Before the
UNITED STATES DEPARTMENT OF ENERGY
ENERGY INFORMATION ADMINISTRATION

Agency Information Collection Activities
Proposed Revision and Extension of EIA Form 767 and Other Electric Power Surveys

Comment Request

Comment of the Staff of the
Bureau of Economics and of Policy Planning
of the Federal Trade Commission(1)

May 14, 2001

I. INTRODUCTION AND SUMMARY

The staff of the Bureau of Economics and of Policy Planning of the Federal Trade Commission (FTC) submit this comment to the United States Department of Energy (DOE) concerning its proposals to expand confidential treatment of data that it collects pursuant to its statutory mandate to manage a centralized, comprehensive, and unified energy information program.(2) EIA proposes to now treat as confidential certain additional operational data that it collects from fossil-fueled steam-electric power plants.(3) The affected information, which is collected on a plant-level basis, includes fuel consumption, quantity, quality, and cost; sales at retail and wholesale; retail sales revenue and number of customers; financial data; thermal output; and cost of purchased power.(4)

The FTC is an independent administrative agency responsible for maintaining competition and safeguarding the interests of consumers. The staff of the FTC often analyzes regulatory or legislative proposals that may affect competition or the efficiency of the economy.(5) In the course of this work, as well as in antitrust research, investigation, and litigation, the staff applies established principles and recent developments in economic theory and empirical analysis to competition issues. The FTC has a longstanding interest in regulation and competition in energy markets, including proposals to reform regulation of the electric power and natural gas industries. The staff has submitted numerous comments concerning these issues at both the federal and state levels,(6) and the FTC has reviewed proposed mergers involving electric power generators and companies that supply fuel for electric generators.

Our basic concern about the EIA proposal to treat additional categories of information as confidential is that it may be premature and may reduce the effectiveness of regulatory reform planning and market monitoring efforts of state and Federal regulatory and law enforcement agencies during the critical, early stages of the transition from regulation to competition. In particular, the ability of state and Federal regulators and competition agencies to understand the complexities of the existing electric transmission system, and proposed changes to the system, may depend upon computer simulations and other analytical techniques that rely on comprehensive, plant-level EIA data.(7) For example, antitrust and regulatory agencies often use these computer simulation models of the electric power grid in designing regulatory reform proposals and evaluating prospective mergers.(8) Although it may be possible for state and Federal agencies to subpoena information to use in modeling, this is not likely to be a workable solution. Under some conditions, all generators within one of the three U.S. transmission interconnects (East, West, and Texas) can affect prices and reliability throughout the interconnect. The task of obtaining, reconciling, and processing this volume of information from several hundred generators in the context of typical regulatory and litigation timetables is nearly impossible.

Potential means to reduce our concern about EIA's proposals include development of a program for selective, confidential access to the data for state and Federal antitrust and regulatory agencies. DOE may wish to consider the costs and benefits of its proposals and this alternative as it updates EIA's electric power survey program.

EIA expresses concern that with increasing competition among generators in U.S. wholesale and retail electric power markets, there is an increased need for confidential treatment of detailed plant-level data.(9) In competitive markets, owners of electric generation facilities may have strong profit incentives to innovate and invest to improve the operation of their facilities. By lowering costs, for example, a generation facility is likely to be dispatched more often (i.e., sell more), and the profit margin between costs and the market clearing price will be increased. EIA is concerned that the incentives to innovate and invest may be blunted if competitors are more likely, due to public disclosure of relevant data, to quickly learn about and emulate the owner's innovations and investments.(10)

Although not explicitly discussed in the Notices, an additional social cost of detailed plant-level disclosures may be an increased likelihood of anticompetitive coordinated interaction among electric power generators.(11) Coordinated interaction generally requires that firms (1) agree (tacitly or explicitly) on pricing or output, (2) monitor the agreement, and (3) punish deviations from the agreement.(12) Detailed knowledge of each other's costs can significantly ease the process of reaching agreement, and detailed, timely knowledge of each other's operating decisions can make monitoring of such an agreement easier as well.

