# Demand Response Compensation in Organized Wholesale Energy Markets > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/fr%3A2011-6490 ## Record - **Collection:** Federal Register - **Document type:** Rule - **Published:** March 24, 2011 - **Citation:** 76 FR 16658 ## Text DEPARTMENT OF ENERGY Federal Energy Regulatory Commission 18 CFR Part 35 [Docket No. RM10-17-000; Order No. 745] Demand Response Compensation in Organized Wholesale Energy Markets AGENCY: Federal Energy Regulatory Commission, Energy. ACTION: Final rule. SUMMARY: In this Final Rule, the Federal Energy Regulatory Commission (Commission) amends its regulations under the Federal Power Act to ensure that when a demand response resource participating in an organized wholesale energy market administered by a Regional Transmission Organization (RTO) or Independent System Operator (ISO) has the capability to balance supply and demand as an alternative to a generation resource and when dispatch of that demand response resource is cost-effective as determined by the net benefits test described in this rule, that demand response resource must be compensated for the service it provides to the energy market at the market price for energy, referred to as the locational marginal price (LMP). This approach for compensating demand response resources helps to ensure the competitiveness of organized wholesale energy markets and remove barriers to the participation of demand response resources, thus ensuring just and reasonable wholesale rates. DATES: Effective Date: This Final Rule will become effective on April 25, 2011. Dates for compliance and other required filings are provided in the Final Rule. FOR FURTHER INFORMATION CONTACT: David Hunger (Technical Information), Office of Energy Policy and Innovation, Federal Energy Regulatory Commission, 888 First Street, NE., Washington, DC 20426, (202) 502-8148, david.hunger@ferc.gov; Dennis Hough (Legal Information), Office of the General Counsel, Federal Energy Regulatory Commission, 888 First Street, NE., Washington, DC 20426, (202) 502-8631, dennis.hough@ferc.gov. SUPPLEMENTARY INFORMATION: Table of Contents (Issued March 15, 2011) Paragraph Nos. I. Introduction 1 II. Background 8 III. Procedural History 15 IV. Discussion 17 A. Compensation Level 18 1. NOPR Proposal 18 2. Comments 20 (a) Capability of Demand Response and Generation Resources to Balance Energy Markets 20 (b) Appropriateness of a Net Benefits Test 38 (c) Standardization or Regional Variations in Compensation 43 3. Commission Determination 45 B. Implementation of a Net Benefits Test 68 1. Comments 68 2. Commission Determination 78 C. Measurement and Verification 86 1. NOPR Proposal 86 2. Comments 88 3. Commission Determination 93 D. Cost Allocation 96 1. NOPR Proposal 96 2. Comments 97 3. Commission Determination 99 E. Commission Jurisdiction 103 1. Comments 103 2. Commission Determination 112 V. Information Collection Statement 116 VI. Environmental Analysis 121 VII. Regulatory Flexibility Act 122 VIII. Document Availability 130 IX. Effective Date and Congressional Notification 133 Regulatory Text Appendix 1—List of Commenters Appendix 2—Dissenting Statement Before Commissioners: Jon Wellinghoff, Chairman; Marc Spitzer, Philip D. Moeller, John R. Norris, and Cheryl A. LaFleur. I. Introduction 1. This Final Rule addresses compensation for demand response in Regional Transmission Organization (RTO) and Independent System Operator (ISO) organized wholesale energy markets, i.e., the day-ahead and real-time energy markets. As the Commission has previously recognized, a market functions effectively only when both supply and demand can meaningfully participate. The Commission, in the Notice of Proposed Rulemaking (NOPR) issued in this proceeding on March 18, 2010, proposed a remedy to concerns that current compensation levels inhibited meaningful demand-side participation. 1 After nearly 3,800 pages of comments, a subsequent technical conference, and the opportunity for additional comment, we now take final action. 1 Demand Response Compensation in Organized Wholesale Energy Markets, Notice of Proposed Rulemaking, 75 FR 15362 (Mar. 29, 2010), FERC Stats. & Regs. ¶ 32,656 (2010) (NOPR). 2. We conclude that when a demand response 2 resource 3 participating in an organized wholesale energy market 4 administered by an RTO or ISO has the capability to balance supply and demand as an alternative to a generation resource and when dispatch of that demand response resource is cost-effective as determined by the net benefits test described herein, that demand response resource must be compensated for the service it provides to the energy market at the market price for energy, referred to as the locational marginal price (LMP). 5 The Commission finds that this approach to compensation for demand response resources is necessary to ensure that rates are just and reasonable in the organized wholesale energy markets. Consistent with this finding, this Final Rule adds section 35.28(g)(1)(v) to the Commission's regulations to establish a specific compensation approach for demand response resources participating in the organized wholesale energy markets administered by RTOs and ISOs. The Commission is not requiring the use of this compensation approach when demand response resources do not satisfy the capability and cost-effectiveness conditions noted above. 6 2 Demand response means a reduction in the consumption of electric energy by customers from their expected consumption in response to an increase in the price of electric energy or to incentive payments designed to induce lower consumption of electric energy. 18 CFR 35.28(b)(4) (2010). 3 Demand response resource means a resource capable of providing demand response. 18 CFR 35.28(b)(5). 4 The requirements of this final rule apply only to a demand response resource participating in a day-ahead or real-time energy market administered by an RTO or ISO. Thus, this Final Rule does not apply to compensation for demand response under programs that RTOs and ISOs administer for reliability or emergency conditions, such as, for instance, Midwest ISO's Emergency Demand Response, NYISO's Emergency Demand Response Program, and PJM's Emergency Load Response Program. This Final Rule also does not apply to compensation in ancillary services markets, which the Commission has addressed elsewhere. See, e.g., Wholesale Competition in Regions with Organized Electric Markets, Order No. 719, 73 FR 64100 (Oct. 28, 2008), FERC Stats. & Regs. ¶ 31,281 (2008) (Order No. 719). 5 LMP refers to the price calculated by the ISO or RTO at particular locations or electrical nodes or zones within the ISO or RTO footprint and is used as the market price to compensate generators. There are variations in the way that RTOs and ISOs calculate LMP; however, each method establishes the marginal value of resources in that market. Nothing in this Final Rule is intended to change RTO and ISO methods for calculating LMP. 6 The Commission's findings in this Final Rule do not preclude the Commission from determining that other approaches to compensation would be acceptable when these conditions are not met. 3. This cost-effectiveness condition, as determined by the net benefits test described herein, recognizes that, depending on the change in LMP relative to the size of the energy market, dispatching demand response resources may result in an increased cost per unit ($/MWh) to the remaining wholesale load associated with the decreased amount of load paying the bill. This is the case because customers are billed for energy based on the units, MWh, of electricity consumed. We refer to this potential result as the billing unit effect of dispatching demand response. By contrast, dispatching generation resources does not produce this billing unit effect because it does not result in a decrease of load. To address this billing unit effect, the Commission in this Final Rule requires the use of the net benefits test described herein to ensure that the overall benefit of the reduced LMP that results from dispatching demand response resources exceeds the cost of dispatching and paying LMP to those resources. When the net benefits test described herein is satisfied and the demand response resource clears in the RTO's or ISO's economic dispatch, the demand response resource is a cost-effective alternative to generation resources for balancing supply and demand. 4. To implement the net benefits test described herein, we direct each RTO and ISO to develop a mechanism as an approximation to determine a price level at which the dispatch of demand response resources will be cost-effective. The RTO or ISO should determine, based on historical data as a starting point and updated for changes in relevant supply conditions such as changes in fuel prices and generator unit availability, the monthly threshold price corresponding to the point along the supply stack beyond which the overall benefit from the reduced LMP resulting from dispatching demand response resources exceeds the cost of dispatching and paying LMP to those resources. This price level is to be updated monthly, by each ISO or RTO, as the historic data and relevant supply conditions change. 7 7 In its compliance filing an RTO or ISO may attempt to show, in whole or in part, how its proposed or existing practices are consistent with or superior to the requirements of this Final Rule. 5. This Final Rule also sets forth a method for allocating the costs of demand response payments among all customers who benefit from the lower LMP resulting from the demand response. 6. The tariff changes needed to implement the compensation approach required in this Final Rule, including the net benefits test, measurement and verification explanation and proposed changes, and the cost allocation mechanism must be made on or before July 22, 2011. All tariff changes directed herein should be submitted as compliance filings pursuant to this Final Rule, not pursuant to section 205 of the Federal Power Act (FPA). 8 Accordingly, each RTO's or ISO's compliance filing to this Final Rule will become effective prospectively from the date of the Commission order addressing that filing, and not within 60 days of submission. 8 16 U.S.C. 824d (2006). 7. In addition, we believe that integrating a determination of the cost-effectiveness of demand response resources into the dispatch of the ISOs and RTOs may be more precise than the monthly price threshold and, therefore, provide the greatest opportunity for load to benefit from participation of demand response in the organized wholesale energy market administered by an RTO or ISO. However, we acknowledge the position of several of the RTOs and ISOs that modification of their dispatch algorithms to incorporate the costs related to demand response may be difficult in the near term. In light of those concerns, we require each RTO and ISO to undertake a study examining the requirements for and impacts of implementing a dynamic approach which incorporates the billing unit effect in the dispatch algorithm to determine when paying demand response resources the LMP results in net benefits to customers in both the day-ahead and real-time energy markets. The Commission directs each RTO and ISO to file the results of this study with the Commission on or before September 21, 2012. 9 9 We note that this report is for informational purposes only and will neither be noticed nor require Commission action. II. Background 8. Effective wholesale competition protects customers by, among other things, providing more supply options, encouraging new entry and innovation, and spurring deployment of new technologies. 10 Improving the competitiveness of organized wholesale energy markets is therefore integral to the Commission fulfilling its statutory mandate under the FPA to ensure supplies of electric energy at just, reasonable, and not unduly discriminatory or preferential rates. 11 10 See, e.g., Wholesale Competition in Regions with Organized Electric Markets, Order No. 719, 73 FR 64100 (Oct. 28, 2008), FERC Stats. & Regs. ¶ 31,281, at P 1 (2008) (Order No. 719); see also Regional Transmission Organizations, Order No. 2000, FERC Stats. & Regs. ¶ 31,089, at P 1 (1999), order on reh'g, Order No. 2000-A, FERC Stats. & Regs. ¶ 31,092 (2000), aff'd sub nom. Pub. Util. Dist. No. 1 of Snohomish County, Washington v. FERC, 272 F.3d 607, 348 U.S. App. DC 205 (DC Cir. 2001). 11 16 U.S.C. 824d (2006); Order No. 719, FERC Stats. & Regs. ¶ 31,281 at P 1. 9. As the Commission recognized in Order No. 719, active participation by customers in the form of demand response in organized wholesale energy markets helps to increase competition in those markets. 12 Demand response, whereby customers reduce electricity consumption from normal usage levels in response to price signals, can generally occur in two ways: (1) Customers reduce demand by responding to retail rates that are based on wholesale prices (sometimes called “price-responsive demand”); and (2) customers provide demand response that acts as a resource in organized wholesale energy markets to balance supply and demand. While a number of States and utilities are pursuing retail-level price-responsive demand initiatives based on dynamic and time-differentiated retail prices and utility investments in demand response enabling technologies, these are State efforts, and, thus, are not the subject of this proceeding. Our focus here is on customers or aggregators of retail customers providing, through bids or self-schedules, demand response that acts as a resource in organized wholesale energy markets. 12 See Order No. 719, FERC Stats. & Regs. ¶ 31,281 at P 48. 10. As the Commission stated in Order No. 719, 13 and emphasized in the NOPR, 14 there are several ways in which demand response in organized wholesale energy markets can help improve the functioning and competitiveness of those markets. First, when bid directly into the wholesale market, demand response can facilitate RTOs and ISOs in balancing supply and demand, and thereby, help produce just and reasonable energy prices. 15 This is because customers who choose to respond will signal to the RTO or ISO and energy market their willingness to reduce demand on the grid which may result in reduced dispatch of higher-priced resources to satisfy load. 16 Second, demand response can mitigate generator market power. 17 This is because the more demand response that sees and responds to higher market prices, the greater the competition, and the more downward pressure it places on generator bidding strategies by increasing the risk to a supplier that it will not be dispatched if it bids a price that is too high. 18 Third, demand response has the potential to support system reliability and address resource adequacy 19 and resource management challenges surrounding the unexpected loss of generation. This is because demand response resources can provide quick balancing of the electricity grid. 20 13 Wholesale Competition in Regions with Organized Electric Markets, Order No. 719-A, FERC Stats. & Regs. ¶ 31,292, at P 48 (2009). 