the study simply raises the take-up rate for technologies deemed cost- effective under a different-and much lower-scenario of energy prices. The assumed take-up rate under the EFF case is bad enough, but the failure to explain how a carbon price would change the decision-making calculus in the HE/LC case is even worse. A number of the examples are based on 1997 technologies. For instance, the refrigerator example assumes adoption of an energy-efficient 1995 model. This ignores the fact that refrigerator efficiency has been improving at a rapid rate for over 20 years (6.2 percent per year from 1972 to 1991). What does the AEO assume for the 1995 to 2010 period? Is it reasonable to assume that the type of saving assumed in the EFF case should be incorporated in the BAU case? Or is the BAU unnecessarily pessimistic? The bulk of the savings in buildings are largely unexplained, coming from "miscellaneous electricity uses." The study admits that because of the importance of these uses, "it is crucial that more research be carried out, both to characterize how energy in used in the miscellaneous category and to identify technologies for improving the efficiency of sub-categories." With no clear understanding of these fundamental aspects of electricity use, how can we be confident to assume that this 2 See, e.g., Gilbert Metcalf and Kevin Hassett, "Measuring the Energy Savings from Home Improvement Investments: Evidence from Monthly Billing Data," National Bureau of Economic Research Working Paper 6074, June 1997. 3 See Robert Pindyck, "Irreversibility, Uncertainty, and Investment," Journal of Economic Literature 29, 1991, for a review of some of these issues. Also, see Gilbert Metcalf and Donald Rosenthal, "The 'New' View of Investment Decisions and Public Policy Analysis: An Application to Green Lights and Refrigerators," Journal of Policy Analysis and Management, Vol. 14, No. 4, 1995, for an application. 3 08/20/97 WED 11:44 FAX 202 6222633 005 category accounts for most of the presumed savings? INDUSTRY For the industrial sector, the study relies on the Long-Term Industrial Energy Forecasting (LIEF) model, an econometric forecasting model, to explain investment in conservation technology; it uses the industrial module of the National Energy Modeling System (NEMS), which does not model investment, to assess the effect of that investment on energy use. The business-as-usual scenario is the Annual Energy Outlook (AEO) reference case. The efficiency scenario assumes that the "capital recovery factor" (CRF) for energy-efficient investment in the LIEF is reduced from 33 percent to 15 percent.4 The high-efficiency/low-carbon scenario assumes, in addition, that the penetration rates for energy-efficient technology are doubled relative to the BAU scenario. This analysis suffers from problems similar to those in the analysis of the buildings sector. The study models investment in the EFF and HE/LC scenarios by arbitrarily assuming that the return criterion for energy-efficient investment is reduced by more than 50 percent. No rationale is provided for this approach, nor are any policies posited that would lead to this result. Rather, it is simply a device to increase investment. As noted above, arbitrary assumptions about hurdle rates beg critical questions about the effect of uncertainty, changing relative prices and irreversibility on investment behavior. The approach used provides no insights into investment behavior; it implicitly assumes that behavior. The HE/LC scenario depends on an arbitrary assumption about penetration rates and ignores price effects. The LIEF model "provides a mechanism for evaluating general investment in conservation technology as a function of energy prices, capital recovery rates, and other prices." Yet in evaluating the HE/LC scenario, the study ignores the postulated $25 and $50 carbon prices that are part of this scenario. Instead it assumes a doubling of penetration rates for new technology and retirement rates for old capital. The study makes no attempt to demonstrate that an increase in penetration/retirement rates is economically motivated. Although no policies are modeled or advocated, the study lists a number of potentially costly and/or intrusive policies that could promote energy efficiency. These include accelerated depreciation, rebates or tax credits, regulation and efficiency standards, pricing and fiscal policies, and "other economic incentive programs." TRANSPORTATION For the transportation sector, the business-as-usual scenario is the Annual 4 A $50 carbon fee would certainly increase investment in energy efficiency, as does the drop in the CRF. The latter approach is not the right way to model expected relative price increases, however. Again, uncertainty and irreversibility should be explicitly modeled, making it clear that a high apparent hurdle rate is not the same as a high discount rate. 4 08/20/97 WED 11:45 FAX 202 6222633 006 Energy Outlook (AEO) reference case, adjusted to reduce the assumed improvement in light-duty vehicle fuel economy. The efficiency scenario assumes earlier introduction of advanced fuel economy technology and adds certain key technologies that are not in the BAU. The high-efficiency/low-carbon scenario postulates future "breakthroughs" in technology. As with the other sectors, this analysis provides few insights for technology policy. The efficiency scenario assumes accelerated adoption of existing technologies, but provides no reason why this might occur. Unlike the analysis of the other sectors, the analysis of the transportation sector does not even include any economic assumptions that might imply its conclusions. As a result, it promises no medium or longer-term gains. The HE/LC scenario is even more ephemeral; it requires technological breakthroughs, primarily in fuel-cell technology. As the report notes, the EFF and HE/LC scenarios "differ from each other less in effort than in outcome. In contrast, because the outcomes postulated in the high-efficiency/low-carbon scenario require technological breakthroughs, they require a certain degree of luck to be achieved by 2010." In other words, we do not need to try any harder to achieve the HE/LC scenario; we just have to be lucky. The analysis raises the AEO energy baseline, but not the emissions baseline. Since it raises the AEO energy baseline, the study should also adjust the emissions baseline. As a result, the savings postulated relative to the AEO baseline would be reduced. ELECTRICITY The analysis of the electricity sector describes how the electricity demands inherent in the BAU, EFF and HE/LC scenarios can be met with lower carbon emissions. The BAU scenario is a fully competitive bulk-power market in the year 2010. The demands on this system are then adjusted to reflect the effects on electricity demand and supply of the EFF and HE/LC scenarios. These results are used in the sector analyses to evaluate carbon savings attributable to reductions in electricity consumption. They are also used to evaluate the opportunities for reducing carbon emissions in electricity generation in response to $25 and $50 carbon prices. The EFF and HE/LC scenarios predict substantial reductions in electricity demand from baseline. The two scenarios predict reductions of 9 and 16 percent, respectively, in end-use electricity demand relative to BAU. In fact, they assume essentially zero growth in electricity demand between 1997 and 2010. To put this into perspective, electricity sales increased at an annual rate of 2.0 percent annual rate so far in the 90s and at more than a 3 percent rate over the past 25 years. IMPLICATIONS OF THE STUDY As the above discussion demonstrates, the study is primarily a catalog of 5 08/20/97 WED 11:46 FAX 202 6222633 1 007 potential improvements in energy efficiency. Neither the determinants of these improvements, nor their implications for energy demand behavior are examined. Rather, the study provides an accounting of two very stylized scenarios. Without developing and detailing the relationship between the study's scenarios and the policy mixes that would have to accompany them, it would be inappropriate to draw any conclusions from the study about the role of technology, much less the role of technology policy, in addressing global warming. Although the study provides no policy recommendations, it does hint at possibilities. Listing them provides some insight into the potential difficulties in making the scenarios "come true." Aggressive R&D, both public and private. The study continually