While wholesale trading is active and retail customer choice programs are in place in some states, we note that the transition to competition may be more prolonged than earlier expected.(13) In addition, wholesale competition among generators has not progressed as smoothly as had been predicted.(14) Moreover, investment in new generation capacity may be delayed as investors seek to understand the uncertain regulatory environment in which their investments will compete. Without detracting from the importance of the innovation and investment incentives identified by EIA or the potential coordinated interaction concerns that may arise from rapid public disclosure of detailed plant-level generation data, we encourage DOE to consider alternative approaches to allay these concerns -- approaches that are likely to preserve the ability of federal and state agencies to design and monitor regulatory reform programs.(15) This ability is likely to be particularly significant during the transition to competition in the electric power industry.

II. DOE MAY WISH TO REVIEW THE IMPORTANCE OF THE EIA DATA TO LAW ENFORCEMENT AND REGULATORY AGENCIES, PARTICULARLY DURING THE TRANSITION TO COMPETITION

Federal law enforcement agencies, state regulatory agencies, and electricity customers and suppliers use EIA data for a variety of purposes. FTC staff use the affected EIA data to help understand the potential competitive effects of proposed mergers in the electric power industry.(16) The EIA data are incorporated in computer simulation models that are an important element in such analysis.(17) State regulatory agencies use computer simulation models in designing regulatory reform programs, in monitoring the transition to competition, and in merger evaluations.(18) One of the key features of any state's restructuring effort is to avoid the potential exercise of existing market power among generators that previously had been state-sanctioned monopoly providers of electricity. It is important to address market power concerns prior to initiating competition because, as the Commission has stated, "[c]urrent antitrust laws are not designed to address the mere possession of market power or the legitimate acquisition or increase in market power through lawful regulatory processes."(19) EIA data can help in this process. Moreover, customers and suppliers utilize models that use EIA data in making location and investment decisions.

The models are particularly important in evaluating electric power mergers and regulatory reforms because there are many variations in demand and supply conditions that substantially affect the definition of the relevant markets and that are likely to accentuate or moderate the potential anticompetitive effects of proposed mergers or regulatory reform plans.(20) Computer simulation modeling is also important because large numbers of generators, transmission lines, and loads (including those of numerous independent third parties) contribute to the supply and demand conditions affecting consumers of the firms involved in any potential merger or in any proposed regulatory reform program. Consequently, it is often extremely difficult to understand the effects of proposed mergers and regulatory reforms without the aid of such comprehensive modeling.

These difficulties in analysis arise in part because large-scale storage of electric power is not practicable either by suppliers or by customers.(21) As a result, generation of electric power must closely match demand continuously in order to maintain electric system reliability. Electric power markets are further complicated by the physics of electric power transmission. In order to move electric power from generators to customers, the electric power must be transmitted over a network of high voltage wires that are subject to congestion, particularly during peak demand periods. As a result of the lack of practicable storage and of congestion on the transmission system, supply and demand conditions affecting a given location often differ dramatically over time. Suppliers that are important in some periods, for example, are largely outside the relevant market at other times.

In our view, the need for public access to affected EIA survey data is likely to be more acute at present than it will be in the future. If the transition from regulated monopolies to effective competition moves forward nationally and in additional states, the urgency of additional transition planning and monitoring through computer simulation models using public EIA data may eventually diminish. Consequently, DOE may wish to periodically review its confidentiality proposals, even if it determines that the present proposals are not advisable at this time.