14 NOPR, FERC Stats. & Regs. ¶ 32,656 at P 4. 15 For example, a study conducted by PJM, which simulated the effect of demand response on prices, demonstrated that a modest three percent load reduction in the 100 highest peak hours corresponds to a price decline of six to 12 percent. ISO-RTO Council Report, Harnessing the Power of Demand How RTOs and ISOs Are Integrating Demand Response into Wholesale Electricity Markets, found at http://www.isorto.org/atf/cf/%7B5B4E85C6-7EAC-40A0-8DC3-003829518EBD%7D/IRC_DR_Report_101607.pdf. 16 Id. (“Demand response tends to flatten an area's load profile, which in turn may reduce the need to construct and use more costly resources during periods of high demand; the overall effect is to lower the average cost of producing energy.”). 17 See Comments of NYISO's Independent Market Monitor filed in Docket No. ER09-1142-000, May 15, 2009 (Demand response “contributes to reliability in the short-term, resource adequacy in the long-term, reduces price volatility and other market costs, and mitigates supplier market power.”). 18 Id. 19 See ISO-RTO Council Report, Harnessing the Power of Demand How RTOs and ISOs Are Integrating Demand Response into Wholesale Electricity Markets at 4, found at http://www.isorto.org/atf/cf/%7B5B4E85C6-7EAC-40A0-8DC3-003829518EBD%7D/IRC_DR_Report_101607.pdf (“Demand response contributes to maintaining system reliability. Lower electric load when supply is especially tight reduces the likelihood of load shedding. Improvements in reliability mean that many circumstances that otherwise result in forced outages and rolling blackouts are averted, resulting in substantial financial savings * * *.”). 20 For instance, in ERCOT, on February 26, 2008, through a combination of a sudden loss of thermal generation, drop in power supplied by wind generators, and a quicker-than-expected ramping up of demand, ERCOT found itself short of reserves. The system operator called on all demand response resources, and 1200 MW of Load acting as Resource (LaaRs) responded quickly, bringing ERCOT back into balance. Oak Ridge Nat'l Lab., Nat'l Renewable Energy Lab., Tech. Rep. NREL/TP-500-43373, ERCOT Event on Feb. 26, 2008: Lessons Learned (Jul. 2008). 11. Congress has recognized the importance of demand response by enacting national policy requiring its facilitation. 21 Consistent with that policy, the Commission has undertaken several reforms to support competitive wholesale energy markets by removing barriers to participation of demand response resources. For example, in Order No. 890, the Commission modified the pro forma Open Access Transmission Tariff to allow non-generation resources, including demand response resources, to be used in the provision of certain ancillary services where appropriate on a comparable basis to service provided by generation resources. 22 Order No. 890-A further required transmission providers to develop transmission planning processes that treat all resources, including demand response, on a comparable basis. 23 21 See Energy Policy Act of 2005, Public Law 109-58, § 1252(f), 119 Stat. 594, 965 (2005) (“It is the policy of the United States that * * * unnecessary barriers to demand response participation in energy, capacity, and ancillary service markets shall be eliminated.”). 22 Preventing Undue Discrimination and Preference in Transmission Service, Order No. 890, FERC Stats. & Regs. ¶ 31,241, at P 887-88 (2007), order on reh'g, Order No. 890-A, FERC Stats. & Regs. ¶ 31,261 (2007), order on reh'g and clarification, Order No. 890-B, 123 FERC ¶ 61,299 (2008), order on reh'g, Order No. 890-C, 126 FERC ¶ 61,228 (2009), order on clarification, Order No. 890-D, 129 FERC ¶ 61,126 (2009). 23 Order No. 890-A, FERC Stats. & Regs. ¶ 31,261 at P 216. 12. In Order No. 719, the Commission required RTOs and ISOs to, among other things, accept bids from demand response resources in their markets for certain ancillary services on a basis comparable to other resources. 24 The Commission also required each RTO and ISO “to reform or demonstrate the adequacy of its existing market rules to ensure that the market price for energy reflects the value of energy during an operating reserve shortage,” 25 for purposes of encouraging existing generation and demand resources to continue to be relied upon during an operating reserve shortage, and encouraging entry of new generation and demand resources. 26 24 Order No. 719, FERC Stats. & Regs. ¶ 31,281 at P 47-49. 25 Order No. 719, FERC Stats. & Regs. ¶ 31,281 at P 194. 26 Order No. 719, FERC Stats. & Regs. ¶ 31,281 at P 247. 13. Additionally, in recent years several RTOs and ISOs have instituted various types of demand response programs. While some of these programs are administered for reliability and emergency conditions, other programs allow wholesale customers, qualifying large retail customers, and aggregators of retail customers to participate directly in the day-ahead and real-time energy markets, certain ancillary service markets and capacity markets. 27 27 Other demand response programs allow demand response to be used as a capacity resource and as a resource during system emergencies or permit the use of demand response for synchronized reserves and regulation service. See, e.g., PJM Interconnection, L.L.C., 117 FERC ¶ 61,331 (2006); Devon Power LLC, 115 FERC ¶ 61,340, order on reh'g, 117 FERC ¶ 61,133 (2006), appeal pending sub nom. Maine Pub. Utils. Comm'n v. FERC, No. 06-1403 (D.C. Cir. 2007); New York Indep. Sys. Operator, Inc., 95 FERC ¶ 61,136 (2001); NSTAR Services Co. v. New England Power Pool, 95 FERC ¶ 61,250 (2001); New England Power Pool and ISO New England, Inc., 100 FERC ¶ 61,287, order on reh'g, 101 FERC ¶ 61,344 (2002), order on reh'g, 103 FERC ¶ 61,304, order on reh'g, 105 FERC ¶ 61,211 (2003); PJM Interconnection, L.L.C., 99 FERC ¶ 61,227 (2002); California Independent System Operator Corp., 132 FERC ¶ 61,045 (2010). 14. To date, the Commission has allowed each RTO and ISO to develop its own compensation methodologies for demand response resources participating in its day-ahead and real-time energy markets. As a result, the levels of compensation for demand response vary significantly among RTOs and ISOs. 28 For example, PJM Interconnection, L.L.C. (PJM) pays the LMP minus the generation and transmission portions of the retail rate. 29 ISO New England Inc. (ISO-NE) and New York Independent System Operator, Inc. (NYISO) pay LMP when prices exceed a threshold level, with the levels differing between the RTOs. 30 The Midwest Independent Transmission System Operator, Inc.'s (Midwest ISO) demand response programs 31 pay LMP for demand response resources in the day-ahead and real-time energy markets. 32 The California Independent System Operator Corporation (CAISO) pays LMP at pricing nodes, or sub-load aggregation points (Sub-LAP) in its Proxy Demand Resource program that allows qualifying resources to provide day-ahead and real-time energy. 33 CAISO also provides for demand response resources to participate in its Participating Load program, which enables certain resources to provide curtailable demand in the CAISO market. CAISO pays nodal real-time LMP for its Participating Load program. The Southwest Power Pool, Inc. (SPP) has filed revisions to its tariff to facilitate demand response in the Energy Imbalance Service Market. 34 28 See New England, Inc., Docket No. ER09-1051-000; ISO New England, Inc., Docket No. ER08-830-000; Midwest Indep. Transmission Sys. Operator, Inc., Docket No. ER09-1049-000. 29 See sections 3.3A.4 and 3.3A.5 (Market Settlements in the Real-Time and Day-Ahead Energy Markets) of the Appendix to Attachment K of the PJM Tariff. 30 For example, under ISO-NE's Real-Time Price Response Program, the minimum bid is $100/MWh and a demand response resource is paid the higher of LMP or $100/MWh. For the Day-Ahead Load Response Program, the minimum offer level is calculated on a monthly basis and is the Forward Reserve Fuel Index ($/MMBtu) multiplied by an effective heat rate of 11.37 MMBtu/MWh. The maximum offer level is $1,000/MWh. See sections III.E.2.1 and III.E.3.2 of Appendix E of the ISO New England Transmission, Markets and Services Tariff. NYISO implements a day-ahead demand response program by which resources bid into the market at a minimum of $75/MWh and can get paid the LMP. See section 4.2.2.9 (“Day-Ahead Bids from Demand Reduction Providers to Supply Energy from Demand Reductions”) of NYISO's Market Services Tariff. 31 Midwest ISO FERC Electric Tariff characterizes Demand Response Resources (DRR) as either DRR-Type I or DRR-Type II. DRR-Type I are capable of supplying a specific quantity of energy or contingency reserve through physical load interruption. DRR-Type II are capable of supplying energy and/or operating reserves over a dispatchable range. See sections 39.2.5A and 40.2.5 of the Tariff. 32 See Charges and Payments for Purchases and Sales for Demand Response Resources. Midwest ISO FERC Electric Tariff, section 39.3.2C. 33 See section 11.2.1.1 IFM Payments for Supply of Energy, CAISO FERC Electric Tariff. CAISO notes that for a Proxy Demand Resource that is made up of aggregated loads, the Resource is paid the weighted average of the LMPs of each pricing node where the underlying aggregate loads reside. See CAISO, 132 FERC ¶ 61,045, at P 26 n.14 (2010). 34 The Commission has directed SPP to report on ways it can incorporate demand response into its imbalance market. Southwest Power Pool, Inc., 128 FERC ¶ 61,085 (2009). As of September 1, 2010, SPP has submitted seven informational status reports regarding its efforts to address issues related to demand response resources. In orders addressing SPP's compliance with Order No. 719, the Commission also directed SPP to make another compliance filing addressing demand response participation in its organized markets. Southwest Power Pool, Inc., 129 FERC ¶ 61,163, at P 51 (2009). On May 19, 2010, SPP submitted revisions to its Open Access Transmission Tariff in Docket Nos. ER09-1050-004 and ER09-748-002 to comply with the Commission's requirements established in Order Nos. 719 and 719-A. These filings are pending before the Commission. III. Procedural History 15. As noted above, the Commission issued the NOPR in this proceeding on March 18, 2010. 35 The NOPR proposed to require RTOs and ISOs to pay the LMP in all hours for demand reductions made in response to price signals. The Commission sought comments on the compensation proposal and, in particular, on the comparability of generation and demand response resources; alternative approaches to compensating demand response in organized wholesale energy markets; whether payment of LMP should apply in all hours, and, if not, any criteria that should be used for establishing hours when LMP should apply; and whether to allow for regional variations concerning approaches to demand response compensation. 36 35 NOPR, FERC Stats. & Regs. ¶ 32,656. 36 See Appendix for a list of commenters. 16. After receiving the first round of comments, the Commission issued a Supplemental Notice of Proposed Rulemaking and Notice of Technical Conference (Supplemental NOPR) in this proceeding on August 2, 2010. 37 The Supplemental NOPR sought additional comment on: Whether the Commission should adopt a net benefits test for determining when to compensate demand response providers, and, if so, what, if any, requirements should apply to the methods for determining net benefits; and what, if any, requirements should apply to how the costs of demand response are allocated. The Commission further directed Staff to hold a technical conference focused on these two issues, which occurred on September 13, 2010. 38 37 Supplemental Notice of Proposed Rulemaking and Notice of Technical Conference, 75 FR 47499 (Aug. 6, 2010), 132 FERC ¶ 61,094 (2010) (Supplemental NOPR). 38 See Notice of Technical Conference (Aug. 27, 2010). IV. Discussion 17. Based upon the record in this proceeding, the Commission herein requires greater uniformity in compensating demand response resources participating in organized wholesale energy markets. This Final Rule also addresses the allocation of costs resulting from the commitment of demand response, directing that such costs be allocated among those customers who benefit from the lower LMP resulting from the demand response. A. Compensation Level 1. NOPR Proposal 18. The NOPR proposed to require RTOs and ISOs to pay the LMP in all hours for demand reductions made in response to price signals. The NOPR sought to provide comparable compensation to generation and demand response providers, based on the premise that both resources provide a comparable service to RTOs and ISOs for purposes of balancing supply and demand and maintaining a reliable electricity grid. 39 Also as stated in the NOPR, the proposed compensation level was designed to allow more demand response resources to cover their investment costs in demand response-related technology (such as advanced metering) and thereby facilitate their ability to participate in organized wholesale energy markets. 40 The Commission sought comments on the compensation proposal and, in particular, on the comparability of generation and demand response resources; alternative approaches to compensating demand response in organized wholesale energy markets; whether payment of LMP should apply in all hours, and, if not, any criteria that should be used for establishing hours when LMP should apply; and whether to allow for regional variations concerning approaches to demand response compensation. 39 NOPR, FERC Stats. & Regs. ¶ 32,656 at P 15. 40 Id. at P 16. 19. In the Supplemental NOPR, the Commission sought additional comments and directed staff to hold a technical conference regarding various net benefits tests. In particular, the Commission sought comment on: whether the Commission should adopt a net benefits test applicable in all or only some hours and what the criteria of any such test would be; how to define net benefits; what costs demand response providers and load serving entities incur and whether they should be included in a net benefits test; whether any net benefits methodology adopted should be the same for all RTOs and ISOs; proposed methodologies for implementing a net benefits test and the advantages and limitations of any proposed methodologies. 41 The September 13, 2010 Technical Conference included an eleven-member panel discussion of net benefits tests representing a wide range of interests and viewpoints. 42 The Commission subsequently received additional written comments addressing these issues. 41 Supplemental NOPR, 132 FERC ¶ 61,094 at P 8-9. 