refers to the need for aggressive R&D, but provides no suggestions as to how it could be fostered and at what cost.5 The EFF scenario is basically an extrapolation of the Climate Change Action Plan; the likely success of this effort is at best problematic. Some have suggested a more vigorous effort, similar to the Apollo project. Although this sounds good, we should remember that the Apollo project required rechanneling 0.4 percent of the entire GDP of the 1960s into this one government project. Many of the climate change modeling efforts predict that reducing emissions levels to 1990 levels in 2010 would not cost much more than that in the absence of a technology program. Tax incentives. Investment tax credits, accelerated depreciation, and other tax incentives are available to spur the development and adoption of energy-saving technology. This approach has two problems. First, energy-efficiency investments should not be singled out unless they impose special costs. If they do, disincentives for using energy are more efficient than incentives for saving energy. Second, any substantial incentives would have to be "paid for" with other tax increases and spending cuts. As recent history has demonstrated, reaching consensus is not easy. Regulations and standards. Implicit in this study is the idea that regulations and standards could go a long way toward achieving the EFF and HE/LC scenarios. The types of standards that would be needed, however, would be very pervasive and potentially very intrusive. For instance, standards would have to be applied to building components (windows, insulation, etc.) and constructions techniques, consumer and commercial appliances, utility and industrial equipment, etc. If standard setting is a major policy instrument, achieving the standards could be very costly. Standards work best if they impose little cost, as the report alleges they often would. Basing policy on this assumption moves us quickly to the idea that better results depend less on effort than on luck. 5 The study (arbitrarily) assumes that government costs to promote energy conservation are equal to 7 percent of energy savings. This is very optimistic. For instance, a 10 percent tax credit would cost 10 percent of the total cost of all qualifying investment, even that which would have taken place in the absence of the credit. The load factor for explicit government subsidies would be roughly 10 times as large. 6 08/20/97 WED 11:47 FAX 202 6222633 008 Higher prices. The most obvious policy to spur energy conservation is to increase energy prices. The study presumes a substantial increase in prices in its HE/LC scenario, but never draws out the implications of higher prices for the development and adoption of advanced technology. The study points out how rapidly energy efficiency improves prior to the late 80s and 90s. This is not surprising, given price trends. As the chart below demonstrates, consumer energy prices, relative to other prices, grew very rapidly during the 70s and into the 80s. A commitment to reduce emissions and internal climate costs would presumably result in substantial increase in relative prices and accompanying increases in efficiency. RELATIVE PRICES OF CONSUMER ENERGY 220 Electricity 200 Natural Gas Gasoline 180 Relative Price (1973=100) 160 140 120 100 80 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 Under almost any feasible scenario that reduces carbon emissions, the pace of technological advance in energy conservation will increase. The importance and role of technology are not in question. What is in question is the value of looking at technology in a vacuum. The 5-labs study provides a plethora of useful information. It does not present this information in a form that is useful for policymaking, however. To be useful, it needs to be merged with careful analysis of the determinants of both consumer and business investment in energy-saving technology. 7 Review of DOE Labs Study Unspecified Policies: No Program Descriptions: The authors do not estimate the costs or even describe the programs necessary to stimulate the adoption of the new energy-efficient technology they find could reduce emissions of approximately 200 million metric tons of carbon equivalent (MMTCE) by 2010. These reductions correspond to about ½ of the reductions needed to attain 1990 levels by 2010. In the absence of cost estimates this information is of little use in the evaluation of alternative policy options. Costs of Programs: Given reasonable assumptions, government programs may cost many tens of billions of dollars per year above the costs of existing programs to reduce 200+ million tons of carbon. The marginal cost of emissions reductions, including the cost of the government program, is likely to greatly exceed the permit price. As a result these programs may be less cost-effective than a cap-and-trade emissions control program. Luck: The transportation sector requires "luck" in terms of technological innovation to achieve the projected emissions reductions. Without a permit price, this luck appears likely only through stringent standards, such as increasing CAFE. Overestimates Reductions/Underestimates Costs: Behavioral Changes Ignored: The analysis appears to ignore the behavioral changes that offset the energy efficiencies promised by the new technologies. For example, more efficient windows lead to houses with more glass, and more efficient cars lead to more vehicle miles traveled. Thus these innovations offer real value to consumers, but smaller energy savings than engineers forecast. Cost Effectiveness Calculated Using Average, Not Marginal Prices: Many of the innovations in electrical use appear to be assumed to be cost effective based on an assumption of average, not marginal electricity prices. The marginal prices may be half as much as the average prices. Price Declines Resulting from Reductions in Energy Demand Are Not Taken into Account: The study assumes a 12% decrease in energy use, but does not account for this effect on price when calculating cost effectiveness for energy efficient technologies. Study Double Counts Emissions Reductions Occurring Under Business-As-Usual: The labs study double counts 22 MMTCE in reductions in the efficiency case that will occur under the business-as-usual case. Interactions Ignored, Double Counting "Likely" in Utilities Sector Analysis: The double counting of dispatching and retooling coal plants to natural gas results in about 9 MMTCE of overestimated reductions. Industry Sector Analysis Ignores the Costs of Accelerating Retirement. Excerpt from the President's speech at American University, 9/9/97 "And we also need people who have the confidence in our ability to break new technological and scientific barriers to stand up and say, you cannot make me believe that we can't reduce greenhouse gas emissions substantially and still grow the American economy. We could reduce them 20 percent tomorrow with technology that is already available at no cost if we just changed the way we do things." The President's statement reflects the result of a study conducted by five Department of Energy research laboratories titled Scenarios of U.S. Carbon Reductions. This study, still in draft form, claims that carbon emissions can be reduced to 1990 levels in 2010 at "net costs to the U.S. economy [that] are near or below zero in this time frame [1998-2020]" (p. 16, chapter 1, 8/29/97 draft). The Council of Economic Advisers and the Department of Treasury have reviewed this report and found the economic analyses to be inadequate to support this assertion. The study suffers two serious flaws. First, it merely assumes "aggressive" and "invigorated" government policies to stimulate technology adoption, without a discussion of the political feasibility of these policies or whether these technologies would be voluntarily adopted by private parties. The study does not provide reliable estimates of the costs of government programs nor does it estimate the costs of extensive and intrusive new regulations implied, including stringent CAFE standards, national building codes, and very aggressive appliance standards. This country's experience with the Climate Change Action Program implies that the government costs of achieving this goal would be in the billions of dollars annually. Second, the analyses on individuals' and firms' adoption decisions do not assess economic behavior, but rather invoke a set of-engineering assumptions that generate substantial underestimates of complete adoption