III. DOE MAY WISH TO CONSIDER GRANTING SELECTIVE ACCESS TO OTHERWISE CONFIDENTIAL DATA

If DOE determines to go forward with confidential treatment of the plant-level data as proposed, it may wish to consider providing selective access to the data for government agencies responsible for regulation and law enforcement that do not already have such access.(22) Such a program could be based on the current one used for Form EIA-412. Incentives to invest and innovate and effects on pricing coordination and monitoring among competitors should be largely unaffected by such access. Providing selective access may help reduce the costs that EIA's proposals impose on state and Federal agencies. Two questions concerning the extent of relief provided by such an arrangement are: (1) whether state and Federal agencies can make use of non-public data in a regulatory proceeding in which parties traditionally have the right to evaluate the same information that is used by the agency;(23) and (2) whether reduced public access to simulation models using the EIA data would raise the costs or reduce the availability or utility(24) of such models.(25) DOE may wish to further address these questions if it considers developing a selective access regime for the EIA data and if the comments received under these Notices are insufficient to form a basis for accurate conclusions.

We note that the FTC has experience in operating a program providing selective access to otherwise confidential data. We have found the confidentiality provisions of the Line of Business (LOB) Program to be satisfactory. EIA may wish to review elements of the FTC's LOB Program if EIA elects to consider establishing a selective access program for its plant-level survey data.(26) Documentation for the FTC's LOB Program is appended to this comment.

IV. CONCLUSION

We agree with DOE that rapid disclosure of detailed electric power generation plant data in a fully competitive environment raises concerns about blunting incentives of owners of these facilities to innovate and invest. Disclosure of such data may also aid coordinated interaction among competitors. At the same time, detailed plant-level data are a key component of the computer simulations that are likely to be important to state and Federal regulatory and law enforcement agencies charged with designing and monitoring the transition from regulated monopolies to effective competition. DOE may wish to consider means to allow public disclosure of plant-level data that reduce the risk that such disclosure will harm incentives and competition during this transition period. In the event that DOE does opt for confidential treatment of the plant-level data as proposed, some of the harm to effective regulatory and law-enforcement oversight might be alleviated by developing a system for selective access to the data for state as well as Federal agencies. We encourage DOE to assess the costs and benefits of the EIA proposals and of alternative approaches before reaching a final determination on expanding confidential treatment of EIA survey data. After a few years, the transition from regulated monopolies to effective competition in the electric power industry may be far enough along that the costs of confidential treatment of EIA survey data will be lower and the benefits will be higher.

Respectfully submitted,

________________________
Jeremy I. Bulow, Director
John C. Hilke, Electricity Project Coordinator
Bureau of Economics

________________________
Susan S. DeSanti, Director
Michael S. Wroblewski, Advocacy Coordinator
Policy Planning
Federal Trade Commission
600 Pennsylvania Ave, NW
Washington, DC 20580

1. This comment represents the views of the staff of the Bureau of Economics and of Policy Planning of the Federal Trade Commission. They are not necessarily the views of the Federal Trade Commission or any individual Commissioner. Inquiries regarding this comment should be directed to John C. Hilke (303-844-3565 or jhilke@ftc.gov).

2. The Federal Energy Administration Act of 1974 (Pub. L. No. 93-275, 15 U.S.C. 761 et seq.) and the DOE Organization Act (Pub. L. No. 95-91, 42 U.S.C. 7101 et seq.) require the EIA to carry out an energy information program in order to access the adequacy of energy resources to meet near and longer term domestic demands. EIA already treats some of the survey data that it collects from industry as confidential. Information on categories of data treated as confidential would continue to be reported in aggregated or averaged form.

3. 65 Fed. Reg. 14562 concerning EIA Form 767 (Mar. 13, 2001) (Notice 1) and 65 Fed. Reg. 14564 (Mar. 13, 2001) (Notice 2). The surviving affected forms covered in Notice 2 include Forms EIA-411, 412, 423, 826, 560, 860A, 861, and 906.

4. Notice 1 at 14563 and Notice 2 at 14565.

5. See, e.g., Letter of the Federal Trade Commission to House Commerce Committee Chairman Thomas Bliley, Analysis of H.R. 2944 (Jan. 14, 2000) (Bliley Letter).