42 See Sept. 13, 2010 Tr. 2. Comments (a) Capability of Demand Response and Generation Resources To Balance Energy Markets 20. Various commenters address the comparability of demand response and generation resources for purposes of compensation in the organized wholesale energy markets. To begin, numerous commenters address the physical or functional comparability of demand response and generation, agreeing that an increment of generation is comparable to a decrement of load for purposes of balancing supply and demand in the day-ahead and real-time energy markets. 43 Equating generation and demand response resources, Dr. Alfred E. Kahn states: 43 DR Supporters Aug. 30, 2010 Comments (Kahn Affidavit at 2); Verso May 13, 2010 Comments at 3-4; Occidental May 13, 2010 Comments at 11; Viridity June 18, 2010 Comments at 5. [Demand response] is in all essential respects economically equivalent to supply response * * * [so] economic efficiency requires * * * that it should be rewarded with the same LMP that clears the market. Since [demand response] is actually—and not merely metaphorically—equivalent to supply response, economic efficiency requires that it be regarded and rewarded, equivalently, as a resource proffered to system operators, and be treated equivalently to generation in competitive power markets. That is, all resources—energy saved equivalently to energy supplied—* * * should receive the same market-clearing LMP in remuneration. 44 44 DR Supporters August 30, 2010 Reply Comments (Kahn Affidavit at 2 (footnote omitted)). Indeed, some commenters believe that, from a physical standpoint, demand response can provide superior services to generation, such as providing a quick response in meeting system requirements and service without having to construct major new facilities. 45 Occidental asserts that the fungibility of demand response and generation output creates greater operational flexibility that, in turn, offers RTOs and ISOs multiple options to solve system issues both in energy and ancillary service markets, and that the fungible nature of demand response and generation supports comparable compensation for each as proposed in the NOPR. 46 45 Verso May 13, 2010 Comments at 3-4; Alcoa May 13, 2010 Comments at 9. 46 Occidental May 13, 2010 Comments at 11. 21. Viridity states that attempts to distinguish the physical characteristics of generation and demand response ignore bid-based security-constrained economic dispatch as the foundation for LMP and are based on the assumption that the value of load management on the grid is limited to periods when the system is stressed, i.e., traditional “super peak shaving.” Viridity states that, while these arguments might have been valid 15 years ago, today competitive markets can offer proactively-managed load control and comparable and non-discriminatory treatment of load-based energy resources. Therefore, Viridity asserts that all resources should be paid LMP if the grid operator accepts their bid to achieve grid balance. 47 47 Viridity June 18, 2010 Comments at 5. 22. At the same time, other commenters argue that generation and demand response are not physically equivalent, pointing out that demand response reduces consumption, whereas generators serve consumption. 48 They argue that a MW reduction in demand does not turn on the lights. 49 EPSA adds that a load reduction does not provide electrons to any other load and, instead, allows the marginal electron to serve a different customer. 50 Some commenters assert that a power system can function solely and reliably on generating plants and without any reliance on demand response, while the system cannot rely exclusively on demand response because demand response by itself cannot keep the lights on. Ultimately, some commenters point out, megawatts produced by generators need to be placed on the system in order for power to flow. 51 Battelle additionally argues that a reduction in consumption is not exactly the same as an increase in production, because elastic demand often comes with attendant future consequences, such as rebound, by virtue of substitution in time. 52 48 ISO-NE May 13, 2010 Comments at 3. 49 See, e.g., APPA May 13, 2010 Comments at 12; Capital Power May 13, 2010 Comments at 2. 50 EPSA May 13, 2010 Comments at 72. 51 See, e.g., PSEG May 13, 2010 Comments at 8. 52 Battelle May 13, 2010 Comments at 3. 23. Some commenters who argue that the physical characteristics of demand response are not comparable to generation frame their arguments in terms of the ability of the system operator to call on demand response and generation resources to provide balancing energy. They argue that generation resources provide superior service to demand response providers, positing that demand response is not intended for long periods of balancing needs, 53 and that, moreover, contracts with demand response providers limit the number of hours and times a customer may be called upon to curtail. For example, ODEC asserts that the degree of physical comparability depends on the extent to which demand response resources can be dispatched similar to a generator. 54 Calpine adds that traditional generators provide system support features that demand response cannot, such as ancillary services including governor response or reactive power voltage support, which are necessary for reliable operation of the electric system. 55 53 AEP May 13, 2010 Comments at 7-8. 54 ODEC May 13, 2010 Comments at 12. 55 Calpine May 13, 2010 Comments at 4-5. 24. Numerous commenters also address the comparability of demand response and generation in economic terms. For example, EEI states that, in finance terms, the demand response product is, unlike generation, essentially an unexercised call option on spot market energy, and the value of that option is well-established in finance theory as the value of the resource (LMP) minus the “strike price,” which EEI contends in this case is the retail tariff rate. 56 EEI and like-minded commenters support, therefore, alternative compensation for demand response to equal LMP minus the generation (or G) component of the retail rate. 57 They posit that payment of LMP without an offset for some portion of the retail rate does not send the proper economic signal to providers of demand response, because it fails to take into account the retail rate savings associated with demand response, and thereby overcompensates the demand response provider. As described by Dr. William W. Hogan on behalf of EPSA, this is sometimes called a double-payment for demand reductions, because demand response providers would “receive” both the cost savings from not consuming an increment of electricity at a particular price, plus an LMP payment for not consuming that same increment of electricity. 58 Viewing LMP as a double-payment, these commenters argue that paying LMP will result in more demand response than is economically efficient. 59 For example, Dr. Hogan states that paying LMP might motivate a company to shut down even though the benefits of consuming electricity outweigh the cost at LMP. 60 Indeed, P3 argues that compensation in excess of LMP-G is unjust and unreasonable, because such a payment level imposes costs on customers that are not commensurate with benefits received. 61 56 EEI May 13, 2010 Comments at 4-5. See also Robert L. Borlick May 13, 2010 Comments at 4. Mr. Borlick argues that the correct price is LMP minus the Marginal Foregone Retail Rate (MFRR), describing the economically efficient price that should be paid to a demand response provider as “its offer price minus the price in its retail tariff at which it would have purchased the curtailed energy.” Mr. Borlick asserts that this amount accurately represents the forgone opportunity costs that result when a demand response provider reduces its load. Id. 57 See May 13, 2010 Comments of: APPPA; AEP; The Brattle Group; Calpine; ConEd; Consumers Energy; CPG; Detroit Edison; Direct Energy; Dominion; Duke Energy; Edison Mission; EEI; EPSA; Exelon; FTC; GDF; NYISO on behalf of the ISO RTO Council; ICC; IPPNY; Indicated New York TOs; IPA; ISO-NE; Midwest TDUs; Mirant; Midwest ISO TOs; NEPGA; NYISO; ODEC; OMS; PJM; PJM IMM; P3; Potomac Economics; PG&E; Ohio Commission; Robert L. Borlick; Roy Shanker; and RRI Energy. 58 See Attachment to Answer of EPSA, Providing Incentives for Efficient Demand Response, Dr. William W. Hogan, Oct. 29, 2009, submitted in Docket No. EL09-68-000. 59 EPSA May 13, 2010 Comments at 23. See also May 13, 2010 Comments of APPA at 13; FTC at 9; Midwest TDUs at 14; Mirant at 2; New York Commission at 5; PJM at 6; PSEG at 5; and Potomac Economics at 6-8. 60 Attachment to Answer of EPSA, Providing Incentives for Efficient Demand Response, Dr. William W. Hogan, Oct. 29, 2009, submitted in Docket No. EL09-68-000. In Dr. Hogan's view, supply should produce when the price of electricity exceeds its cost of production and demand should decline to consume when the costs in terms of convenience of delaying use are less than the price of electricity. 61 P3 June 14, 2010 Comments at 2, 7-8. 25. ISO-NE argues that paying full LMP to demand response providers without taking into account the bill savings produced by demand response provides a significant financial incentive to dispatch demand response with marginal costs exceeding LMPs. By dispatching higher-cost demand response, ISO-NE asserts, lower-cost generation resources are displaced. 62 At the same time, ISO-NE argues, generation is not dispatched and paid for only when the generation reduces LMP—generation is dispatched and paid for when it is cost-effective. 63 62 ISO-NE May 13, 2010 Comments at 3-4. 63 Id. at 28. 26. Dr. Hogan further disputes arguments equating a MW of energy supplied to a MW of energy saved on economic grounds. Dr. Hogan draws a distinction between reselling something that one has purchased, and selling something that one would have purchased without actually purchasing it. Dr. Hogan argues that from the perspective of economic efficiency and welfare maximization, the aggregate effect of demand response is a wash producing no economic net benefit. Dr. Hogan asserts that Commission policy citing the benefits of price reduction in support of demand response compensation would amount to no less than an application of regulatory authority to enforce a buyers' cartel. He states that the Commission has been vigilant and aggressive in preventing buyers and sellers from engaging in market manipulation to influence prices, and it would be fundamentally inconsistent for the Commission to design demand response compensation policies that coordinate and enforce such price manipulation. 27. Dr. Hogan argues that the ideal and economically efficient solution regarding demand response compensation is to implement retail real-time pricing at the LMP, thereby eliminating the need for demand response programs. Realizing that this is unattainable at the present time, Dr. Hogan goes on to propose a next-best solution, which he believes is to pay demand response compensation in the amount of LMP-G, or some amount that simulates explicit contract demand response (such as “buy-the-baseline” approach discussed below). These options, he argues, more than paying LMP, better support notions of comparability between demand response resources and generation. 64 64 Hogan Affidavit, ISO RTO Council May 13, 2010 Comments at 5. 28. The New York Commission, however, argues that requiring payment of LMP-G would result in an administrative burden of tracking retail rates for the multiple utilities, ESCOs and power authorities and create undue confusion for retail customers and administrative difficulties for State commissions and ISOs and RTOs. 65 65 New York Commission May 13, 2010 Comments at 8. 29. Consistent with Dr. Hogan's arguments, some commenters assert that demand response providers should actually own or pay for electricity prior to, what commenters characterize as, an effective reselling of the electricity back to the market in the form of demand response. For example, these commenters suggest that the demand response provider purchase the power in the day-ahead market and resell it in the real-time markets. 66 EPSA argues that there must be some purchase requirement or representative offset to allow a demand response provider to “sell” a commodity that it owns to the ISO or RTO. 67 EPSA argues that such a requirement would send an efficient price signal, reduce incentives for gaming the system, and help address difficulties with measurement and verification of a demand reduction. EPSA highlights an ISO-NE IMM recommendation that, if the Commission permits LMP payment, it should also adopt a “buy-the-baseline” approach requiring demand response resources to purchase an expected amount of energy consumption in the day-ahead energy market and subsequently sell any demand reduction from that level in the real-time market. 68 66 See, e.g., ISO-NE IMM May 13, 2010 Comments at 4-5; Midwest ISO TOs May 13, 2010 Comments at 14; PJM May 13, 2010 Comments at 5; and Duke Energy May 13, 2010 Comments at 2. 67 EPSA June 30, 2010 Comments at 3. 68 EPSA June 30, 2010 Comments at 23. 30. Viridity, on the other hand, argues that forcing customers to buy and then resell electricity will lead to too little demand response and that adopting a “buy-the-baseline” approach would constitute an inappropriate exercise of Commission authority to effectively force parties into contracts. Viridity and DR Supporters state that any characterization of demand response as a purchase and then resale of energy is erroneous 69 and based on the flawed assumption that demand response resources are reselling energy. They state that the description of demand response as a reselling of energy has been correctly rejected by the Commission in EnergyConnect, where the Commission stated that it was establishing a policy of treating demand response as a service rather than a purchase and sale of electric energy. 70 69 Viridity Energy June 18, 2010 Comments at 25. 70 DR Supporters Aug. 30, 2010 Reply Comments at 10 (citing EnergyConnect, Inc., 130 FERC ¶ 61,031 at P 30-31 (2010)). 31. DR Supporters further argues that, despite claims to the contrary, paying full LMP to demand response providers does not constitute a subsidy for demand response any more than the remunerations of generators for the power that they sell. As Dr. Kahn states: Does this plan involve double compensation, as [Dr.] Hogan asserts, at the expense of power generators—of successful bidders promising to induce efficient demand curtailment and of consumers induced to practice it? Certainly not: The decrease in the revenue of the generators is (and consequent savings by consumers are) matched by the savings in their (marginal) costs of generating that power; the successful bidders for the opportunity to induce that consumer response are compensated for the costs of those efforts by the pool, whose (marginal) costs they save by assisting consumers to reduce their purchases. 