costs. The study ignores likely individuals' behavioral responses to climate policy (e.g., people drive more miles when cars become more fuel efficient), implementation costs, and the costs of accelerating capital retirement, but assumes that individuals live in model houses, drive model cars, and purchase products manufactured in model factories. The economic literature on technology adoption indicates that the actual net costs of energy efficient technologies are much greater than predicted by engineering studies like this study. Scenarios of U.S. Carbon Reductions is an engineering study, not an economic study. Given these inadequacies, neither CEA nor Treasury has signed off on the report. pre-tyto comment on President's GCC Statement Preliminary Draft MEMORANDUM TO: Jon Gruber FROM: Joe Aldy DATE: 12/8/97 RE: Umbrella Annex I Trading with neither EU nor Eastern European Participation If neither the EU nor Eastern European countries participate in full Annex I permit trading market, the total cost for Annex I countries to comply with a target and timetable increases, but the U.S., Japan, and Canada are all better off with a smaller Annex I "Umbrella" (trading between U.S., Japan, Canada, Australia, and FSU countries). Australia has roughly the same costs (domestic abatement costs plus transfers for emissions reductions abroad) under this umbrella as under full Annex I trading. Eastern Europe actually benefits under a regime where it does not participate in the umbrella but it does sell permits with the EU (a European-only trading regime) relative to full Annex I trading. As the only seller of permits to the EU, Eastern Europe does not have to compete with cheap permits from FSU countries, and thus can sell more permits at higher permit prices than under full Annex I trading. The three major buyers of permits in the Umbrella, U.S., Canada, and Japan, benefit, because as buyers of permits, the loss of a major competitor means they can buy more emissions reductions in EEFSU at lower permit prices. Former Soviet Union countries and the EU are worse under this scenario of two trading blocks (the Umbrella and the European-only trading) than under full Annex I trading. The cost for the EU increases under European-only trade relative to full trading because it cannot take advantage of low cost emissions reductions in FSU countries, such as Russia. Former Soviet Union countries are worse off even though they don't have to compete with Eastern European countries in the sellers market because the EU would be the second largest buyer of permits under full Annex I-wide trading. Losing a major buyer decreases total demand for their permits and thus the price more so than losing competitors contracts the supply and pushes up the permit price. To illustrate the effects of EU non-participation, I assessed country-specific marginal abatement cost curves derived from SGM outputs for the 1990 in 2010 target and timetable. The first chart illustrates the difference in the total economic effects of these two trading regimes. The two charts that follow show permit prices, domestic emissions abatement, purchases and transfers of permits on the Annex I market, and total costs of Annex I trading with EU participation and a trading regime without EU participation. Umbrella: U.S, Canada, Japan, Australia, FSU European-Only Bloc: EU, Eastern Europe Winners with Two Trading Blocs Relative to Full Annex I Trading: U.S, Canada, Japan, Eastern Europe Losers with Two Trading Blocs Relative to Full Annex I Trading: EU, FSU No Difference Between the Trading Blocs: Australia Preliminary Draft Difference in Economic Costs of Trading Regimes with and without EU, Eastern European Participation Country/Region Total Costs under Total Costs under Change in Costs Full Trading Trading without EU, (- means costs (abatement cost + Eastern Europe decrease w/o EU, transfers) (abatement cost + Eastern Europe transfers) participation) U.S. $12.0b $10.2b -$1.8b Australia $0.8b $0.8b 0 Canada $1.9b $1.5b -$0.4b Japan $3.5b $2.9b -$0.6b Former Soviet Union net gain: $12.3b net gain: $9.7b $2.6b Russia¹ net gain: $7.9b net gain: $6.2b $1.7b Eastern Europe net gain: $3.7b net gain: $5.8b -$1.9b EU² $8.1b $10.6b $2.5b ANNEX I, total $10.3b $12.2b $1.9b abatement cost Preliminary Draft Annex I Trading with Full Participation Country/ Permit Domestic Reductions Domestic Capital Total Cost Region Price Reductions Purchased Abatement Flow Abroad Cost U.S. $41.3 177 202 $3.7b -$8.3b $12.0b Australia $41.3 11 15 $0.2b -$0.6b $0.8b Canada $41.3 13 39 $0.3b -$1.6b $1.9b Japan $41.3 21 75 $0.4b -$3.1b $3.5b Former $41.3 168 -382 $3.5b +$15.8b net gain: Soviet $12.3b Union Russia¹ $41.3 109 -248 $2.3b +$10.2b net gain: $7.9b Eastern $41.3 27 -105 $0.6b +$4.3b net gain: Europe $3.7b EU² $41.3 83 156 $1.7b -$6.4b $8.1b Preliminary Draft Annex I Trading without EU or Eastern Europe Participation Country/ Permit Domestic Reductions Domestic Capital Total Cost Region Price Reductions Purchased Abatement Flow Abroad Cost U.S. $33.6 151 228 $2.5b $7.7b $10.2b Australia $33.6 9 17 $0.2b $0.6b $0.8b Canada $33.6 12 40 $0.2b $1.3b $1.5b Japan $33.6 18 78 $0.3b $2.6b $2.9b Former $33.6 149 -363 $2.5b -$12.2b net gain: Soviet $9.7b Union Russia¹ $33.6 97 236 $1.7b -$7.9b net gain: $6.2b Eastern $59.7 40 -118 $1.2b -$7.0b net gain: Europe $5.8b EU² $59.7 121 118 $3.6b $7.0b $10.6b Notes: 1. Russia is calculated as 65% of FSU because its economy in 1995 was 65% the size of the economies of all FSU countries. 2. SGM models the Western Europe region including 19 countries, 4 of which are not in the EU bubble. However, the 15 countries in the bubble comprised about 93% of CO₂ emissions in 1990 of the 19 countries in the modeled region. See the attached list of countries to review the distinctions between the Western Europe region modeled in SGM and the EU bubble. Preliminary Draft EU Bubble Countries: Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom Western Europe in SGM: EU bubble countries plus Iceland (1990: 0.6 MMTCE CO₂) Norway (1990: 14.3 MMTCE CO₂) Switzerland (1990: 11.6 MMTCE CO₂) Turkey (1990: 39.7 MMTCE CO₂) Preliminary Draft MEMORANDUM TO: Jon Gruber FROM: Joe Aldy DATE: 12/8/97 RE: Annex I Trading without EU Participation If the EU does not participate in an Annex I permit trading market, the total cost for Annex I countries to comply with a target and timetable increases, but the U.S., Japan, Australia, and Canada are all better off without EU participation. The EU and Eastern European and Former Soviet Union countries are all made worse off by EU non-participation. The cost for the EU increases if it does not participate because it must achieve all of its emissions reductions at home, instead of in lower cost countries such as Russia. Eastern European and Former Soviet Union countries are worse off because the EU would be the second largest buyer of permits under full Annex I-wide trading. Losing a major buyer decreases total demand for their permits, resulting in EEFSU countries selling fewer permits at lower prices. All other Annex I countries benefit, because as buyers of permits, the loss of a major competitor means they can buy more emissions reductions in EEFSU at lower permit prices. To illustrate the effects of EU non-participation, I assessed country-specific marginal abatement cost curves derived from SGM outputs for the 1990 in 2010 target and timetable. The first chart illustrates the difference in the total economic effects of these two trading regimes. The two charts that follow show permit prices, domestic emissions abatement, purchases and transfers of permits on the Annex I market, and total costs of Annex I trading with EU participation and a trading regime without EU participation. Preliminary Draft Difference in Economic Costs of Trading Regimes with and without EU Participation Country/Region Total Costs under Total Costs under Change in Costs Trading with EU Trading without EU (- means costs (abatement cost + (abatement cost + decrease w/o EU transfers) transfers) participation) U.S. $12.0b $7.6b -$4.4b Australia $0.8b $0.5b -$0.3b Canada $1.9b $1.1b -$0.8 Japan $3.5b $2.0b -$1.5b Eastern Europe net gain: $3.7b net gain: $1.9b $1.8b Former Soviet Union net gain: $12.3b net gain: $6.2b $6.1b Russia¹ net gain: $7.9b net gain: $4.0b $3.9b EU² $8.1b $18.9b $10.8b ANNEX I, total $10.3b $21.9b $11.6b abatement cost Preliminary Draft Annex I Trading with EU Participation Country/ Permit Domestic Reductions Domestic Capital Total Cost Region Price Reductions Purchased Abatement Flow Abroad Cost U.S. $41.3 177 202 $3.7b -$8.3b $12.0b Australia $41.3 11 15 $0.2b -$0.6b $0.8b Canada $41.3 13 39 $0.3b -$1.6b $1.9b Japan $41.3 21 75 $0.4b -$3.1b $3.5b Eastern $41.3 27 -105 $0.6b +$4.3b net gain: Europe $3.7b Former $41.3 168 -382 $3.5b +$15.8b net gain: Soviet $12.3b Union Russia¹ $41.3 109 -248 $2.3b +$10.2b net gain: $7.9b EU² $41.3 83 156 $1.7b -$6.4b $8.1b Preliminary Draft Annex I Trading without EU Participation Country/ Permit Domestic Reductions Domestic Capital Total Cost Region Price Reductions Purchased Abatement Flow Abroad Cost U.S. $23 101 278 $1.2b $6.4b $7.6b Australia $23 7 19 $0.08b $0.4b $0.5b Canada $23 8 44 $0.09b $1.0b $1.1b Japan $23 12 84 $0.1b $1.9b $2.0b Eastern $23 15 -93 $0.2b -$2.1b net gain: Europe $1.9b Former $23 118 -332 $1.4b -$7.6b net gain: Soviet $6.2b Union Russia' $23 77 -216 $0.9b -$4.9b net gain: $4.0b EU² $158 239 0 $18.9b 0 $18.9b Notes: 1. Russia is calculated as 65% of FSU because its economy in 1995 was 65% the size of the economies of all FSU countries. 