6. The staff of the FTC has commented to FERC on electric power regulation, for example, in Docket Nos. EL00-95-000 et al. (San Diego Gas & Electric Company and California regulatory issues more generally) (Nov. 22, 2000); Docket No. RM99-2-000 (regional transmission organizations) (Aug. 16, 1999) (FTC RTO Comment); Docket No. EL99-57-000 (Entergy transco proposal) (May 27, 1999); and Docket No. RM98-4-000 (merger filing requirements) (Sept. 11, 1998). The staff of the FTC also has submitted comments to numerous state agencies regarding electric power industry restructuring that have been compiled in an FTC Staff Report: Competition and Consumer Protection Perspectives on Electric Power Regulatory Reform (July 2000). The FTC staff comments and report are available at: <http://www.ftc.gov/be/advofile.htm>.

7. The ability of potential entrants to better model the transmission system using the EIA plant-level data also may be particularly valuable during the transition from regulated monopolies to competition.

8. EIA will grant access to Federal regulatory and law enforcement agencies. (EIA provisions of confidentiality, Form EIA-412, for example.) In Section III, we describe how the costs of using the affected EIA data may increase for Federal agencies if public access is curtailed (even if Federal regulatory and law enforcement agencies continue to have access).

9. Notice 1 at 14563. In the past, when the vast majority of generation facilities were owned and operated by regulated utilities that did not directly compete, detailed plant-level data either were not of commercial consequence or were already accessible to the public through regulatory processes at the state level.

10. The incentives to innovate discussed here parallel the economic rationale for the patent system. To warrant making front-end investments in innovation and in commercial development of innovation, an individual inventor or corporation must expect that once commercialization begins, product prices can be held above postinvention production and marketing costs long enough so that the discounted present value of the profits (or more accurately, quasi rents) will exceed the value of the front-end investments. Patents and trade secrets are intended to exclude imitating users for long enough to create or strengthen this expectation. Without such an expectation, investments required for innovation and commercialization of innovations are less likely to be made. See, e.g., F.M. Scherer and David Ross, Industrial Market Structure and Economic Performance, Houghton Mifflin, 1990, Ch. 17.

11. There are many ways in which electric power generators could coordinate their actions in the electric power industry. In wholesale spot markets, generating firms have an incentive to bid their marginal opportunity costs for energy in each of the hours in which they offer power. They have this incentive because the last unit accepted to meet demand sets the market clearing price and all bids are compensated at that price. An example of possible coordinated interaction would be if all of the electric power generators with peaking capacity agreed (tacitly or explicitly) to submit bids that exceed the marginal costs of these units during peak demand periods. This strategy would be in the suppliers' joint interest because all would enjoy higher market clearing prices during peak demand periods as a result. Acting in this manner together reduces the risk of not being dispatched that any one generator would face if it alone raised its bids on peaking capacity.

12. U.S. Department of Justice and the Federal Trade Commission, Horizontal Merger Guidelines (as revised Apr. 8, 1997), Section 2.1.

13. Most states do not yet allow customers to choose their generator and are still served by the monopoly utility provider. Further, several states that had established specific dates in the future for inauguration of customer choice have decided to delay customer choice due to the price and reliability problems encountered under California's customer choice program. One state, Nevada, has repealed its customer choice legislation altogether.

14. FERC Order No. 2000 at 9 ("Indeed, the nature of emerging markets and the remaining impediments to full competition that became apparent in the nearly four years since the issuance of Orders Nos. 888 and 889 . . . have made clear that the Commission must take further action if we are to achieve the fully competitive power markets envisioned by those orders.") See also FERC, Market Order Proposing Remedies for California's Wholesale Markets, Docket No. EL00-95-000 (Nov. 1, 2000).