71 71 DR Supporters Aug. 30, 2010 Reply Comments, Kahn Affidavit at 10. 32. Viridity further disputes Dr. Hogan's argument that payment of LMP for demand response will distort an otherwise optimal market. Viridity posits that such arguments ignore dislocations in the wholesale power markets, the existence of market power that must be mitigated, imperfect information available to customers, barriers to entry and uneconomic resources dispatched to fulfill must-run requirements. 72 Viridity further states that Dr. Hogan's arguments fail to acknowledge the limits of the Commission's jurisdiction and widespread dislocations and distortions in virtually all economic aspects of relevant energy markets (including fuels, facilities, pricing, environmental attributes, information and participation) and fail to account for any market benefits of demand response. 73 Finally, Viridity argues that Dr. Hogan's arguments fail to reflect the many complex interactions between price, equipment operational requirements, and customer processes, which point to a complex demand response decision. 74 72 Viridity June 18, 2010 Comments at 13 (“Importantly, Dr. Hogan (and others) in opposing the proposed rulemaking fails to acknowledge the limits of the Commission's jurisdiction, and wide spread dislocations and distortions in virtually all economic aspects of relevant energy markets (including fuels, facilities, pricing, environmental attributes, information and participation).” (Affidavit of John C. Tysseling, PhD)). 73 Viridity Reply Comments at 13. 74 Viridity Reply Comments at 14. 33. In addition to physical and economic comparability, some commenters contrast the environmental effects of generation and demand response resources. EDF notes that current market prices fail to internalize environmental externalities—including toxic air pollution, greenhouse gas pollution, and land and water use impacts—and other social costs. EDF asserts that the social impact of these environmental externalities is especially acute at peak times, positing that generation sources used for marginal supply at such times (“peaker plants”) are among the oldest, dirtiest, and most inefficient in the fleet. 75 The American Clean Skies Foundation contends that fossil-fuel generators are typically mispriced because wholesale prices radically understate the full environmental and health costs associated with such generators. 76 Indeed, some commenters, such as Alcoa, argue that because demand response does not result in the external costs associated with generation ( e.g., greenhouse gas emissions), instead resulting in less greenhouse gas emissions than generation, it should be compensated at more than LMP. 77 75 EDF Oct. 13, 2010 Comments at 2. 76 American Clean Skies Foundation May 13, 2010 Comments at 4. 77 Alcoa May 13, 2010 Comments at 9. 34. Taking the opposite view concerning environmental externalities, EPSA states that paying LMP for demand response will merely encourage load to switch to off-grid power (or behind-the-meter generation), while still being compensated, and that such behind-the-meter generation produces more greenhouse gases and other air emissions than electricity from the regional energy market. 78 78 EPSA May 13, 2010 Comments at 60. 35. Some commenters discuss comparability of generation and demand response in terms of the market rules that apply to each resource, arguing that both resources should be comparably compensated only if the same rules for participation apply to both resources, and both resources are held to the same standards for dispatchability. 79 They also argue that similar penalty structures should apply to demand response resources as apply to generation, and that demand response participation must be subject to market monitoring. 80 Calpine adds that to the extent demand response resources are used and treated on par with generators for purposes of compensation, they should be subject to the same performance testing, penalties, and other similar requirements as generators. 81 79 ODEC May 13, 2010 Comments at 12; Westar May 13, 2010 Comments at 5-6. 80 Id. 81 Calpine May 13, 2010 Comments at 5. 36. Some commenters address the comparability of demand response providers and generators in terms of maintaining system reliability. PIO argues that reductions in consumption provide additional reliability. 82 According to the NEMA, North American Electric Reliability Corporation (NERC) standards suggest that, from a reliability perspective, load reductions are equivalent or even superior to generator increases for balancing purposes. For example, while specific to the Western Interconnection, BAL-002-WECC-1 lists interruptible load as comparable to generation deployable within 10 minutes. 83 EPSA maintains that demand response resources are not full substitutes based on the nature of their participation and the rules applicable to each resource in the energy markets, pointing out, for example, that, unlike generators, demand response providers are not subject to regional and NERC mandatory reliability standards. 84 82 PIO May 13, 2010 Comments at 8. 83 NEMA May 13, 2010 Comments at 2. 84 EPSA May 13, 2010 Comments at 7. 37. On the other hand, PSEG argues that a MW of demand response does not make the same contribution towards system reliability as a MW of generation, because demand response committed as a capacity resource is only required to perform for a limited number of times over the peak period. PSEG refers to PJM's capacity market, for example, in which demand response only has to perform 10 times during the entire summer peak period, and then only for six hours per response. In contrast, PSEG argues, generators are available for dispatch, 24 hours a day, 365 days per year, except for a small percentage of time for forced and planned outages. PSEG further asserts that additional reliability standards—applicable to generating facilities, but not to demand response—increase the relative reliability value of generating resources to the system. 85 85 PSEG May 13, 2010 Comments at 8. (b) Appropriateness of a Net Benefits Test 38. Some commenters assert that demand response providers should be paid LMP only when the benefits of demand response compensation outweigh the energy market costs to consumers of paying demand response resources, i.e., when cost-effective, as determined by some type of net benefits or cost-effectiveness test. 86 They maintain that paying LMP for demand response in all hours, including off-peak hours, might not result in net benefits to customers, because the payments might be substantially more than the savings created by reducing the clearing price at that time. 87 According to these commenters, net benefits are most likely to be positive and greatest when the supply curve is steepest, which typically occurs in highest-cost, peak hours. 88 They argue that experience to date has shown positive benefits from demand response as a peak system resource, and that, during peak periods, the positive economics of demand response are generally very clear and a cost-benefit analysis may not be needed. 89 Furthermore, some commenters suggest that limiting the hours in which demand response resources are paid LMP could help establish better baselines for measuring whether a demand response provider has, in fact, responded. 90 86 See generally May 13, 2010 Comments of NYSCPB; NECA; Capital Power; NECPUC; Maryland Commission; New York Commission; NSTAR; National Grid; NE Public Systems. 87 Capital Power May 13, 2010 Comments at 5; P3 May 13, 2010 Comments at 5. 88 NECPUC May 13, 2010 Comments at 13; see also Sept. 13, 2010 Tr. 13:6-19 (Mr. Keene); Maryland Commission May 13, 2010 Comments at 4-5. 89 See, e.g., ACEEE Oct. 13, 2010 Comments 3-4. See also National Grid May 13, 2010 Comments at 4-5; NSTAR Electric Company (NSTAR) May 14, 2010 Comments at 3; Maryland Commission May 13, 2010 Comments, submitting Analysis of Load Payments and Expenditures under Different Demand Response Compensation Schemes at 10-11 (discussing PJM analysis showing that paying demand response providers LMP for all hours after compensating LSEs for lost revenues would not benefit customers in general but that positive economic benefits results when demand response providers receive LMP during at least the top 100 hours (the highest priced energy hours)). 90 See, e.g., CDWR May 13, 2010 Comments at 11; National Grid May 13, 2010 Comments at 8; ISO-NE May 13, 2010 Comments at 34; ACEEE Oct. 13, 2010 Comments 4. But see ISO-NE May 13, 2010 Comments at 32-33 (contending that no baseline estimation methodology that relies upon historical customer meter data can accurately and reliably estimate an individual customer's normal energy usage pattern if that customer responds frequently to price signals). 39. Some commenters who oppose paying LMP in all hours for demand response also suggest various approaches, including net benefits tests, for determining when LMP should apply. The stated purpose of any of these tests would be to determine the point at which the incremental payment for demand response equals the incremental benefit of the reduction in load; payment of LMP would apply only up to that point. 91 91 NECAA May 13, 2010 Comments at 11; NYSCPB May 13, 2010 Comments at 5; National Grid May 13, 2010 Comments at 4-5. 40. Opposition to use of a net benefits test comes from several directions. Numerous commenters, primarily industrial consumers and some consumer advocates, argue that a net benefits test will reduce competition, 92 have a “chilling effect” on the development of demand response, 93 and be costly and complex to implement. 94 Some commenters further state that no net benefits test is needed because the merit-order bid stack and market clearing function in a wholesale market, by definition, assures that the benefits to the system of demand response exceed the costs, and that the resource that clears is the lowest cost resource; otherwise, demand response would not dispatch ahead of competing alternatives. 95 92 Viridity Oct. 13, 2010 Comments at 14. 93 NAPP Oct. 13, 2010 Comments at 2. 94 Viridity Oct. 13, 2010 Comments at 14; NAPP Oct. 13, 2010 Comments at 3; AMP Oct. 13, 2010 Comments at 4; CAISO Oct. 13, 2010 Comments at 5 and 16. 95 EDF Oct. 13, 2010 Comments at 2; Viridity Oct. 13, 2010 Comments at 10; ELCON Oct. 13, 2010 Comments at 3. 41. Another set of commenters argues that a net benefits test is unnecessary and inappropriate for different reasons. 96 These commenters assert that a net benefits test would be very costly and difficult to implement, that RTOs and ISOs cannot implement a net benefits test, 97 and that such a test is unnecessary with the economically efficient compensation level for demand response resources. 98 According to Andy Ott of PJM, “[t]he implicit assumption in developing a benefits test for purposes of compensation would be that you could actually determine individual customers, whether they benefitted or not. That type of analysis would be very costly to implement.” 99 Midwest ISO TOs further assert that it would be difficult to prescribe by regulation the hours in which demand response provides net benefits because system conditions and load patterns change across seasons and over time. 100 NEPGA argues that compensating demand response resources at LMP whenever a reduction in consumption suppresses energy prices enough to provide net benefits to load is neither just and reasonable, nor in the public interest. 101 NEPGA states that the Commission recognized in Amaranth Advisors 102 that, if prices are suppressed below competitive, market levels, society as a whole is worse off. According to NEPGA, the goal is to get the right price—the economically efficient price produced by competitive markets. 96 See, e.g., Oct. 13, 2010 Comments of: Midwest TDUs at 4-5; NEPGA at 8, NJBPU at 2-3; NAPP at 2-3; P3; SPP at 3-4; SDG&E, SoCal Edison, and PG&E at 4-6; Viridity Energy at 2; ELCON at 2; AMP at 2; CDWR at 1, 4-5; CAISO at 4, 15; Detroit Edison at 2; Smart Grid Coalition at 2; Duke Energy at 2; EDF at 2; FTC at 1; EPSA at 4; Indicated New York TOs at 3; Midwest ISO at 9; Steel Manufacturers Ass'n at 3. 97 P3 Oct. 13, 2010 Comments at 5. 98 Sept. 13, 2010 Tr. 155:21-24 (Mr. Robinson); Sept. 13, 2010 Tr. 141-42 (Mr. Centolella); Dr. Hogan Sept. 13, 2010 Comments at 5; Sept. 13, 2010 Tr. 60 (Dr. Shanker); Sept. 13, 2010 Tr. 27 (Mr. Newton); SDG&E May 13, 2010 Comments at 4. 99 Sept. 13, 2010 Tr. 19 (Mr. Ott). 100 Midwest ISO TOs May 13, 2010 Comments at 16. 101 NEPGA June 21, 2010 Comments at 1-2. 102 120 FERC ¶ 61,085 (2007). 42. NYISO posits that a rule mandating payment of LMP-G avoids the need to develop a net benefits test. NYISO further states, however, that if the Commission decides to move forward with LMP for demand response, it should craft a net benefits test that minimizes any opportunities for distorting market prices or exploiting market inefficiencies. Citing support for Dr. Hogan's arguments, NYISO states that “a net benefits test should ensure that the demand response program does not have negative net benefits compared to no program at all. The criterion to apply would focus on the bid-cost savings of generation and load, with the load bids adjusted for the effects of avoidance of the retail rate.” 103 103 NYISO Oct. 13, 2010 Comments at 3-4. (c) Standardization or Regional Variations in Compensation 43. With regard to potential regional variations for compensation mechanisms across RTO and ISO markets, many commenters, mostly those in support of the NOPR's proposed compensation level, endorse standardization. 104 Some parties, primarily industrial customers and some customer advocates, argue that, regardless of location, both demand response providers and generators provide a comparable service in terms of balancing supply and demand, as discussed above, and therefore should be comparably compensated at the LMP. 105 They argue that fair, non-discriminatory markets must adapt and eliminate barriers to entry to the use and incorporation of traditional and non-traditional resources—where non-traditional resources include actively-managed demand—in the dispatch and management of the electric system. 106 They further posit that the lack of a unified policy itself represents a regulatory barrier to demand response, 107 and that a consistent set of rules reduces the costs and complexities of demand response participation and facilitates training and transfer of personnel across regions. 108 To that end, many commenters argue that adopting a unified approach to demand response compensation at the LMP, as opposed to allowing regional variation including payment of something less than LMP, is necessary to overcome the barriers to entry of demand response providers. 