2. SGM models the Western Europe region including 19 countries, 4 of which are not in the EU bubble. However, the 15 countries in the bubble comprised about 93% of CO₂ emissions in 1990 of the 19 countries in the modeled region. See the attached list of countries to review the distinctions between the Western Europe region modeled in SGM and the EU bubble. Preliminary Draft EU Bubble Countries: Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom Western Europe in SGM: EU bubble countries plus Iceland (1990: 0.6 MMTCE CO₂) Norway (1990: 14.3 MMTCE CO₂) Switzerland (1990: 11.6 MMTCE CO₂) Turkey (1990: 39.7 MMTCE CO₂) T&T: U.S., Japan, Canada: 2% below 1990 Australia: 5% above 1990 FSU, Eastern Europe: 1990 EU: 10% below 1990 Full Annex I Trading Country/ permit domestic reductions domestic transfers total cost Region price reductions purchased abatement abroad cost U.S. $51.6 215 191 $5.5b $9.9b $15.4b Canada $51.6 16 38 $0.4b $2.0b $2.4b Japan $51.6 26 75 $0.6b $3.9b $4.6b Australia $51.6 13 5 $0.3b $0.2b $0.6b FSU $51.6 415 0 $5.2b -$21.4b net gain: $16.2b EU $51.6 105 218 $2.7b $11.2b $13.9b Eastern $51.6 112 0 $0.9b -$5.8b net gain: Europe $4.9b Two Permit Trading Blocs: Annex I Umbrella and Europe-Only Country/ permit domestic reductions domestic transfers total cost Region price reductions purchased abatement abroad cost U.S. $36.6 159 247 $2.9b $9.0b $11.9b Canada $36.6 13 41 $0.2b $1.5b $1.7b Japan $36.6 19 82 $0.3b $3.0b $3.3b Australia $36.6 10 8 $0.2b $0.3b $0.5b FSU $36.6 378 0 $3.0b -$13.8b net gain: $10.8b EU $101 178 145 $9.0b $14.6b $23.6b Eastern $101 145 0 $3.4b -$14.6 net gain: Europe $11.2b Two Permit Trading Blocs: Annex I Umbrella and Europe-Only and 50% Constraint Country/ permit domestic reductions domestic transfers total cost Region price reductions purchased abatement (domestic) abroad cost U.S. $48 203 203 $4.9b $3.0b $7.9b Canada $101 27 27 $1.4b $0.4b $1.8b Japan $119 50 51 $3.0b $0.8b $3.8b Australia $32 9 9 $0.1b $0.1b $0.2b FSU $15 289 0 $0.5b -$4.3b net gain: $3.8b EU $101 178 145 $9.0b $14.6b $23.6b Eastern $101 145 0 $3.4b -$14.6b net gain: Europe $11.2b Two Permit Trading Blocs: Annex I Umbrella + Global JI and Europe-Only (no JI) Country/ permit domestic reductions domestic transfers total cost Region price reductions purchased abatement abroad cost U.S. $12 50 356 $0.3b $4.4b $4.7b Canada $12 4 50 $0.02b $0.6b $0.6b Japan $12 7 94 $0.04b $1.2b $1.2b Australia $12 4 14 $0.02b $0.2b $0.2b FSU $12 280 0 $0.4b -$3.5b net gain: $3.1 EU $101 178 145 $9.0b $14.6b $23.6b Eastern $101 145 0 $3.4b -$14.6b net gain: Europe $11.2b China $12 170 0 $1.1b -$2.1b net gain: $1.0b India $12 21 0 $0.1b -$0.3b net gain: $0.2b Korea $12 2 0 $0.01b -$0.02b net gain: $0.01b Mexico $12 11 0 $0.07b -$0.1b net gain: 0.03b ROW $12 30 0 $0.2b -$0.4b net gain: 0.2b US 2010 GDP 9185 Required Sinks Elect Res Scenario Permit Domestic Permits US US US TC TC/GDP Cut Under Price Reductions Purchased Abatement Capital ($b) Kyoto (92$) (MMTCE) Abroad Cost Outflow 600 0 0 Dom Only 171 600 0 51.3 0.0 51.3 0.005585 600 52 0 Dom Only 146 548 0 40.0 0.0 40.0 0.004355 600 52 33 Dom Only 130 515 0 33.5 0.0 33.5 0.003645 600 104 33 Dom Only 105 463 0 24.3 0.0 24.3 0.002646 600 0 33 Dom Only 155 567 0 43.9 0.0 43.9 0.004784 497 -55 0 10/97 position 148 552 0 40.8 0.0 40.8 0.004447 600 104 0 Dom Only 121 496 0 30,0 0.0 30.0 0.003267 600 0 0 Annex I 48 281 319 6.7 15.3 22.1 0.002401 600 52 0 Annex I 26 187 361 2.4 9.4 11.8 0.001287 600 104 0 Annex I 11 95 401 0.5 4.4 4.9 0.000537 600 0 0 Umb - EU 26 187 413 2.4 10.7 13.2 0.001434 600 52 0 Umb - EU 11 95 453 0.5 5.0 5.5 0.000599 600 104 0 Umb - EU 0 0 496 0.0 0.0 0.0 0 600 0 0 Umb - EU, EE 43 258 342 5.5 14.7 20.3 0.002205 600 52 0 Umb - EU, EE 20 157 391 1.6 7.8 9.4 0.001022 600 104 0 Umb - EU, EE 5 51 445 0.1 2.2 2.4 0.000256 600 0 0 A1 + JI 19 150 450 1.4 8.6 10.0 0.001086 600 52 0 A1 + JI 12 102 446 0.6 5.4 6.0 0.000649 600 104 0 A1 + JI 5 51 445 0.1 2.2 2.4 0.000256 600 0 0 Umb - EU + JI 10 84 516 0.4 5.2 5.6 0.000608 600 52 0 Umb - EU + JI 4 45 503 0.1 2.0 2.1 0.000229 600 104 0 Umb - EU + JI 0 0 496 0.0 0.0 0.0 0 Umb - EU, EE + JI 600 0 0 14 116 484 0.8 6.8 7.6 0.000826 Umb - EU, EE + JI 600 52 0 8 66 482 0.3 3.9 4.1 0.000449 Umb - EU, EE + JI 600 104 0 2 27 469 0.0 0.9 1.0 0.000105 600 0 0 A1 w/ Key LDC 20 150 450 1.5 9.0 10.5 0.001143 600 52 0 A1 w/ Key LDC 12 102 446 0.6 5.4 6.0 0.000649 600 104 0 A1 w/ Key LDC 5 51 445 0.1 2.2 2.4 0.000256 Umb - EU w/ Key 600 0 0 LDC 12 102 498 0.6 6.0 6.6 0.000717 Umb - EU w/ Key 600 52 0 LDC 5 51 497 0.1 2.5 2.6 0.000284 Umb - EU w/ Key 600 104 0 LDC 0 0 496 0.0 0.0 0.0 0 SGM/non-CO2 cost curve analysis 1. Required Sinks Elect Res Scenario Permit Domestic Permits US US US TC TC/GDP Cut Under Price Reductions Purchased Abatement Capital ($b) Kyoto (92$) (MMTCE) Abroad Cost Outflow Umb - EU, EE w/ 600 0 0 Key LDC 16 130 470 1.0 7.5 8.6 0.000932 Umb - EU, EE w/ 600 52 0 Key LDC 9 72 476 0.3 4.3 4.6 0.000502 Umb EU, EE w/ 600 104 0 Key LDC 2 27 469 0.0 0.9 1.0 0.000105 600 0 0 A1 + 20% JI 36 229 371 4.1 13.4 17.5 0.001903 600 52 0 A1 + 20% JI 21 164 384 1.7 8.1 9.8 0.001065 600 104 0 A1 + 20% JI 9 72 424 0.3 3.8 4.1 0.000451 Umb - EU + 20% 0 0 JI 20 157 443 1.6 8.9 10.4 0.001136 600 Umb - EU + 20% 600 52 0 JI 9 72 476 0.3 4.3 4.6 0.000502 Umb - EU + 20% 600 104 0 JI 0 0 496 0.0 0.0 0.0 0 Umb - EU, EE + 600 0 0 20% JI 30 204 396 3.1 11.9 14.9 0.001627 Umb - EU, EE + 600 52 0 20% JI 16 130 418 1.0 6.7 7.7 0.000841 Umb - EU, EE + 600 104 0 20% JI 4 45 451 0.1 1.8 1.9 0.000206 A1 w/ Key LDC + 600 0 0 20% JI 20 157 443 1.6 8.9 10.4 0.001136 A1 w/ Key LDC + 600 52 0 20% JI 12 102 446 0.6 5.4 6.0 0.000649 A1 w/ Key LDC + 600 104 0 20% JI 5 51 445 0.1 2.2 2.4 0.000256 Umb - EU w/ Key 600 0 0 LDC + 20% JI 11 95 505 0.5 5.6 6.1 0.000662 Umb - EU w/ Key 600 52 0 LDC + 20% JI 5 51 497 0.1 2.5 2.6 0.000284 Umb - EU w/ Key 600 104 0 LDC + 20% JI 0 0 496 0.0 0.0 0.0 0 Umb - EU, EE w/ Key LDC + 20% JI 600 0 0 15 123 477 0.9 7.2 8.1 0.000879 Umb - EU, EE w/ 600 52 0 Key LDC + 20% JI 9 72 476 0.3 4.3 4.6 0.000502 Umb - EU, EE w/ 600 104 0 Key LDC + 20% JI 2 27 469 0.0 0.9 1.0 0.000105 SGM/non-CO2 cost curve analysis The Effects of Annex I Trading and Ideal Global JI on Russia Scenario Permit Price FSU Permit FSU TR Russia TR FSU TC Russia TC Russia ($/ton) Sales (billions $) (billions $) (billions $) (billions $) TR-TC (MMTC) (billions $) 100% Paper $15.8 294 $4.65 $3.02 $0.63 $0.41 $2.61 Tons* 50% Paper $20.3 215 $4.36 $2.84 $1.10 $0.71 $2.13 Tons No Paper $25.9 130 $3.37 $2.19 $1.68 $1.09 $1.10 Tons* Model: SGM * From model runs. 50% paper tons case based on interpolations from constructed marginal abatement cost curves. Russian revenues and costs are assumed to be 65% of FSU revenues and costs, given that the Russian economy was approximately 65% the size of all FSU countries' economies in 1995. Total costs are calculated assuming that the marginal abatement cost curve is linear in those sections of the curve. All dollars are expressed in 1992 $. These analyses assume that there are no constraints on the ability of other Annex I countries to purchase emissions reductions abroad. The Effects of Annex I Trading Alone on Russia Scenario Permit Price FSU Permit FSU TR Russia TR FSU TC Russia TC Russia ($/ton) Sales (billions $) (billions $) (billions $) (billions $) TR-TC (MMTC) (billions $) 100% Paper $41.3 382 $15.8 $10.27 $3.5 $2.28 $7.99 Tons* 50% Paper $55 319 $17.5 $11.38 $5.8 $3.77 $7.61 Tons No Paper $71.6 251 $18.0 $11.70 $9.0 $5.85 $5.85 Tons* Model: SGM * From model runs. 50% paper tons case based on interpolations from constructed marginal abatement cost curves. Russian revenues and costs are assumed to be 65% of FSU revenues and costs, given that the Russian economy was approximately 65% the size of all FSU countries' economies in 1995. Total costs are calculated assuming that the marginal abatement cost curve is linear in those sections of the curve. All dollars are expressed in 1992 $. These analyses assume that there are no constraints on the ability of other Annex I countries to purchase emissions reductions abroad. The Effects of Annex I Trading and Ideal Global JI on Russia Scenario Permit Price FSU Permit FSU TR Russia TR FSU TC Russia TC Russia ($/ton) Sales (billions $) (billions $) (billions $) (billions $) TR-TC (MMTC) (billions $) 100% Paper $15.8 294 $4.65 $3.02 $0.63 $0.41 $2.61 Tons* 50% Paper $20.3 215 $4.36 $2.84 $1.10 $0.71 $2.13 Tons No Paper $25.9 130 $3.37 $2.19 $1.68 $1.09 $1.10 Tons* Model: SGM * From model runs. 50% paper tons case based on interpolations from constructed marginal abatement cost curves. Russian revenues and costs are assumed to be 65% of FSU revenues and costs, given that the Russian economy was approximately 65% the size of all FSU countries' economies in 1995. Total costs are calculated assuming that the marginal abatement cost curve is linear in those sections of the curve. All dollars are expressed in 1992 $. These analyses assume that there are no constraints on the ability of other Annex I countries to purchase emissions reductions abroad. The Effects of Annex I Trading Alone on Russia Scenario Permit Price FSU Permit FSU TR Russia TR FSU TC Russia TC Russia ($/ton) Sales (billions $) (billions $) (billions $) (billions $) TR-TC (MMTC) (billions $) 100% Paper $41.3 382 $15.8 $10.27 $3.5 $2.28 $7.99 Tons* 50% Paper $55 319 $17.5 $11.38 $5.8 $3.77 $7.61 Tons No Paper $71.6 251 $18.0 $11.70 $9.0 $5.85 $5.85 Tons* Model: SGM * From model runs. 50% paper tons case based on interpolations from constructed marginal abatement cost curves. Russian revenues and costs are assumed to be 65% of FSU revenues and costs, given that the Russian economy was approximately 65% the size of all FSU countries' economies in 1995. Total costs are calculated assuming that the marginal abatement cost curve is linear in those sections of the curve. All dollars are expressed in 1992 $. These analyses assume that there are no constraints on the ability of other Annex I countries to purchase emissions reductions abroad. Preliminary Draft MEMORANDUM TO: Ray Squitieri FROM: Joe Aldy DATE: December 3, 1997 RE: Constraints on Annex I Permit Trading: Paper Tons and Purchase Caps To assess the effects of constraining the percentage of paper tons EEFSU countries can sell as well as the percentage of emissions reductions occurring abroad other Annex I countries can buy, I reviewed model outputs from SGM. Both of these constraints can increase the price of a carbon permit, depending on the stringency of the constraint. In most cases, a stringent purchase cap (e.g., no country can buy more than 10% of its emissions reductions abroad) drives the U.S. permit price (see attached table). For the scenarios with paper tons and 10% purchase caps, EEFSU countries cannot even sell all of their paper tons. For the scenarios with paper tons and 25% purchase caps, EEFSU countries do sell all of their paper tons, and some tons representing actual emissions reductions relative to BAU. However, constraints on the extent of allowable paper ton sales and loose purchase caps (e.g., 50%) are driven by the paper ton constraint. In the case of a 50% paper ton constraint and a 50% purchase cap, non-EEFSU countries buy 49% of their emissions reductions from EEFSU countries. In the case of no paper tons and a 50% purchase cap, non-EEFSU countries buy 38% of their emissions reductions from EEFSU countries. Preliminary Draft Constraints U.S. Permit Price Permits Purchased Abroad by ($/ton) U.S. (MMTCE) 100% paper tons, no purchase $41* 202 caps (no constraints) 50% paper tons, no purchase $55 149 caps no paper tons, no purchase $72* 97 caps 100% paper tons, 10% $93 38 purchase cap 100% paper tons, 25% $72 95 purchase cap 100% paper tons, 50% $45 190 purchase cap 50% paper tons, 10% $93 38 purchase cap 50% paper tons, 25% $72 95 purchase cap 50% paper tons, 50% $55 151 purchase cap no paper tons, 10% purchase $93 38 cap no paper tons, 25% purchase $72 95 cap no paper tons, 50% purchase $72 97 cap *From model runs. All other estimates are derived from assessing the constructed marginal abatement cost curves depicted in attached graphs. Preliminary Draft MEMORANDUM TO: Ray Squitieri FROM: Joe Aldy DATE: December 2, 1997 RE: The Effects of Trading Constraints on Permit Prices and Income Flows To assess the effects of trading constraints on U.S. permit prices and income flows, I reviewed several model runs from SGM. While unconstrained trading would result in minimizing the compliance costs with binding targets, some have proposed constraining the amount of emissions reductions allowed to be achieved through international permit trading and joint implementation. The following table presents the permit prices and the quantity of permits purchased abroad under various percentage-based constraints. U.S. Permit Prices for Meeting 1990 in 2010 under Various Trading Constraints % of Reductions Allowed to Annex I Trading Only Annex I Trading + be Purchased from Other (permits purchased abroad) Joint Implementation Countries (permits purchased abroad) 100% (no constraint) $41/ton (202 MMTCE) $16/ton (312 MMTCE) 0% (no trading, no JI) $108/ton (0 MMTCE) $108/ton (0 MMTCE) 10% $93/ton (38 MMTCE) $93/ton (38 MMTCE) 25% $72/ton (95 MMTCE) $72/ton (95 MMTCE) 50% $45/ton (190 MMTCE) $45/ton (190 MMTCE) model: SGM Constraints on Trade This analysis provides four interesting points. First, in an unconstrained world, the U.S. buys 53% of its emissions reductions abroad with Annex I trading and 82% of reductions with JI and Annex I trading. Note in the second table that the U.S. buys less as a share of reductions than other Annex I countries (however, EEFSU countries are sellers). Second, any split between JI and Annex I trading (e.g., no more than 5% through trading, no more than 5% through JI) is irrelevant to the U.S. permit price so long as the total constraint (e.g., 10%) is less than the percentage of emissions reductions the U.S. would buy under no constraints. Third, Russia sells more permits than China under JI (but less than it would with an Annex I trading only world) because it has so many paper tons. Fourth, any constraint less than 60% implies that EEFSU countries would not be able to sell all of their paper tons. Preliminary Draft Percent of Emissions Reductions Purchased Abroad with No Constraints on Trading and JI Country/Region Annex I Trading Only Annex I Trading + Joint Implementation Canada 75% 88% Japan 78% 91% Western Europe 65% 86% Australia 58% 81% U.S. 53% 82% model: SGM Income Flows Not only do constraints on trade negatively impact the buyers of permits by increasing permit prices, but they also decrease the flows of income to countries that sell. In a no-constraint world, the sellers of permits would receive $20 billion in 2010 under Annex I trading and more than $10 billion in 2010 under Annex I trading and JI. However, a constraint of 10% emissions reduction through foreign purchases would result in income transfers of only $600 million under Annex I trading and $200 million under Annex I trading plus JI. This 10% constraint results in EIT and developing countries receiving only 2-3% of the income they would have received without the constraint. August 19, 1997 Comments on "Additionality in the Context of Joint Implementation", August 13 Draft Office of Economic Policy, Treasury Department 1) The central problem with this paper is that it does not sufficiently acknowledge the fundamentally difficult task of establishing that emissions reductions are incremental. Each of the approaches proposed involves comparisons to a counterfactual that is not observed, only modelled. As such, the guarantee of additionality is only as good as the models themselves. The memo presents no evidence on the quality of the predictions from these types of models, nor a sufficiently rich discussion of the technical difficulties inherent in applying these types of modelling exercises in the context of JI. To see this, compare the approaches listed in this paper to the