15. An alternative could be to increase the delay in disclosure of EIA survey information. This option, however, is not without potential problems for Federal antitrust authorities. A lag in disclosure may not be sufficient to prevent competitors from effectively using public disclosures in support of anticompetitive coordinated interaction. This concern is heightened, for example, when cost conditions persist over long periods of time so that information on past cost conditions accurately reflects current cost conditions. We do not have a specific recommendation regarding delays in disclosure of plant-level data. DOE may wish to review with state and Federal agencies the costs and benefits of different publication lags as well as how frequently the data are updated to determine the degree to which each item of information is competitively sensitive. In assessing coordinated interaction, FTC staff have sometimes found that lags of several months to a year or more are not conducive to reaching or effectively monitoring anticompetitive coordinated interaction. As for too short a lag time, the Court of Appeals for the District of Columbia Circuit recently concluded that disclosure lags of up to eight weeks are not sufficient to discourage coordinated interaction. FTC v. H.J. Heinz Co. and Milnot Holding Corp., No. 00-5362, slip op. at 17 (D.C. Cir. Apr. 27, 2001).

16. See, e.g., "Analysis of Proposed Consent Order to Aid Public Comment in the Matter of PacifiCorp et al.," FTC File No. 971-0091, at 4 (Feb. 18, 1998). The FTC withdrew from the proposed consent as of June 30, 1998 because PacifiCorp abandoned the merger. See News Release at <www.ftc.gov/opa/9807/petapp38.98.htm>.

17. According to staff discussions with various commercial computer simulation modeling providers, the fuel and thermal data are the most important affected EIA data.

18. Arkansas, Colorado and Indiana, for example, made extensive use of computer simulation modeling in considering retail competition policies. States may also find the affected EIA data useful in monitoring environmental claims contained in generators' promotional materials. For a general discussion of state use of EIA data, see Statement of Commissioner William Nugent, Maine Public Utilities Commission, on behalf of the National Association of Regulatory Utility Commissioners (Mar. 21, 2001) before the Senate Energy Committee.

19. See Bliley Letter, supra n. 5, at 5.

20. July 2000 FTC Staff Report, supra n. 6, Chapter V, Section A.

21. In other industries, inventories held by suppliers or by customers may moderate demand and supply variations that would otherwise result in increased price volatility.

22. As noted earlier, DOE already has a provision to provide confidential EIA survey data to Federal law enforcement agencies on a confidential basis. DOE could also use a selective access approach to providing more recent data to state or local regulatory and law enforcement agencies in the event that it determines to increase the lag before public disclosure and in the event that a state or local agency (or an exempted party such as counsel for a litigant, or an interested party in a regulatory proceeding) believes that early access would materially aid its investigation or deliberation.

23. In FTC litigated matters, outside counsel for parties often obtain access to confidential data from third parties after signing a confidentiality order issued by a judge. These confidentiality orders typically shield third party data from management and operating employees of the parties. Post-litigation confidentiality orders generally include specific periods of time for which different types of information used in reaching the decision are to remain confidential. (Commission Rule 3.45, 16 CFR 3.45 -- revised six weeks ago -- includes a rebuttable presumption that in camera treatment in administrative proceedings before the FTC should expire after three years.)

24. A concern could be that, due to new confidential treatment of data integral to existing models, computer simulation vendors would face insufficient demand for their services to sustain the complexity and sophistication of their modeling efforts. If so, agencies seeking such services could find fewer suppliers and require additional customization, both of which could increase prices.

25. The physics of electric power transmission are such that modeling of even a small section of the transmission grid requires detailed plant-level data for a large number of generators, including generators not involved in a proposed merger or other restructuring. While regulators and investigators may be able to obtain detailed plant-level information under subpoena (or through other procedures), the costs of doing so are likely to be high for both the agencies and the generators complying with government data requests. Compiling and processing this information on a one-time basis from many sources that potentially maintain it in different formats using different software is also likely to be time consuming and prone to inconsistencies.

26. If EIA were to operate such a selective access program, the program could include the requirement that all relevant personnel from a requesting agency sign a confidentiality statement before being granted access to the data. If the requesting agency sought to involve a computer simulation vendor in its use of EIA confidential data, the vendor's employees could also be required to sign the confidentiality statement. Under the FTC's LOB Program, researchers can obtain access to confidential line-of-business data (for example, annual sales, profits, and advertising expenditures of a specific line of business within a firm) on the condition that each of these individuals signs a confidentiality agreement and publishes only aggregated statistics when utilizing the confidential data.