109 Reciting the many benefits of demand reductions in energy use, these commenters support a compensation level that will provide a catalyst for private sector engagement in improved energy management practices. Viridity argues that the near absence of demand response participating in energy markets is powerful empirical proof that current, varying levels of compensation are inadequate—especially in markets that start with a market-based level of compensation and then reduce it by the generation portion of a customer's retail rate (LMP-G). 110 104 See May 13, 2010 Comments of: ArcelorMittal; Alcoa; ACENY; ACC; AFPA; CDWR; Mayor Bloomberg; Consert; CDRI; CPower; DR Supporters; Derstine's; Durgin; Electricity Committee; ELCON; Electrodynamics; ECS; EnerNOC; ICUB; IECA; IECPA; Irving Forest; Joint Consumers; Limington; Madison Paper; Massachusetts AG; NEMA; National Energy; National League of Cities; NJBPU; NAPP; Occidental; Okemo; Partners; Pennsylvania Department of Environment; Pennsylvania Commission; Rep. Chris Ross; Precision; PRLC; Raritan; SDEG, SoCal; PG&E; Schneider; Governor O'Malley; Steel Manufacturers Ass'n; Verso; Viridity; Virginia Committee; Wal-Mart; Waterville. 105 See, e.g., Steel Manufacturers Ass'n May 13, 2010 Comments at 12; NEMA May 13, 2010 Comments at 5. 106 Steel Manufacturers Ass'n May 13, 2010 Comments at 12. 107 PIO May 13, 2010 Comments at 9; DR Supporters Aug. 30, 2010 Comments at 6-7. 108 See, e.g., Alcoa May 13, 2010 Comments at 13. 109 NECPUC May 13, 2010 Comments at 4; NYISO May 13, 2010 Comments at 16. 110 Viridity Energy May 13, 2010 Comments at 4. 44. Other commenters caution against standardizing the compensation level for demand response, pointing to regional differences in market structure, State regulatory environment, and resource mix. 111 111 See, e.g., May 13, 2010 Comments of: ConEd at 3-4; Consumers Energy at 2; California Commission at 9; CMEEC at 2-3, 14-15; Detroit Edison at 3-5; Dominion at 8; Duke Energy at 4; EPSA at 6; Hess at 4; Indicated New York TOs at 3; Maryland Commission at 5; Midwest TDUs at 2, 6; Midwest ISO TOs at 16; National Grid at 5-6; 11-12; New York Commission at 4, 11; NCPA at 3; NYISO at 2-3; ODEC at 27; PJM at 5-6; SPP at 1. 3. Commission Determination 45. The Commission acknowledges the diverging opinions of commenters regarding the appropriate level of compensation for demand response resources. As discussed above, commenters are split on this issue, with some in favor of paying the LMP for demand reductions in the day-ahead and real-time energy markets in all hours, others arguing that paying the LMP for demand reductions under any conditions will result in over-compensation or distortions in incentives to reduce consumption, and still others arguing that paying the LMP for demand reductions is only appropriate when it is reasonably certain to be cost-effective. 46. In the face of these diverging opinions, the Commission observes that, as the courts have recognized, “ `issues of rate design are fairly technical and, insofar as they are not technical, involve policy judgments that lie at the core of the regulatory mission.' ” 112 We also observe that, in making such judgments, the Commission is not limited to textbook economic analysis of the markets subject to our jurisdiction, but also may account for the practical realities of how those markets operate. 113 112 Elec. Consumers Res. Council v. FERC, 407 F.3d 1232, 1236 (DC Cir. 2005) (quoting Pub. Util. Comm'n of the State of Cal. v. FERC, 254 F.3d 250, 254 (DC Cir. 2001)); see also Town of Norwood v. FERC, 962 F.2d 20, 22 (DC Cir. 1992). 113 See Elizabethtown Gas Co. v. FERC, 10 F.3d 866, 872 (DC Cir. 1993) (“It is the FERC's established policy to consider equitable factors in designing rates, and to allow for phasing in of changes where appropriate. * * * It is hardly arbitrary or capricious so to temper the dictates of theory by reference to their consequences in practice.”); Vermont Dep't of Pub. Serv. v. FERC, 817 F.2d 127, 135 (DC Cir. 1987) (“Indeed, `the congressional grant of authority to the agency indicates that the agency's interpretation typically will be enhanced by technical knowledge.' ” (quoting Nat'l Fuel Gas Supply Corp. v. FERC, 811 F.2d 1563, 1570 (DC Cir. 1987))); Columbia Gas Transmission Corp. v. FERC, 750 F.2d 105, 112 (DC Cir. 1984) (“the Commission is vested with wide discretion to balance competing equities against the backdrop of the public interest”). 47. As discussed further below, the Commission agrees with commenters who support payment of LMP under conditions when it is cost-effective to do so, as determined by the net benefits test described herein. 114 We have previously accepted a variety of ISO and RTO proposals for compensation for demand response resources participating in organized wholesale energy markets. We find, based on the record here that, when a demand response resource has the capability to balance supply and demand as an alternative to a generation resource, and when dispatching and paying LMP to that demand response resource is shown to be cost-effective as determined by the net benefits test described herein, payment by an RTO or ISO of compensation other than the LMP is unjust and unreasonable. When these conditions are met, we find that payment of LMP to these resources will result in just and reasonable rates for ratepayers. 115 As stated in the NOPR, we believe paying demand response resources the LMP will compensate those resources in a manner that reflects the marginal value of the resource to each RTO and ISO. 116 114 See generally May 13, 2010 Comments of NYSCPB; NECA; Capital Power; NECPUC; Maryland Commission; New York Commission; NSTAR; National Grid; NE Public Systems. 115 The Commission's findings in this Final Rule do not preclude the Commission from determining that other approaches to compensation would be acceptable when these conditions are not met. 116 NOPR at P 12. 48. The Commission emphasizes that these findings reflect a recognition that it is appropriate to require compensation at the LMP for the service provided by demand response resources participating in the organized wholesale energy markets only when two conditions are met: • The first condition is that the demand response resource has the capability to provide the service, i.e., the demand response resource must be able to displace a generation resource in a manner that serves the RTO or ISO in balancing supply and demand. • The second condition is that the payment of LMP for the provision of the service by the demand response resource must be cost-effective, as determined by the net benefits test described herein. 49. With respect to the first, capability-related condition, we note that a power system must be operated so that there is real-time balance of generation and load, supply and demand. An RTO or ISO dispatches just the amount of generation needed to match expected load at any given moment in time. The system can also be balanced through the reduction of demand. 117 Both can have the same effect of balancing supply and demand at the margin either by increasing supply or by decreasing demand. 117 Andrew L. Ott Sept. 13, 2010 Statement at 1. Economic and Capacity-based demand response clearly provides benefits to regional grid operation and the wholesale market operation. * * * These demand resources provide benefits by providing valuable alternatives to PJM in maintaining operational reliability and in promoting efficient market operations. Id. at 1; see also CDRI May 13, 2010 Comments at 10; CDWR May 13, 2010 Comments at 5; NJPBU May 13, 2010 Comments at 2. 50. With respect to the second cost-effectiveness condition, the record leads us to alter the proposal set forth in the NOPR in this proceeding. As various commenters explain, dispatching demand response resources may result in an increased cost per unit to load associated with the decreased amount of load paying the bill, depending on the change in LMP relative to the size of the energy market. As stated above, this is the billing unit effect of dispatching demand response resources. 118 However, when reductions in LMP from implementing demand response results in a reduction in the total amount consumers pay for resources that is greater than the money spent acquiring those demand response resources at LMP, such a payment is a cost-effective purchase from the customers' standpoint. 119 In comparison, when wholesale energy market customers pay a reduced price attributable to demand response that does not reduce total costs to customers more than the costs of paying LMP to the demand response dispatched, customers suffer a net loss. Implementation of the net benefits test described herein will allow each RTO or ISO to distinguish between these situations. 118 As stated above, dispatching generation resources does not produce this billing unit effect because it does not result in a decrease of load. 119 As a simple example, assume a market of 100 MW, with a current LMP of $50/MWh without demand response, and an LMP of $40/MWh if 5 MW of demand response were dispatched. Total payments to generators and load would be $4,000 with demand response compared to the previous $5,000. Even though, the reduced LMP is now being paid by less load, only 95 MW compared to 100 MW, the price paid by each remaining customer would decrease from $50/MWh to $42.11/MWh ($4,000/95). Therefore, the payment of LMP to demand resources is cost-effective. 51. This billing unit effect and the net benefits test through which it is addressed herein, warrant more detailed discussion. In the organized wholesale energy markets, the economic dispatch organizes offers from lowest to highest bid in order to balance supply and demand, taking into account other parameters such as requirements for a generator to operate at a minimum level of output or minimum amount of time, reserve requirements and so forth. With dispatch of a demand response resource, the load also goes down, that is, the level of remaining load falls. However, the “supply” of resources deployed—which includes both generation and demand response—does not fall. The total costs to the system for these resources must then be allocated among the reduced quantity of remaining load. 52. In the absence of the net benefits test described herein, the RTO's or ISO's economic dispatch ordinarily would select demand response when it is the incremental resource with the lowest bid. However, if the next unit of generation is not sufficiently more expensive than the demand response resource, the decrease in LMP multiplied by the remaining load would not be greater than the costs of dispatching the demand response resource. In this situation, dispatching the demand response resource would result in a higher price to remaining customers than the dispatch of the next unit of generation in the bid stack. While the demand response resource appears cost competitive in the dispatch order, selection of the demand response resource increases the total cost per unit to remaining load, and it would not be cost-effective to dispatch the demand response resource. 53. For this reason, the billing unit effect associated with dispatch of a demand response resource in an energy market must be taken into account in the economic comparison of the energy bids of generation resources and demand response resources. Therefore, rather than requiring compensation at LMP in all hours, the Commission requires the use of the net benefits test described herein to ensure that the overall benefit of the reduced LMP that results from dispatching demand response resources exceeds the cost of dispatching those resources. When the above-noted conditions of capability and of cost-effectiveness are met, it follows that demand response resources that clear in the day-ahead and real-time energy markets should receive the LMP for services provided, as do generation resources. LMP represents the marginal value of an increase in supply or a reduction in consumption at each node within an ISO or RTO, i.e., LMP reflects the marginal value of the last unit of resources necessary to balance supply and demand. Indeed, LMP has been the primary mechanism for compensating generation resources clearing in the organized wholesale energy markets since their formation. 120 120 See DR Supporters Aug. 30, 2010 Reply Comments (Kahn Affidavit at 2 (footnote omitted)). 54. The Commission finds that demand response resources that clear in the day-ahead and real-time energy markets should receive the same market-clearing LMP as compensation in the organized wholesale energy markets when those resources meet the conditions established here as a cost-effective alternative to the next highest-bid generation resources for purposes of balancing the energy market. We discuss below the comments filed on these issues. 55. Some commenters dispute that the foregone consumption of energy by demand response resources performs the service of balancing supply and demand in the energy market as would energy supplied by generators in the day-ahead and real-time energy markets, arguing that it is inappropriate to pay electric consumers to not consume. 121 The Commission disagrees. Generation and load must be balanced by the RTOs and ISOs when clearing the day-ahead and real-time energy markets, and such balancing can be accomplished by changes in either supply or demand. The Commission finds that in the organized wholesale energy markets demand response can balance supply and demand as can generation. 121 See, e.g., ISO-NE May 13, 2010 Comments at 3; APPA May 13, 2010 Comments at 12; Capital Power May 13, 2010 Comments at 2; EPSA May 13, 2010 Comments at 72. 56. Commenters that oppose this finding do not adequately recognize a distinctive and perhaps unique characteristic of the electric industry. The electric industry requires instantaneous balancing of supply and demand at all times to maintain reliability. It is in this context that the Commission finds that demand response can balance supply and demand as can generation when dispatched, in the organized wholesale energy markets. 57. Due to a variety of factors, demand responsiveness to price changes is relatively inelastic in the electric industry and does not play as significant a role in setting the wholesale energy market price as in other industries. The Commission has recognized that barriers remain to demand response participation in organized wholesale energy markets. For example, in Order No. 719, the Commission stated: [D]espite previous Commission and RTO and ISO efforts to facilitate demand response, regulatory and technological barriers to demand response participation persist, thereby limiting the benefits that would otherwise result. A market functions effectively only when both supply and demand can meaningfully participate, and barriers to demand response limit the meaningful participation of demand in electricity markets. 