best possible approach: JI with binding developing country committments on emissions levels. This simply amounts to international trading with developing countries included. In such a world, we are guaranteed incrementality: to claim credit for a unit of emissions reduction, a company in the U.S. will have to buy a permit in the developing country for one unit of emissions. This will by definition guarantee that emissions in the developing country drop by one unit, since there is a fixed total of emissions credits in that country. Thus, additionality is assured, since there is now one fewer units of total emissions from other sources in the developing country. Now, compare to this the best possible modelling approach. Under this approach, engineering specialists draw up a prediction about the future path of generation power sources (for example) in the developing country. But any prediction is just that, and as such will have some error. In some cases, the prediction will understate the pace of technological advance in the developing country, SO that companies will get credit for building a type of plant that would have been built anyway. In some cases, the prediction will overstate the pace, so that companies will not build a plant because they are not getting credit. In any case, given that all models make prediction errors, there will be resulting error rates in our ability to assign JI credit. So there is no way to guarantee additionality without binding developing country committments. This is not to say that there is JI is completely non-additional under the proposed approaches. Indeed, the methodologies suggested are clever and are probably the best that we can do without committments from developing countries. But while the paper acknowledges potential problems such as those described above in the fourth paragraph, it then immediately turns into the fifth to "assuring additionality". A footnote in that paragraph offers the appropriate caveat, but the implication is that additionality can be assured nonetheless. This problem can be partially addressed by making three straightforward changes: a) A new fourth paragraph on the first page should discuss the benchmark case: international trading. It should bring out the key points made above: that any Л approach which revolves around modelling baselines can only approach, and can never achieve, the additionality that would be guaranteed if developing countries were brought into an international trading regime. b) Footnote 4 should be brought into the text and amplified. c) Terms like "high accuracy" should be prefaced by "relatively"; these options are the highest accuracy of those listed, but may be very low accuracy relative to the ideal benchmark of international trading. 2) Do we have any evidence on how accurate the forecasts of counterfactual technology paths are likely to be? If we had done forecasts based on available data for the U.S. or some other country 10 years ago, would we have predicted anything like our current pattern of technology adoption for generation and other activities? This goes to the heart of these type of modelling approaches to establishing additionality. 3) As CEA and Treasury highlighted in their comments on an earlier draft, different approaches to modelling the processes of technological diffusion are likely to yield very different answers for the counterfactual. How will this uncertainty be resolved? This is a particularly salient question given the "all or nothing" aspect of JI. If there is some penetration of a particular type of generation, but it has not fully diffused, the company might get no credit at all. This will provide a particularly strong incentive for firms to promote the engineering estimates that support their case. How will these pressures be resisted, given the true uncertainty in the estimates? One particular logistical difficulty will be deciding which country will be responsible for determining the counterfactual. If we let the developing countries themselves model the counterfactual, one could imagine an "engineering bidding war", where each of several countries engineers compete to make their baseline look the least advanced (thus making their country more attractive for investment by presenting more opportunity for additionality). This could be rectified by an objective U.S. or third party engineer, but this might raise a difficult set of sovereignty issues. 4) Option 3 provides the closest alternative to true additionality through developing country committments, by measuring changes in sector specific emissions patterns. But once again, it must be recognized that emissions are projected based on out of sample modelling. How accurate are our models? Once again, it would be constructive to assess how well existing models would do at predicted observed emissions history based on data that ended 10 years ago. 5) Options 4 and 5 strike us as the most problematic, since they are the most subjective. Any option which is strongly influenced by the type of proposal made by the firm, rather than by objective evidence to support additionality, will simply reward firms that are best at making their case, not those that are guaranteeing the most additionality. This will lead to huge inefficiencies ("rent-seeking losses") as firms devote substantial resources to making a case look as good as possible, rather than devoting those resources to the investment itself. 9/18/97 Outputs Modelers Target Timetable Trading Permit Revenue Burden BAU Paper Tons AEEI Ramp-up Time Path PDV (5%; GDP in 2010 ID Scenario Model Allocation Recycling Sharing Emissions or Corre- 2000-2050) (deviation Path sponding Foregone from BAU) Assumptio Reductions Consumption n SGM1 BAU SGM, Battelle n/a n/a n/a n/a n/a n/a IAT n/a 1.0 n/a -->2050, $121,650 billion $9,185 billion -->2100 (BAU (BAU GDP) MAGICC (climate) consumption) -10% of SGM, Battelle -10% 1990 stabilize in Annex I auction lump-sum no LDC IAT PT 1.0 no -->2050, $485 billion $9,155 billion SGM9 1990 in MAGICC emissions 2010 part. -->2100 (-$30 billion) 2010 level (climate) SGM8 -10% of SGM, Battelle -10% 1990 stabilize in domestic auction lump-sum no LDC IAT PT 1.0 no -->2050, $980 billion $9,149 billion part. -->2100 (-$36 billion) 1990 in MAGICC emissions 2010 only (climate) 2010 level SGM10 -10% of SGM, Battelle -10% 1990 stabilize in worldwide auction lump-sum no LDC IAT PT 1.0 no -->2050, $50 billion $9,171 billion -->2100 1990 in MAGICC emissions part. (-$14 billion) 2010 (climate) 2010 level SGM36 1990 in SGM, Battelle 1990 stabilize in Annex 1 auction lump-sum no LDC IAT PT 1.0 yes -->2050, $195 billion $9,179 billion 2020 MAGICC emissions 2020 part. -->2100 (-$6 billion) (climate) level SGM39 1990 in SGM, Battelle 1990 stabilize in Annex I auction lump-sum no LDC IAT PT 1.0 in 2005 -->2050, $230 billion $9,172 billion -->2100 2010 MAGICC emissions part. (-$13 billion) 2010 (climate) level SGM3 1990 in SGM, Battelle 1990 stabilize in Annex I auction lump-sum no LDC IAT PT 1.0 no -->2050, $225 billion $9,172 billion (-$13 billion) 2010 MAGICC emissions 2010 part. -->2100 level (climate) SGM18 1990 in SGM, Battelle 1990 stabilize in Annex I auction lump-sum LDC IAT PT 1.0 in 2005 -->2050, $255 billion $9,172 billion 2010 MAGICC emissions 2010 stabilizes -->2100 (-$13 billion) level at 2030 in (climate) 2030 SGM, Battelle 1990 stabilize in Annex I auction lump-sum LDC IAT PT 1.0 in 2005 -->2050, $260 billion $9,172 billion SGM21 1990 in 2010 MAGICC emissions 2010 BAU to -->2100 (-$13 billion) 2030, (climate) level equal per capita in 2050 SGM17 1990 in 1990 stabilize in domestic auction lump-sum LDC IAT PT 1.0 in 2005 -->2050, $740 billion $9,164 billion SGM, Battelle 2010 MAGICC emissions 2010 stabilizes -->2100 (-$21 billion) only level at 2030 in (climate) 2030 no LDC IAT PT 1.0 no -->2050, $665 billion $9,165 billion SGM2 1990 in SGM, Battelle 1990 stabilize in domestic auction Tump-sum -->2100 (-$20 billion) 2010 MAGICC emissions 2010 only part. (climate) level Battelle 1990 stabilize in domestic auction Tump-sum no LDC IAT PT 1.0 in 2005 -->2050, $740 bilion $9,164 billion SGM38 1990 in SGM, 2010 MAGICC emissions part. -->2100 (-$21 billion) 2010 only (climate) level PT 1.0 in 2005 -->2050, $740 billion $9,164 billion SGM20 1990 in SGM, Battelle 1990 stabilize in domestic auction Tump-sum LDC IAT 2010 MAGICC emissions 2010 only BAU to -->2100 (-$21 billion) level 2030, (climate) equal per capita in 2050 SGM40 1990 in SGM, Battelle 1990 stabilize in worldwide auction lump-sum no LDC IAT PT 1.0 in 2005 -->2050, $5 billion $9,180 billion 2010 part. -->2100 (-$5 billion) MAGICC emissions 2010 (climate) level 9/18/97 Scenario Model Modelers Target Timetable Trading Permit Revenue Burden BAU Paper Tons AEEI Ramp-up Time Path PDV (5%; GDP in 2010 ID Allocation Recycling Sharing Emissions or Corre- 2000-2050) (deviation Path sponding Foregone from BAU) Assumptio Reductions Consumption n SGM4 1990 in SGM, Battelle 1990 stabilize in worldwide auction lump-sum no LDC IAT PT 1.0 no -->2050, $5 billion $9,180 billion -->2100 (-$5 billion) 2010 MAGICC emissions 2010 part. level (climate) SGM22 1990 in SGM, Battelle 1990 stabilize in worldwide auction lump-sum LDC IAT PT 1.0 in 2005 -->2050, $25 billion $9,180 billion 2010 MAGICC emissions 2010 BAU to -->2100 (-$5 billion) level 2030, (climate) equal per capita in 2050 SGM, Battelle 1990 stabilize in worldwide auction lump-sum LDC IAT PT 1.0 in 2005 -->2050, $20 billion $9,180 billion SGM19 1990 in 2010 MAGICC emissions 2010 stabilizes -->2100 (-$5 billion) level at 2030 in (climate) 2030 SGM35 1990 in SGM, Battelle 1990 stabilize in domestic auction Tump-sum no LDC IAT PT 1.0 yes -->2050, $660 billion $9,172 billion 2020 MAGICC emissions 2020 only part. -->2100 (-$13 billion) level (climate) 1990 worldwide auction lump-sum no LDC IAT PT 1.0 yes -->2050, -$1 billion $9,183 billion SGM37 1990 in SGM, Battelle stabilize in 2020 MAGICC part. -->2100 emissions 2020 (-$2 billion) level (climate) SGM6 1995 in SGM, Battelle 1995 stabilize in Annex I auction lump-sum no LDC IAT CR 1.0 no -->2050, $380 billion $9,175 billion 2010 MAGICC emissions 2010 part. -->2100 (-$10 billion) level (climate) stabilize in domestic auction lump-sum no LDC IAT CR 1.0 no -->2050, $425 billion $9,174 billion SGM5 1995 in SGM, Battelle 1995 2010 MAGICC emissions 2010 part. -->2100 (-$11 billion) only level (climate) SGM7 1995 in Battelle 1995 stabilize in worldwide auction lump-sum no LDC IAT CR 1.0 no -->2050, $30 billion $9,179 billion SGM, 2010 MAGICC emissions part. -->2100 2010 (-$6 billion) level (climate) Annex 1 auction lump-sum no LDC IAT CR 1.0 yes -->2050, $410 billion $9,176 billion SGM33 1995 in SGM, Battelle 1995 stabilize in 2020 part. -->2100 (-$9 billion) MAGICC emissions 2020 level (climate) SGM32 1995 in Battelle 1995 stabilize in domestic auction lump-sum no LDC IAT CR 1.0 yes -->2050, $445 billion $9,176 billion SGM, 2020 MAGICC emissions 2020 part. -->2100 (-$9 billion) only level (climate) auction lump-sum no LDC IAT CR 1.0 yes -->2050, -$40 billion $9,180 billion SGM34 1995 in SGM, Battelle 1995 stabilize in worldwide part. -->2100 (-$5 billion) 2020 MAGICC emissions 2020 level (climate) SGM12 Battelle 2010 BAU stabilize in Annex I auction lump-sum no LDC TAT PT 1.0 no -->2050, $55 billion $9,185 billion Peak in SGM, 2015 MAGICC emissions in part. -->2100 ($0 billion) 2040 2015; 1990 (climate) level in 2040 lump-sum no LDC IAT PT 1.0 in 2005 -->2050, $75 billion $9,185 billion SGM30 Peak in SGM, Battelle 2010 BAU stabilize in Annex I auction 2015 MAGICC emissions in 2040 part. -->2100 (0 billion) 2015; 1990 (climate) level in 2040 SGM, Battelle 2010 BAU stabilize in Annex 1 auction lump-sum LDC IAT PT 1.0 in 2005 -->2050, $75 billion $9,185 billion SGM24 Peak in 2015 MAGICC stabilizes -->2100 (0 billion) emissions in 2040 2015; 1990 at 2030 in (climate) level in 2040 2030 9/18/97 ID Scenario Model Modelers Target Timetable Trading Permit Revenue Burden BAU Paper Tons AEEI Ramp-up Time Path PDV (5%; GDP in 2010 Allocation Recycling Sharing Emissions or Corre- 2000-2050) (deviation Path sponding Foregone from BAU) Assumptio Reductions Consumption D Annex I auction lump-sum LDC IAT PT 1.0 in 2005 -->2050, $75 billion $9,185 billion SGM27 Peak in SGM, Battelle 2010 BAU stabilize in 2015 MAGICC emissions in 2040 BAU to -->2100 (0 billion) 2015; 1990 2030, (climate) level in 2040 equal per capita in 2050 SGM11 Peak in SGM, Battelle 2010 BAU stabilize in domestic auction lump-sum no LDC IAT PT 1.0 no --->2050, $235 billion $9,185 billion MAGICC emissions in 2040 only part. -->2100 ($0 billion) 2015 2015; 1990 (climate) level in 2040 SGM29 Peak in SGM, Battelle 2010 BAU stabilize in domestic auction lump-sum no LDC IAT PT 1.0 in 2005 -->2050, $270 billion $9,184 billion 2015 MAGICC emissions in 2040 only part. -->2100 (-$1 billion) 2015; 1990 (climate) level in 2040 stabilize in domestic auction lump-sum LDC IAT PT 1.0 in 2005 -->2050, $270 billion $9,184 billion SGM26 Peak in SGM, Battelle 2010 BAU 2015 MAGICC emissions in 2040 only BAU to -->2100 (-$1 billion) 2015; 1990 2030, (climate) level in 2040 equal per capita in 2050 SGM23 Peak in SGM, Battelle 2010 BAU stabilize in domestic auction lump-sum LDC IAT PT 1.0 in 2005 -->2050, $270 billion $9,184 billion MAGICC emissions in 2040 only stabilizes -->2100 (-$1 billion) 2015 2015; 1990 at 2030 in (climate) level in 2040 2030 SGM25 Peak in SGM, Battelle 2010 BAU stabilize in worldwide auction lump-sum LDC TAT PT 1.0 in 2005 -->2050, -$3 billion $9,185 billion 2015 MAGICC emissions in 2040 stabilizes -->2100 (0 billion) 2015; 1990 at 2030 in (climate) level in 2040 2030 lump-sum no LDC IAT PT 1.0 in 2005 -->2050, -$5 billion $9,185 billion SGM31 Peak in SGM, Battelle 2010 BAU stabilize in worldwide auction 2015 MAGICC emissions in 2040 part. -->2100 (0 billion) 2015; 1990 (climate) level in 2040 Peak in SGM, Battelle 2010 BAU stabilize in worldwide auction lump-sum no LDC IAT PT 1.0 no -->2050, -$8 billion $9,185 billion SGM13 2015 MAGICC emissions in 2040 part. -->2100 ($0 billion) 2015; 1990 (climate) level in 2040 SGM28 Peak in SGM, Battelle 2010 BAU stabilize in worldwide auction Tump-sum LDC IAT PT 1.0 in 2005 -->2050, $1 billion $9,185 billion 2015 2040 BAU to -->2100 emissions in (0 billion) MAGICC 2015; 1990 2030, (climate) level in 2040 equal per capita in 2050 +10% 1990 Annex I auction lump-sum no LDC IAT PT 1.0 in 2005 -->2050, $95 billion $9,181 billion SGM15 +10% of SGM, Battelle stabilize in 1990 in MAGICC 2010 part. -->2100 emissions (-$4 billion) level (climate) 2010 SGM14 +10% of SGM, Battelle +10% 1990 stabilize in domestic auction lump-sum no LDC IAT PT 1.0 in 2005 -->2050, $460 billion $9,174 billion 1990 in MAGICC emissions 2010 only part. -->2100 (-$11 billion) 2010 level (climate) 9/18/97 Modelers Target Timetable Trading Permit Revenue Burden BAU Paper Tons AEEI Ramp-up Time Path PDV (5%; GDP in 2010 ID Scenario Model Allocation Recycling Sharing Emissions or Corre- 2000-2050) (deviation Path sponding Foregone from BAU) Assumptio Reductions Consumption n worldwide auction lump-sum no LDC IAT PT 1.0 in 2005 -->2050, $1 billion $9,184 billion SGM16 +10% of SGM, Battelle +10% 1990 stabilize in 1990 in part. -->2100 (-$1 billion) MAGICC emissions 2010 2010 level (climate) MM1 n/a n/a n/a n/a n/a n/a IAT n/a ~T.0 n/a -->2025 n/a $9,205 billion BAU Markal DOE (BAU GDP) auction lump-sum no LDC IAT n/a ~1.0 n/a -->2025 n/a $9,137 billion MM2 1990 in Markal DOE 1990 stabilize in domestic 2010 emissions 2010 only part. (-$68 billion) level MM3 1995 in Markal 1995 stabilize in domestic auction lump-sum no LDC IAT n/a ~1.0 n/a -->2025 n/a $9,152 billion DOE 2010 emissions 2010 only part. (-$53 billion) level n/a ~1.0 n/a -->2025 n/a $9,197 billion MM4 1990 in Markal DOE 1990 stabilize in domestic auction lump-sum no LDC IAT 2020 emissions 2020 part. (-$8 billion) only level DOE 2010 BAU stabilize in domestic auction lump-sum no LDC IAT n/a ~1.0 n/a -->2025 n/a $9,201 billion MM5 Peak in Markal (-$4 billion) 2015 emissions in 2040 only part. 2015; 1990 level in 2040 9/18/97 ID Permit Permit Permit Conc. Conc. Year Conc. Change in Change in Emissions Year Intl. Trade of Intl. Trade of Date File (h:\jaldy\) Prices: Prices: Prices: (ppmv) in (ppmv) in (550 ppmv) Temp. (deg. Temp. (deg. Peak, Returns to Permits, U.S., Permits, U.S., Received 2010 2025 2050 2050 2100 Reaches 2x C) from 1990 C) from 1990 mmtce 1990 2010 2050 Run (Deviation (Deviation Pre-Ind. Level in 2050 in 2100 (year) (MMTCE), ($) (MMTCE), ($) from BAU) from BAU) (Deviation (Deviation (Deviation from BAU) from BAU) from BAU) SGM1 $0 $0 $0 502 711 2065 1.06 2.36 no peak: n/a n/a n/a 08/21/97 cea90_~1.xls 2245 (2050) SGM9 $91 $137 $238 481 645 2074 0.97 2.11 1637 never returns -180, -332, 08/21/97 cea90m_~1.xls (-21) (-66) (+9) (-0.09) (-0.25) (2005) (-$16.4 billion) (-$79.0 billion) SGM8 $175 $304 $924 481 645 2074 0.97 2.11 1637 2010 n/a n/a 08/21/97 cea90m_~1.xls (-21) (-66) (+9) (-0.09) (-0.25) (2005) SGM10 $33 $37 $45 481 645 2074 0.97 2.11 