122 122 Order No. 719, FERC Stats. & Regs. ¶ 31,281 at P 83 (citing Federal Energy Regulatory Commission Staff, A National Assessment of Demand Response Potential (June 2009), found at http://www.ferc.gov/legal/staff-refports/06-09-demand-response.pdf; Barriers to Demand Side Response in PJM (2009)). In compliance filings submitted by RTOs and ISOs and their market monitors pursuant to Order No. 719, as well as in responsive pleadings, parties have mentioned additional barriers, such as the inability of demand response resources to set LMP, minimum size requirements, and others. Barriers to demand response participation at the wholesale level identified by commenters include the lack of a direct connection between wholesale and retail prices, 123 lack of dynamic retail prices (retail prices that vary with changes in marginal wholesale costs), the lack of real-time information sharing, and the lack of market incentives to invest in enabling technologies that would allow electric customers and aggregators of retail customers to see and respond to changes in marginal costs of providing electric service as those costs change. For example, Dr. Kahn states: 123 See, e.g., Monitoring Analytics May 13, 2010 Comments at 4-6. These circumstances—specifically, the fact that pass-through of the LMP is costly and (perhaps) politically infeasible, the possibly prohibitive cost of the metering necessary to charge each ultimate user, moment-by-moment, the often dramatic changes in true marginal costs for each—can justify direct payment at full LMP to distributors and ultimate customers who promise to guarantee their immediate response to such increases in true marginal costs of supplying them. 124 124 DR Supporters Sept. 16, 2009 Comments filed in Docket No. EL-09-68-000 (Kahn Affidavit at 6). See also id. at 4 (Customers offering to reduce consumption should be induced “to behave as they would if market mechanisms alone were capable of rewarding them directly for efficient economizing.”). Furthermore, EnerNOC states: On a more fundamental level, the inadequate compensation mechanisms in place today in wholesale energy markets fail to induce sufficient investment in demand response resource infrastructure and expertise that could lead to adequate levels of demand response procurement. Without sufficient investment in the development of demand response, demand response resources simply cannot be procured because they do not yet exist as resources. Such investment will not occur so long as compensation undervalues demand response resources. 125 125 EnerNOC May 13, 2010 Comments at 4; s ee also Alcoa May 13, 2010 Comments at 4; Viridity May 13, 2010 Comments at 5-6. 58. The Commission concludes that paying LMP can address the identified barriers to potential demand response providers. 59. Removing barriers to demand response will lead to increased levels of investment in and thereby participation of demand response resources (and help limit potential generator market power), moving prices closer to the levels that would result if all demand could respond to the marginal cost of energy. To that end, the Commission emphasizes that removing barriers to demand response participation is not the same as giving preferential treatment to demand response providers; rather, it facilitates greater competition, with the markets themselves determining the appropriate mix of resources, which may include both generation and demand response, needed by the RTO and ISO to balance supply and demand based on relative bids in the energy markets. In other words, while the level of compensation provided to each resource affects its willingness and ability to participate in the energy market, ultimately the markets themselves will determine the level of generation and demand response resources needed for purposes of balancing the electricity grid. 126 126 Generation and demand response resources have the potential to earn other revenues through bilateral arrangements, capacity markets where they exist, and ancillary services. 60. Another issue raised by a number of commenters, largely representing generators, is whether a lower payment based on LMP-G is the economically-efficient price that sends the proper price signal to a potential demand response provider. These commenters argue that, by not consuming energy, demand response providers already effectively receive “G,” the retail rate that they do not need to pay. They therefore contend that demand response providers will be overcompensated unless “G” is deducted from payments made by the RTO or ISO for service in the wholesale energy market, resulting in a payment of LMP-G. These commenters suggest that payment of LMP-G will result in a price signal to demand response providers equivalent to the LMP ( i.e., (LMP−G) + G). Similarly, some commenters argue that paying demand response resources the LMP will lead to a wholesale electricity price that is not economically efficient. 127 127 See NEPGA June 21, 2010 Comments at 1-2. 61. The Commission disagrees with commenters who contend that demand response resources should be paid LMP-G in all hours. First, as discussed above, demand response resources participating in the organized wholesale energy markets can be cost-effective, as determined by the net benefits test described herein, for balancing supply and demand and, in those circumstances, it follows that the demand response resource should also receive compensation at LMP. Second, such comments largely rely on arguments about economic efficiency, analogizing to incentives for individual generators to bid their marginal cost. These arguments fail to acknowledge the market imperfections caused by the existing barriers to demand response, also discussed above. In Order No. 719, the Commission found that allowing demand response to bid into organized wholesale energy markets “expands the amount of resources available to the market, increases competition, helps reduce prices to consumers and enhances reliability.” 128 Furthermore, Dr. Kahn argues that paying demand response LMP sets “up an arrangement that treats proffered reductions in demand on a competitive par with positive supplies; but the one is no more a [case of overcompensation] than the other: the one delivers electric power to users at marginal costs—the other— reductions in cost —both at competitively-determined levels.” 129 128 Order No. 719, FERC Stats. & Regs. ¶ 31,281 at P 154. 129 DR Supporters Aug. 30, 2010 Reply Comments (Kahn Affidavit at 9-10). 62. Several other considerations also support this Commission conclusion. In the absence of market power concerns, the Commission does not inquire into the costs or benefits of production for the individual resources participating as supply resources in the organized wholesale electricity markets and will not here, as requested by some commenters, single out demand response resources for adjustments to compensation. The Commission has long held that payment of LMP to supply resources clearing in the day-ahead and real-time energy markets encourages “more efficient supply and demand decisions in both the short run and long run,” 130 notwithstanding the particular costs of production of individual resources. Commenters have not justified why it would be appropriate for the Commission to continue to apply this approach to generation resources yet depart from this approach for demand response resources. 130 See New England Power Pool, 101 FERC ¶ 61,344, at P 35 (2002). 63. In addition, we agree with the New York Commission that given the differences in retail rate structures across RTO footprints and even within individual States, requiring ISOs and RTOs to incorporate such disparate retail rates into wholesale payments to wholesale demand response providers would, even though perhaps feasible, create practical difficulties for a number of parties, including State commissions and ISOs and RTOs. Moreover, incorporating such rates could result in customer uncertainty as to the prevailing wholesale rate. 64. Some arguments advocating paying LMP-G rather than LMP are based on an assumption that demand response resources need to purchase the energy in day-ahead markets or by other means and then “resell” the energy to the market in the form of demand response. However, as the Commission previously stated in EnergyConnect, the Commission does not view demand response as a resale of energy back into the energy market. 131 Instead, as the Commission also explained in EnergyConnect and in Order No. 719-A, the Commission asserts jurisdiction with respect to demand response in organized wholesale energy markets because of the effect of demand response and related RTO and ISO market rules on Commission-jurisdictional rates. 132 131 See EnergyConnect, 130 FERC ¶ 61,031 at P 32. 132 Id.; see also Order No. 719-A, FERC Stats. & Regs. ¶ 31,292, at P 47. 65. With regard to the “buyers' cartel” argument, the Commission disagrees that market rules establishing circumstances in which particular resources can participate and receive the LMP represents cooperative price setting. RTOs and ISOs evaluate the bids from generation and demand response resources to establish the order of dispatch which secures the most economical supplies needed, consistent with the reliability constraints imposed on the system. Imposing a cost-effectiveness condition does not convert this unit commitment process by the RTO or ISO into collusion among bidders, whether generation or demand response. Furthermore, the market rules administering such a program would be approved by this Commission and demand response resources would be subject to Commission-approved rules, just like any other participants in the organized wholesale energy markets. In addition, arguments that the subject of this proceeding is equivalent to the types of market manipulation investigated in Amaranth and ETP are groundless and without merit. In Amaranth, the trader was accused of engaging in a fraudulent scheme with scienter in connection with a jurisdictional transaction. Here, there is no such allegation, merely speculation that the Commission is somehow facilitating coordination of demand-side bidders in order to lower prices. 66. Some commenters argue that demand response providers and generators should both be compensated at the market clearing price only if both are subject to the same market participation rules, penalty structures, testing requirements, and market monitoring provisions. The ISOs and RTOs already consider how to ensure comparability between demand response and generation in terms of market rules. 133 The Commission agrees that as a general matter demand response providers and generators should be subject to comparable rules that reflect the characteristics of the resource, and expect ISOs and RTOs to continue their evaluation of their existing rules in light of this Final Rule and make appropriate filings with the Commission. 133 See PJM Interconnection, L.L.C., 129 FERC ¶ 61,081 (2009). 67. Some commenters argue that the Commission should not impose a single pricing rule due to differences in market structure, State regulatory environment, and resource mix among the ISOs and RTOs. While such differences may exist, the commenters have not shown why such differences warrant a different compensation level among the ISOs and RTOs. As discussed above, regardless of the resource mix or the State regulatory environment, demand response, which satisfies the net benefits test described herein and can balance the system, is a cost-effective alternative to generation in the organized wholesale energy markets, and payment of LMP represents the marginal value of a decrease in demand. B. Implementation of a Net Benefits Test 1. Comments 68. In response to questions that the Commission posed in the Supplemental NOPR, some commenters advocate a net benefits trigger based on a particular price threshold. 134 The NYISO currently has a static bid threshold of $75/MWh in its day-ahead demand response program. 135 134 For example, National Grid states that the threshold could be triggered by a particular price on the supply offer curve at which the additional cost of paying LMP to demand response resources is most likely to be outweighed by LMP reductions in the wholesale energy market as a result of the demand reductions produced by these resources. National Grid May 13, 2010 Comments at 6. Those in favor of a price threshold include National Grid (but allow the ISO or RTO to identify threshold based on analysis); NE Public Systems; NECPUC; ISO-NE (minimum offer price based on fixed heat rate, times a fuel price index); New York Commission (supports ISO-NE's heat rate indexed price threshold). 135 NYISO implements a day-ahead demand response program by which resources bid into the market at a minimum of $75/MWh and can get paid the LMP. See section 4.2.2.9 (“Day-Ahead Bids from Demand Reduction Providers to Supply Energy from Demand Reductions”) of NYISO's Market Services Tariff. 69. However, other commenters assert that using a static threshold based on historical data misses the changes that occur within electricity markets across seasons and years, and that it is erroneous to assume that all demand response occurring above a certain threshold price (for instance, at the very highest loads or highest priced hours) will result in lower costs to wholesale customers and that demand response is not cost-effective at prices below the static threshold price. 136 They argue that a static threshold offer price cannot easily adjust with changing energy market prices which may result in inefficient dispatch of demand resources, excluding demand response participation in hours when demand response can provide beneficial savings and including demand response participation in hours when there are no beneficial savings. 137 The New York Commission supports a dynamic, rather than a static bid threshold, arguing that, while a static bid threshold helps prevent demand response providers from gaming the system by seeking compensation for reducing electricity consumption for reasons other than market prices, it can also limit participation in a demand response program because prices might not exceed the threshold on a consistent basis. 138 136 Sept. 13, 2010 Tr. 52-53 (Mr. Peterson); Massachusetts AG Oct. 13, 2010 Comments at 23. 137 Massachusetts AG Oct. 13, 2010 Comments (attachment, Demand Response Potential in ISO New England's Day-Ahead Energy Market, Synapse Energy Economics, Inc. Oct. 11, 2010 at 9). See generally, NECPUC May 13, 2010 Comments at 18. 138 Id. 70. In a similar vein, some commenters suggest utilizing a dynamic bid threshold for determining when LMP payment would apply. 