no peak: never returns -332, -845, 08/21/97 cea90m_~1.xls (-21) (-66) (+9) (-0.09) (-0.25) 2060 (-$11.0 billion) (-$38.0 billion) (2050) SGM36 $23 $86 $150 486 656 2072 0.99 2.16 no peak: n/a -180 -364 09/10/97 case20-1.xls (-16) (-55) (+7) (-0.07) (-0.20) 1714 (-$4.14 billion) (-$54.6 billion) (2050) SGM39 $41 $84 $149 486 656 2072 0.99 2.16 no peak: never returns -202, -364, 09/10/97 case10~1.xls (-$8.3 billion) (-$54.2 billion) (-16) (-55) (+7) (-0.07) (-0.20) 1714 (2050) SGM3 $42 $84 $149 486 655 2072 0.99 2.15 no peak: never returns -202, -364, 08/21/97 cea90_~2.xls (-16) (-56) (+7) (-0.07) (-0.21) 1714 (-$8.5 billion) (-$54.2 billion) (2050) SGM18 $39 $83 $150 478 565 2091 0.96 1.82 no peak: never returns -209, -362, 08/27/97 case2.xls, (-24) (-146) (+26) (-0.10) (-0.54) 1712 (-$8.2 billion) (-$54.3 billion) 9/4/97 fax (2050) SGM21 $39 $83 $150 n/a n/a n/a n/a n/a no peak: never returns -209, -362, 08/27/97 case3.xls 1712 (-$8.2 billion) (-$54.3 billion) (2050) SGM17 $108 $188 $582 476 564 2092 0.95 1.82 1550 2010 n/a n/a 08/27/97 case2.xls, (2000) 9/4/97 fax (-26) (-147) (+27) (-0.11) (-0.54) SGM2 $110 $191 $582 485 658 2072 0.99 2.16 1637 2010 n/a n/a 08/21/97 cea90_~2.xls (-17) (-53) (+7) (-0.07) (-0.20) (2005) SGM38 $108 $188 $582 484 656 2073 0.98 2.15 1550 2010 n/a n/a 09/10/97 case10~1.xls (-18) (-55) (+8) (-0.08) (-0.21) (2000) n/a n/a n/a n/a n/a 1550 2010 n/a n/a 08/27/97 case3.xls SGM20 $108 $188 $582 (2000) $23 $33 486 656 2072 0.99 2.16 no peak: never returns -312, -751, 09/10/97 case10~1.xis SGM40 $16 (-16) (-55) (+7) (-0.07) (-0.20) 2101 (-$5.0 billion) (-$24.8 billion) (2050) 9/18/97 ID Permit Permit Permit Conc. Conc. Year Conc. Change in Change in Emissions Year Intl. Trade of Intl. Trade of Date File (h:\jaldy\) Prices: Prices: Prices: (ppmv) in (ppmv) in (550 ppmv) Temp. (deg. Temp. (deg. Peak, Returns to Permits, U.S., Permits, U.S., Received 2010 2025 2050 2050 2100 Reaches 2x C) from 1990 C) from 1990 mmtce 1990 2010 2050 Run (Deviation (Deviation Pre-Ind. Level in 2050 in 2100 (year) (MMTCE), ($) (MMTCE), ($) from BAU) from BAU) (Deviation (Deviation (Deviation from BAU) from BAU) from BAU) SGM4 $16 $23 $32 486 655 2072 0.99 2.15 no peak: never returns -313, -752, 08/21/97 cea90_~2.xls (-16) (-56) (+7) (-0.07) (-0.21) 2102 (-$5.0 billion) (-$24.1 billion) (2050) $0 n/a n/a n/a n/a n/a no peak: never returns n/a n/a 08/27/97 case3.xls SGM22 $15 $23 2239 (2050) SGM19 $15 $23 $110 478 565 2091 0.96 1.82 1883 never returns -316, -500, 08/27/97 case2.xls, (-24) (-146) (+26) (-0.10) (-0.54) (2040) (-$4.7 billion) (-$55.0 billion) 9/4/97 fax SGM35 $71 $192 $582 484 656 2073 0.98 2.15 1550 2020 n/a n/a 09/10/97 case20~1.xls (-18) (-55) (+8) (-0.08) (-0.21) (2000) SGM37 $9 $24 $33 486 656 2072 0.99 2.16 no peak: n/a -246 -751 09/10/97 case20~1.xls (-16) (-55) (+7) (-0.07) (-0.20) 2101 (-$2.21 billion) (-$24.78 (2050) billion) SGM6 $74 $119 $202 483 648 2074 0.98 2.12 1637 never returns 36, -122, 08/21/97 cea95_-1.xls (-19) (-63) (+9) (-0.08) (-0.24) (2005) (+$2.7 billion) (-$24.6 billion) SGM5 $62 $131 $317 483 648 2074 0.98 2.12 1637 never returns n/a n/a 08/21/97 cea95_~1.xls (-19) (-63) (+9) (-0.08) (-0.24) (2005) SGM7 $27 $32 $41 483 648 2074 0.98 2.12 no peak: never returns -131, -593, 08/21/97 cea95_~1.xls (-19) (-63) (+9) (-0.08) (-0.24) 2073 (-$3.5 billion) (-$24.3 billion) (2050) SGM33 $57 $120 $203 482 648 2074 0.97 2.12 no peak: n/a 20 -119, 09/10/97 case20~2.xls (-20) (-63) (+9) (-0.09) (-0.24) 1599 ($1.14 billion) (-$24.16 (2050) billion) SGM32 $51 $131 $318 482 648 2074 0.97 2.12 1550 n/a n/a n/a 09/10/97 case20~2.xls (-20) (-63) (+9) (-0.09) (-0.24) (2000) $33 $41 482 648 2074 0.97 2.12 no peak: n/a -122 -592 09/10/97 case20-2.xls SGM34 $21 (-20) (-63) (+9) (-0.09) (-0.24) 2072 (-$2.56 billion) (-$24.27 (2050) billion) SGM12 $0 $47 $155 491 n/a n/a 1.02 n/a 1807 never returns 0, (n/a) -367, 08/21/97 cea90_~1.xls (-11) (-0.04) (2015) (-$56.9 billion) 08/27/97 case6.xls, $153 489 658 2071 1.00 2.17 1755 never returns -5, -367, SGM30 $11 $56 (-13) (-53) (+6) (-0.06) (-0.19) (2015) (-$0.06 billion) (-$56.2 billion) 9/4/97 fax SGM24 $11 $56 $153 481 568 2089 0.98 1.84 1755 never returns is, -367, 08/27/97 case4.xls, (-21) (-143) (+24) (-0.08) (-0.52) (2015) (-$0.06 billion) (-$56.2 billion) 9/4/97 fax 9/18/97 ID Permit Permit Permit Conc. Conc. Year Conc. Change in Change in Emissions Year Intl. Trade of Intl. Trade of Date File (h:\jaldy\) Prices: Prices: Prices: (ppmv) in (ppmv) in (550 ppmv) Temp. (deg. Temp. (deg. Peak, Returns to Permits, U.S., Permits, U.S., Received 2010 2025 2050 2050 2100 Reaches 2x C) from 1990 C) from 1990 mmice 1990 2010 2050 Run (Deviation (Deviation Pre-Ind. Level in 2050 in 2100 (year) (MMTCE), ($) (MMTCE), ($) from BAU) from BAU) (Deviation (Deviation (Deviation from BAU) from BAU) from BAU) SGM27 $56 $153 n/a n/a n/a n/a n/a 1755 never returns -5, -367, 08/27/97 case5.xls $11 (2015) (-$0.06 billion) (-$56.2 billion) SGM11 $0 $84 $559 489 n/a n/a 1.01 n/a 1807 2040 n/a n/a 08/21/97 cea90_~1.xls (-13) (-0.05) (2015) $12 $99 $563 489 658 2071 1.00 2.17 1729 2040 n/a n/a 08/27/97 case6.xls, SGM29 (-13) (-53) (+6) (-0.06) (-0.19) (2015) 9/4/97 fax SGM26 $12 $99 $563 n/a n/a n/a n/a n/a 1729 2040 n/a n/a 08/27/97 case5.xls (2015) SGM23 $563 481 568 2089 0.98 1.84 1729 2040 n/a n/a 08/27/97 $12 $99 case4.xls, (-21) (-143) (+24) (-0.08) (-0.52) (2015) 9/4/97 fax SGM25 $4 $16 $111 481 568 2089 0.98 1.84 1893 never returns -30, -500, 08/27/97 case4.xls, (-21) (-143) (+24) (-0.08) (-0.52) (2030) (-$0.1 billion) (-$55.5 billion) 9/4/97 fax 08/27/97 case6.xls, SGM31 $4 $16 $33 489 658 2071 1.00 2.17 no peak: never returns -30, -751, (-13) (-53) (+6) (-0.06) (-0.19) 2101 (-$0.12 billion) (-$24.8 billion) 9/4/97 fax (2050) SGM13 $14 $33 491 n/a n/a 1.02 n/a no peak: never returns 0, (n/a) -752, 08/21/97 $0 cea90_~1.xls (-11) (-0.04) 2102 (-$24.8 billion) (2050) SGM28 $4 $16 $0 n/a n/a n/a n/a n/a no peak: never returns n/a n/a 08/27/97 case5.xls 2241 (2050) $113 489 662 2071 1.00 2.18 no peak: never returns -172, -339, 08/27/97 casel.xls SGM15 $17 $56 (-13) (-49) (+6) (-0.06) (-0.18) 1824 (-$2.9 billion) (-$38.3 billion) (2050) SGM14 $60 $128 $310 488 664 2071 0.99 2.18 1550 never returns n/a n/a 08/27/97 case1.xls (-14) (-47) (+6) (-0.07) (-0.18) (2000) 9/18/97 ID Permit Permit Permit Conc. Conc. Year Conc. Change in Change in Emissions Year Intl. Trade of Intl. Trade of Date File (h:\jaldy\) Prices: Prices: Prices: (ppmv) in (ppmv) in (550 ppmv) Temp. (deg. Temp. (deg. Peak, Returns to Permits, U.S., Permits, U.S., Received 2010 2025 2050 2050 2100 Reaches 2x C) from 1990 C) from 1990 mmtce 1990 2010 2050 Run (Deviation (Deviation Pre-Ind. Level in 2050 in 2100 (year) (MMTCE), ($) (MMTCE), ($) from BAU) from BAU) (Deviation (Deviation (Deviation from BAU) from BAU) from BAU) SGM16 $7 $16 $25 489 662 2071 1.00 2.18 no peak: never returns -219, -643, 08/27/97 casel.xls (-13) (-49) (+6) (-0.06) (-0.18) 2128 (-$1.5 billion) (-$16.1 billion) (2050) MM1 $0 $0 n/a n/a n/a n/a n/a n/a no peak: n/a n/a n/a 08/26/97 8_26runl.wk4 2066 (2025) MM2 $148 $192 n/a n/a n/a n/a n/a n/a 1586 2010 n/a n/a 08/26/97 8_26runl.wk4 (2005) MM3 $136 $146 n/a n/a n/a n/a n/a n/a 1600 n/a n/a n/a 08/26/97 8_26runl.wk4 (2005) MM4 $0 $198 n/a n/a n/a n/a n/a n/a 1749 2020 n/a n/a 08/26/97 8_26runl.wk4 (2010) MM5 $0 $99 n/a n/a n/a n/a n/a n/a 1767 2040* n/a n/a 08/26/97 8_26runl.wk4 (2015) c.maccracken @ pnl.gov 10/13/97 01:47:00 PM Record Type: Record To: joseph e. aldy CC: Subject: SGM Results Joe - I assume you're not at work today ... I finished the $50 Safety Valve runs that you requested on Friday. I have a disk waiting at the front desk to be picked up by your courier. I will be out Tuesday and Wednesday at the Technology Strategy Workshop (that I think Jeff Frankel and Adele Morris are attending on Tuesday). If, for some reason, there is a problem with the data on the disk or if you have any questions, give Ron Sands a call at 646-7791. He has copies of the results. Or you can leave me a voice mail message and I'll get back to you. The runs on the disk are: sv5s.xls $10 tax on Annex I in 2005, $30 in 2010, $50 in 2015 and hold at $50 thereafter (Non-Annex I BAU) sv6s.xls $10 tax on Annex I in 2005, $30 in 2010, $50 in 2015 and grow at same rate thereafter (Non-Annex I BAU) sv7s.xls $10 tax on Annex Lin 2005. ramp to $50 in 2020, and hold at $50 thereafter (Non-Annex I BAU) sv8s.xls $10 tax on Annex I in 2005. ramp to $50 in 2020, and grow at same rate thereafter (Non-Annex I BAU) Enjoy! Chris