139 For example, NECPUC favors use of a dynamic mechanism such as a price threshold based on a preset heat rate of marginal generation and fuel price, like that currently used in New England's Day-Ahead Load Response Program (DALRP), 140 for the ISO-NE control area. 141 National Grid suggests a trigger, determined by each ISO or RTO, using a particular price on the supply offer curve at which the additional cost of paying LMP to demand resources is most likely to be outweighed by LMP reductions in the wholesale energy market as a result of the demand reductions. 142 139 National Grid May 13, 2010 Comments at 6; New York Commission May 13, 2010 Comments at 10; Viridity May 13, 2010 Comments at 24. See generally NECPUC, New York Commission; ISO-NE; NSTAR; ACEEE; and NYSCPB Oct. 13, 2010 Comments. 140 The DALRP establishes a minimum offer price by approximating the variable cost component, in the form of a fuel cost, of a hypothetical peaking unit sufficiently high enough in the supply stack to ensure net benefits. On a monthly basis, this minimum offer price is reset to reflect the product of an appropriate fuel price index and a proxy heat rate. See NECPUC Oct. 13, 2010 Comments at 15. 141 NECPUC Oct. 13, 2010 Comments at 14-16; NECPUC May 13, 2010 Comments at 17. 142 Id. at 5-6. 71. Still other commenters urge compensating demand response during an ISO- or RTO-defined period of critical high-cost hours in which it is cost-effective to pay LMP. These commenters argue that the effect of demand response on the market clearing price is greatest during a limited number of hours during the year. 143 Therefore, identifying the hours in which to pay LMP to demand response resources could be used as a cost-effective net benefits test with potential savings for ratepayers. According to PJM, further analysis is needed to ascertain the critical high-cost hours in which it will be cost-effective to pay full LMP for demand response. 144 143 Maryland Commission May 13, 2010 Comments at 4-5; see generally NSTAR, ACEEE and NYSCPB Oct. 13, 2010 Comments. 144 Maryland Commission May 13, 2010 Comments at 4 n.9. 72. The Consumer Demand Response Initiative (CDRI) proposes a mechanism for determining what demand response resources are cost-effective in any hour. 145 This dispatch algorithm tests whether the money necessary to compensate demand response is less than the cost savings due to the decreased market-clearing price resulting from implementing demand response. In a sense, it is a dynamic cost/benefit analysis built into the dispatch algorithm. This cost/benefit analysis accounts for the billing unit effect. The billing unit effect occurs when demand response resources are dispatched to balance the system; the associated reduction in load results in fewer MWh of realized load (demand) paying for the sum of generator and demand response resource MWh, so load pays an effective rate which is greater than the LMP set to procure resources. Some commenters assert that if the Commission finds that a net benefits test is needed, it should require organized wholesale energy market operators to implement a proposal similar to that submitted by CDRI. 146 145 The approach submitted by CDRI was developed for implementation in the ISO-NE day-ahead energy market. The discussion here is generalized to be applicable to any energy market that uses security-constrained economic dispatch to select the least-cost resources and establish a market-clearing price. 146 PIO July 27, 2010 Comments at 6; Massachusetts AG Oct. 13, 2010 Comments at 11; Viridity Oct. 13, 2010 Comments at 2. See CDRI May 13, 2010 Comments for a full description of the algorithms. 73. Under the proposal submitted by CDRI, the demand response bids are part of the supply stack to which a security-constrained economic dispatch process is applied. All demand response bids that result in a lower price to customers, including consideration of the reduced number of billing units, are selected while those bids that raise the price, as compared to selecting the next generation bid in the supply stack, are not. This dispatch algorithm, as proposed, would be used by the ISO or RTO to determine a revised LMP that would be charged to load. The revised LMP creates a surplus (or over-collection) of revenue for the ISO or RTO that is then distributed to the LSEs through a settlement algorithm with the goal of holding LSEs harmless. 147 147 CDRI May 13, 2010 Comments Attachment B at 18. CDRI states that the dispatch and settlement algorithms “could be employed to evaluate dispatch and assure customer benefits, without being employed to perform allocations and settlements.” CDRI Oct. 13, 2010 Comments at 4. 74. During the September 2010 Technical Conference, Dr. Ethier of ISO-NE stated that a dynamic net benefits test done on an hourly basis that examines the effect of the demand response resource on LMPs, similar to that proposed by CDRI, would become very complicated to implement and require essentially an iterative process. 148 Dr. Ethier states that the ISO would have to run the dispatch model to formulate a base LMP with no demand response and then re-run it with demand response in the market; however those two iterations alone do not “cover the whole waterfront” in terms of the possible iterations required. According to Dr. Ethier, the ISO could dispatch too much demand response the first time, or if the ISO first rejected dispatching demand response, it may need to go back and dispatch smaller amounts of demand response to determine what would happen to the LMPs. Dr. Ethier stated that it is unclear where the ISO would stop the iteration of testing the impact on LMPs of dispatching demand response. 149 Andy Ott of PJM also stated during the technical conference that implementing a net benefits test would entail an iterative process that would be costly and difficult, if the RTO could even do it. 150 148 Sept. 13, 2010 Tr. 80-81 (Dr. Ethier). 149 Id. 150 Sept. 13, 2010 Tr. 82:16-21 (Mr. Ott). 75. Other commenters do not support the use of a net benefits test, but state that if one is adopted it should be based on general principles that RTOs and ISOs must apply to their systems in determining when LMP payments will apply. 151 A few commenters articulated specific criteria to be used in a net benefits test. 152 AEP believes that the objective of an incentive payment for demand response resources on the basis of broad market benefits can be achieved through a review of the costs and benefits of individual providers. Constellation states that any net benefits test should be based on the difference between the value consumers receive from energy and the cost of energy production. 153 151 See generally AEP, Midwest ISO, Occidental, NYISO, Constellation Oct. 13, 2010 Comments. 152 See, e.g., Midwest ISO October 13, 2010 Comments at 9-14 and Table 1 (setting forth comprehensive list of benefits and costs of demand response by type of market participants); Occidental October 13, 2010 Comments at 4-5 (any net benefits test must take into consideration offsetting variables, such as higher LMPs in the subsequent periods where demand rebound increases market price, and capacity market price effects); AEP October 13, 2010 Comments at 3-4 (AEP does not recommend the use of a societal benefits component ( i.e., health, environment, or employment efforts)). 153 Constellation October 13, 2010 Comments at 3-4. 76. ISO-NE argues that a net benefits test should be based on economic efficiency, the sum of producer and consumer surplus, which suggests that demand response incentives ought to be provided to encourage demand reductions when the cost of energy production exceeds the value of consumption, and to encourage usage when the cost of energy production is less than the value of consumption. ISO-NE further states that a net benefits test that focuses solely on consumer savings ignores the value lost by consumers when energy consumption levels are reduced in response to incentive payments. ISO-NE posits that any variant of a LMP payment should be limited to a very small number of high-priced hours to minimize the economic distortions and avoid significant administrative complexities. 154 154 ISO-NE Oct. 13, 2010 Comments at 4-5 and 21. 77. A few commenters state that policies affecting energy prices will also impact capacity prices because generation owners with fixed costs must raise capacity price offers to remain financially viable at lower energy prices. 155 ISO-NE and Pepco argue, therefore, that the Commission should adopt a net benefits test that considers the impact of demand response compensation on both energy and capacity markets. 156 According to ISO-NE, when considering capacity market impacts under full-LMP compensation, long-term increases in capacity prices in response to suppressed LMPs offset consumer savings and leaves consumers worse off over time. 157 Robert Weishaar of the DR Supporters argues that properly compensating demand response should flatten the load profile and decrease the forecast of load projections, which would reduce capacity clearing prices. 158 Donald Sipe of CDRI adds that to the extent that scarcity revenues are not sufficient, capacity markets are designed to ensure that a generator's capital costs are recovered; in a forward market that looks ahead as load adjusts, one can see whether a resource is performing or not. For purposes of long-run reliability, he argues, as long as compensation is in the amount that is necessary to induce new investment and reflects market value, the argument that demand response in the bid stack will push out generators is only true if generators are higher priced than the consumer resources that are brought by demand response. 159 155 See, e.g., Sept. 13, 2010 Tr. 94:13-22 (Dr. Shanker); Sept. 13, 2010 Tr. 98:4-24 (Mr. Peterson); Sept. 13, 2010 Tr. 99:2-7 (Mr. Sunderhauf); ISO-NE Oct. 13, 2010 Comments at 5. 156 Sept. 13, 2010 Tr. 99:1-24 (Mr. Sunderhauf); ISO-NE Oct. 13, 2010 Comments at 5. 157 ISO-NE Oct. 13, 2010 Comments at 6. 158 Sept. 13, 2010 Tr. 103-104 (Mr. Weishaar). 159 Sept. 13, 2010 Tr. 106:16-24 (Mr. Sipe). 2. Commission Determination 78. For the reasons discussed previously, the Commission is requiring each RTO and ISO to implement the net benefits test described herein to determine whether a demand response resource is cost-effective. More specifically, the Commission is adopting two distinct requirements with respect to the net benefits test. While we find that the integration of the billing unit effect into the RTO/ISO dispatch processes has the potential to more precisely identify when demand response resources are cost-effective, we also recognize and understand the position of several of the RTOs and ISOs that modification of their dispatch algorithms may be difficult in the near term. Given these technical difficulties, we will require to RTOs and ISO to perform (1) the net benefits test described below to determine on a monthly basis under which conditions it is cost-effective to pay full LMP to demand resources; 160 and (2) a study of the feasibility of developing a mechanism for determining the cost-effective dispatch of demand resources. 160 There will be inherent differences in the supply curves determined by each RTO and ISO under the net benefits test required herein due to decisions the RTOs and ISOs must make based on supply data for their regions, the mathematical methods each RTO and ISO chooses to use for smoothing the supply curves, the certainty of changes in supply due to outages in each region, local generation heat rates, and the choice of relevant fuel price indices. 79. First we direct each RTO and ISO to undertake an analysis on a monthly basis, based on historical data and the RTO's or ISO's previous year's supply curve, to identify a price threshold to estimate where customer net benefits, as defined herein, would occur. The RTO or ISO should determine the threshold price corresponding to the point along the supply stack for each month beyond which the benefit to load from the reduced LMP resulting from dispatching demand response resources exceeds the increased cost to load associated with the billing unit effect, and update the calculation monthly. The ISOs and RTOs are to determine monthly threshold prices based on historical data. The threshold prices would be updated monthly as new data becomes available and posted on the RTO Web site. For example, the RTO should conduct an analysis of supply curves for January through December 2010 to be used as a starting point to establish threshold prices for 2011. Those numbers would be updated monthly during 2011 for significant changes in resource availability and fuel prices, with the process repeated monthly to reflect that month's data from the previous year. 161 The supply curve analysis should be updated monthly, by the 15th day of the preceeding month in advance of the effective date, to allow demand response providers as well as other market participants to plan, while still reflecting current supply conditions. 162 161 The ISOs and RTOs are to select a representative supply curve for the study month, smooth the supply curve using numerical methods, and find the price/quantity pair above which a one megawatt reduction in quantity that is paid LMP would result in a larger percentage decrease in price than the corresponding percentage decrease in quantity (billing units). Beyond that point, a reduction in quantity everywhere along an upward sloping supply curve would be cost-effective. 162 Thus, the test is to determine where: (Delta LMP × MWh consumed) > (LMP new × DR); where LMP new is the market clearing price after demand response (DR) is dispatched and Delta LMP is the price before DR is dispatched minus the market clearing price after DR is dispatched. 80. Based on historical evidence and analysis submitted in this proceeding, the threshold point along the supply stack for each month will fall in the area where the supply curve becomes inelastic, rather than the extreme steep portion at the peak or in the flat portion of the supply curve. 163 In other words, LMP will be paid to demand response resources during periods when the nature of the supply curve is such that small decreases in generation being called to serve load will result in price decreases sufficient to offset the billing unit effect. The Massachusetts AG noted that the actual supply stack has locally flat and steep sections at all bid prices. We recognize that the threshold price approach we adopt here may result in instances both when demand response is not paid the LMP but would be cost-effective and when demand response is paid the LMP but is not cost-effective. We accept this result given the apparent computational difficulty of adopting a dynamic approach that incorporates the billing unit effect in the dispatch algorithms at this time. 164 163 Supply elasticity is defined as the percentage change in quantity supplied divided by the percentage change in price. When the elasticity is less than or equal to one, supply is considered inelastic. So, for example, in the inelastic portion of the supply curve, a reduction in quantity supplied by one percent will result in more than a one percent decrease in price. Using the terms related to demand response compensation, the billing unit effect (percentage change in quantity supplied) will be more than offset by lower LMP (percentage change in price), thus resulting in lower prices for wholesale load. 164 See supra note 114. 81. We direct each RTO and ISO to file its analysis as supporting documentation to the accompanying tariff revisions with the Commission on or before July 22, 2011, along with proposed tariff revisions necessary to comply with this Final Rule. The filing should include the data, analytical methods and the actual supply curves used to determine the monthly threshold prices for the last 12 months to show how the RTO or ISO would calculate the curves. 165 The Commission-approved net benefits test methodology must be posted on the RTO or ISO's Web site, with supporting documentation. The RTO or ISO must also post the price threshold levels that would have been in effect in the previous 12 months. In addition, when the net benefits test becomes effective, the supply curve analysis for the historic month that corresponds to the effective month should be updated for current fuel prices, unit availabilities, and any other significant changes to historic supply curve and posted on the RTO Web site (for example, the supply curve analysis for the March price threshold would be posted in mid-February). Finally, the supply curve analyses for all months should be updated and posted on the RTO Web site if a significant change to the composition or slope of the historic monthly curves occurs, such as extended outages or retirements not previously reflected. 165 See supra P 6. 82. Some commenters argue that that there would be no need for a net benefits test if demand response resources were paid LMP-G, while others argue that use of a net benefits test otherwise undermines our decision to compensate demand response resources at the LMP. As stated above, the Commission finds that when a demand response resource participating in an organized wholesale energy market is capable of balancing supply and demand in the energy market and is cost-effective, as determined by the net benefits test described herein, that demand response resource should receive the same compensation, the LMP, as a generation resource when dispatched. We see no reason to reduce that compensation simply to avoid the use of the net benefits test that will ensure benefits to load. 83. Nearly every participant in the net benefits panel at the September 13, 2010 Technical Conference agreed that it would be counterproductive to defer to the RTO or ISO stakeholder process to determine when demand response provides net benefits without explicit guidance from the Commission. 166 We believe that this result, and the guidance provided in this Final Rule will provide for timely improvements to RTO and ISO market pricing for demand response resources participating in organized wholesale energy markets. 166 “[G]etting this decision resolved is an impediment to all the other stuff we want to do with price response to demand, and DR generally in our market * * * so until we get through this, we're not going to make much progress * * * the implication of that is if you send something back that leaves a lot of room for debate, it's going to be a while on all those other things.” Testimony of Robert Ethier, Vice President, Market Design, ISO-NE, Sept. 13, 2010 Tr. at 136. 84. In addition to requiring each RTO and ISO to construct the net benefits test described herein, the Commission also imposes a second requirement for each RTO and ISO to undertake a study, examining the requirements for and impacts of implementing a dynamic approach to determine when paying demand response resources LMP results in net benefits to customers. We believe that integration of the billing unit effect into RTO and ISO dispatch algorithms holds promise for more accurately integrating demand resources on a dynamic basis into the dispatch of the RTOs and ISOs. In theory, this could help ensure that the cost-effective level of demand response resources is dispatched or scheduled into the organized wholesale energy markets. Given the potential of software enhancements to determine the amount of cost-effective demand response resources purchased in the day-ahead and real-time energy markets, we believe that it would be useful for the Commission to know more about the feasibility of and requirements for implementing improvements to the existing dispatch algorithms. Therefore, we will require each RTO and ISO to undertake a study, either individually or collectively, examining the requirements for, costs of, and impacts of implementing a dynamic net benefits approach to the dispatch of demand resources that takes into account the billing unit effect in the economic dispatch in both the day-ahead and real-time energy markets, and to file the results of their study with the Commission on or before September 21, 2012. 85. ISO-NE and Pepco suggest that the net benefits test also consider the impact of demand response compensation on both energy and capacity markets. However, this Final Rule is focused only on organized wholesale energy markets, not capacity markets. 167 Given the differences in capacity markets among the ISOs and RTOs, the record in this proceeding provides neither a reasonable basis for including capacity market effects in net benefits calculations in the energy markets, nor have ISO-NE and Pepco provided a methodology for taking such effects into account. Indeed, in some cases, the capacity markets already reflect energy and ancillary service revenue in determining capacity prices. 167 Additionally, the arguments presented for focusing on the effect of demand response compensation in wholesale energy markets on capacity markets were not convincing—that decreases in energy market revenues by generators will be recouped in the form of increased capacity prices. First, they fail to consider how the increased participation by demand resources could actually increase potential suppliers in the capacity markets by reducing barriers to demand resources, which would tend to drive capacity prices down. Second, they did not examine the way in which capacity markets already may take into account energy revenues. C. Measurement and Verification 1. NOPR Proposal 86. In the NOPR, the Commission explained that demand response curtailment is a reduction in actual load as compared to the demand response provider's expected level of electricity consumption. 168 The NOPR did not address measurement and verification of demand response. 168 Demand Response Compensation in Organized Wholesale Energy Markets, FERC Stats. & Regs. ¶ 32,656, at P 1 (2010). 87. Each RTO and ISO with a demand response program has procedures for the measurement and verification of demand response. These procedures include techniques to establish a customer baseline for each demand response participant. This customer baseline then becomes the basis for measuring the quantity of demand response delivered to the wholesale market. Customer baselines are often based on historic load information, such as an average of five of the last ten comparable days' hourly load profile. Techniques vary among RTOs and ISOs and most have several techniques that may be allowed, depending on the demand response provider's characteristics. 169 169 See, e.g., ISO/RTO Council, North American Wholesale Electricity Demand Response 2010 Comparison, under the tab for “Performance Evaluation Methods” ( http://www.isorto.org/atf/cf/%7B5b4e85c6-7eac-40a0-8dc3-003829518ebd%7D/IRC%20DR%20M&V%20STANDARDS%20IMPLEMENTATION%20COMPARISON%20(20100524).XLS ). 2. Comments 88. Commenters assert that the integrity of a demand response program is heavily dependent on measurement and verification. 170 Some commenters raise the issue that paying LMP in all hours presents a significant challenge to the accurate measurement and verification of demand response. 171 ISO-NE argues that when a market participant schedules demand reductions for many consecutive days, baselines may become stale—no longer reflecting a customer's “normal” electricity usage. 172 ISO-NE goes on to argue that “it is necessary to limit the number of hours or days that a demand resource could clear in the energy market so that the customer's `normal' load can be estimated” to avoid the potential for manipulation. 173 In the context of the Commission's proposal to pay demand response the LMP in all hours, ISO-NE goes on to advocate requiring demand response to establish baselines by purchasing energy in the day-ahead market as a way to overcome its concerns with statistical baseline methods. 174 ISO-NE IMM makes similar arguments and recommendations. 175 Westar also appears to support this approach. 176 170 Illinois CUB May 14, 2010 Comments at 16-17; Joint Consumers May 13, 2010 Comments at 12; P3 May 12, 2010 Comments at 38; Westar May 13, 2010 Comments at 3. 171 See, e.g., ISO-NE May 13, 2010 Comments at 32. 172 Id. 173 ISO-NE May 13, 2010 Comments at 34. ISO-NE identifies several practices that, in its view, might be deployed by a demand responder to receive payment when it has not, in fact, responded to price. ISO-NE states that observations of such behavior in the Fall of 2007 led it to limit the hours demand response offers could clear the market. Citing ISO New England Inc., Docket No. ER08-538-000 (February 5, 2008 filing). ISO-NE May 13, 2010 Comments at 32-34. 174 Id. 175 ISO-NE IMM May 13, 2010 Comments at 9-13 and Attachment A. 176 Westar May 13, 2010 Comments at 3. 89. Similarly, CPower notes that with some baseline methods, paying LMP in all hours could reward demand responders for any shift in demand from the baseline, not just shifting load from high LMP hours to low LMP hours, or could simply shift load from day-to-day in different hours to affect the calculation of actual curtailment, which it labels “checkerboarding.” However, CPower believes that the capability of consumption management to shed or shift load for many hours is well into the future, and perhaps not a current concern. CPower also believes that baseline standards along with market monitoring will develop to meet these concerns. 177 177 CPower May 13, 2010 Comments at 4-5. 90. ISO-NE IMM asserts that “[if] the Commission adopts any proposal that permits the use of an administrative baseline it should explicitly state that any demand reductions offered into Commission-jurisdictional markets that are not genuine, even if they are the result of `normal' activity * * * may be violations of the Commission's anti- manipulation rules and subject to penalties thereunder.” 178 178 ISO-NE IMM May 13, 2010 Comments at 14 (footnotes omitted) (ISO-NE MMU also notes that “[i]n assessing whether demand reductions are genuine, allowance should be made for non-performance analogous to a generator's forced outage.”). 91. Noting the ongoing efforts by the industry and the North American Energy Standards Board (NAESB) on measurement and verification, EnerNOC takes the view that resolution of customer baseline issues should not delay the issuance of this Final Rule. 179 179 EnerNOC, Inc. May 13, 2010 Comments at 4. 92. Finally, some commenters assert that measurement and verification methods should not be standardized, but left to the RTOs and ISOs to reflect the unique features of their individual energy, ancillary services, and capacity markets. 180 180 ECS May 13, 2010 Comments at 3; Indicated New York TOs May 13, 2010 Comments at 2-3; Midwest ISO May 13, 2010 Comments at 17, 21; National Grid May 13, 2010 Comments at 11-12; NSTAR May 14, 2010 Comments at 9; PPL May 13, 2010 Comments at 4. 3. Commission Determination 93. The Commission agrees with commenters who assert that measurement and verification are critical to the integrity and success of demand response programs. Without a determination of a demand response provider's expected use of power, the ISOs and RTOs cannot determine whether that provider has in fact reduced its energy usage when paid to do so. Towards that end, all the RTOs and ISOs already have measurement and verification protocols for their demand response programs. 181 In addition, we have adopted Phase I standards for measurement and verification published by the North American Energy Standards Board, 182 and have recognized the potential benefits of the continuing NAESB effort to craft Phase II standards with more substantive and consistent wholesale standards for measurement and verification. 183 181 See, e.g., PJM Interconnection, L.L.C., 123 FERC ¶ 61,257 (2008). 182 Standards for Business Practices and Communication Protocols for Public Utilities, Final Rule, 131 FERC ¶ 61,022 (2010). 183 Id., at P 32-34. 94. A number of commenters maintain that compensating demand response resources at the LMP during all hours could make determining baselines for demand response providers exceedingly difficult. However, the impact of our adopting the net benefits test described herein is that the LMP will not be paid to demand response resources in all hours. Accordingly, implementation of this Final Rule would not appear to prevent the determination of appropriate baselines. Nonetheless, we direct ISOs and RTOs to review their current requirements in light of the changes in this Final Rule and develop appropriate revisions and modifications, if necessary, to ensure that their baselines remain accurate and that they can verify that demand response resources have performed. Specifically, we direct each RTO and ISO to include as part of the compliance filing required herein, an explanation of how its measurement and verification protocols will continue to ensure that appropriate baselines are set, and that demand response will continue to be adequately measured and verified as necessary to ensure the performance of each demand response resource. If necessary, each RTO and ISO should propose any changes needed to ensure that measurement and verification of demand response will adequately capture the performance (or non-performance) of each participating demand response market participant to be consistent with the requirements of this Final Rule. 95. Finally, we agree with ISO-NE IMM that demand reductions that are not genuine may be violations of the Commission's anti-manipulation rules. 184 Allegations of such behavior will continue to be investigated, and when appropriate, sanctions will be brought to bear. 184 18 CFR 1.c (2010). D. Cost Allocation 1. NOPR Proposal 96. In response to the NOPR and September 13, 201 [Text truncated at 120,000 characters. The full text is on the page linked above.] --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A2011-6490. Public record. Not legal advice.