DOJ Antitrust Division, Business Review Request Letter 92-8, Super Efficient Refrigerator Program (SERP) (1992)
Robert A. Collier (1917-1984)
Thomas F. Shannon
William W. Scott
David A. Hartquist
R. Timothy Columbus
Lauren R. Howard
Paul D. Cullen
Kathleen E. McDermott
Michael D. Sherman
Mark L. Austrian
John B. Williams
Paul C. Rosenthal
James R. Loftis, III
John L. Wittenborn
Jeffrey L. Leiter
Michael R. Kershow
Jeffrey S. Beckington
Judith L. Oldham
Jeanne M. Forch
Laurence J. Lasoff
Christopher J. MacAvoy
Patrick J. Coyne
Karen M. Lockwood
Dennis J. Whittlesey
Douglas W. Charnas
William C. MacLeod*
Sean F. X. Boland
Patrick B. Fazzone*
K. Michael O'Connell
William J. Rodgers
*RESIDENT IN AUSTRALIA
Attorneys-at-Law
3050 K Street, N.W.
Washington, D.C. 20007
Telephone: (202) 342-8400
Telecopier: (202) 338-5534
Writer's Direct Dial Number
(202) 342-8480
June 17, 1992
Kathleen Weaver Cannon
Daniel J. Harrold
T. Michael Jankowski
Mary T. Staley
Robert M. Huber
R. Randal Black
Marcy M. Rehberger
William A. Henry
J. Keith Ausbrook
Robin A. Fastenau
Robin H. Gilbert
Martin A. Wright
William M. Guerry, Jr.
Bernard A. Nigro, Jr.
Dawnmarie D. Sanok*
Carolyn O. Tillman
Kerrie L. Hook
Alexander H. Pitofsky
Virginia R. Metallo
Joanna K. McIntosh
Catherine A. Micklitsch
Sean L. Collin
Andrea B. Wenderoth*
Jason M. Branciforte
Stephen A. Jones
Alan A. B. McDowell*
William W. Funderburk, Jr.*
Carole Klein
Lisa A. Jose
Jeffrey S. Longsworth*
John E. Villafranco*
Elise Kirban
D. Hamilton Peterson*
The Honorable Charles A. James
Acting Assistant Attorney General -- Antitrust Division
U.S. Department of Justice
10th Street & Pennsylvania Avenue, N. W., Room 3109
Washington, D.C. 20530
Dear Mr. James:
This is a request for a business review letter pursuant to the Business Review Procedure of the Department of Justice, 28 C.F.R. § 50.6. This request is submitted on behalf of the Super Efficient Refrigerator Program, Inc. ("SERP"). SERP requests a statement of the Antitrust Division's present enforcement intention with regard to SERP's plan to provide a financial incentive to a manufacturer who will develop and market commercial models of refrigerator/freezers ("R/Fs") meeting energy efficiency and environmental criteria adopted by SERP. The full scope of SERP is summarized below and is set forth in detail in the SERP Membership Information Memorandum (Attachment 1), and the Draft Request for Proposals to Produce Super Efficient Refrigerators (Attachment 2).
SERP is a nonprofit corporation organized under the laws of California by certain utilities. SERP's goal is to design and implement a program to offer an incentive to a manufacturer of R/Fs who will develop and market between 1994-1997, approximately 250,000 R/Fs that are at least 25%-50% more efficient than 1993 federal standards
The Honorable Charles A. James
June 17, 1992
Page 2
mandate. (See Attachment 1). The technology necessary to achieve these objectives is, for the most part, available today, and in SERP's opinion, under presently-anticipated circumstances, it will be commercially practicable.1/ The mission of SERP is to offer incentives to manufacturers to help advance the technology of super efficient R/Fs and bring energy efficient and environmentally friendly R/Fs to consumers years ahead of normal market expectations. SERP itself would not manufacture R/Fs or purchase R/Fs for resale.
In the next five or six years, it will be necessary for R/F manufactures to redesign their existing models to change the insulation from chlorofluorocarbon ("CFC") to a material that is CFC-free because R/Fs relying on CFC technology will be banned by 1999, at the latest and perhaps as early as 1995. SERP wants to use this opportunity to influence the new design in a way that makes it more energy efficient.
The driving force behind its members' support for the SERP program is the goal to reduce electricity consumption. Because of the significant costs of new power plant construction and maintaining aging fossil fuel and nuclear facilities, and federal and state governmental programs creating economic incentives to the utilities to reduce consumption, the utilities benefit from decreased rather than increased electricity consumption. In addition, reduced consumption of electricity reduces the external costs that generating electricity creates, e.g., air, water, and thermal pollution.
By engaging utilities to commit to market-motivating, efficiency incentives now, SERP can bring highly efficient, non-CFC R/Fs into the market in the mid-1990's. Utilizing this forward-thinking approach and a unique program design, SERP's program will help participating utilities maximize the economic and environmental benefits of their demand-side management investments.
SERP will continue, through the summer of 1992, to solicit commitments from utilities to contribute to a winner-take-all bid pool. (See Attachment 2). Manufacturers will bid to win the pool through a Request for Proposal process with the winner committing to develop and distribute the super efficient R/Fs. (See Attachment 3). It is anticipated that over $30,000,000 will be invested into the award pool. As of May 1, 1992, prospective participating utilities have begun the process of internal and regulatory approvals for commitments of $27,500,000 to the pool. Currently, a national campaign is ongoing to recruit as many utilities as possible to join SERP.
1/ Some of the technology and materials necessary for construction of the super efficient CFC-free R/Fs include improved insulation, higher efficiency motors and compressors, and improved sealing capabilities.
The Honorable Charles A. James
June 17, 1992
Page 3
The following utilities are active in the program design: Atlantic City Electric Company, Arizona Public Service, Bonneville Power Administration, Central Maine Power & Light, Long Island Lighting Company, Los Angeles Department of Water and Power, New England Power Service, Northern States Power Company of Minnesota, Pacific Gas & Electric Company, PacifiCorp, Sacramento Municipal Utility District, Southern California Edison, and Western Area Power Administration. Many additional utilities have expressed interest and are expected to join SERP before the end of 1992.
The following organizations have endorsed and provided assistance to SERP: American Council for an Energy-Efficient Economy, Consortium for Energy Efficiency, Inc., Electric Power Research Institute, National Association of Regulatory Utility Commissioners, Natural Resources Defense Council, and U.S. Environmental Protection Agency. (See Attachment 3).
The SERP program contemplates the assembly by SERP of funds and commitments from members, a competitive solicitation by SERP of offers from manufacturers of R/Fs, and the payment by SERP of an incentive specified in the contract awarded to the successful manufacturer. SERP anticipates that during the course of the program, substantially all of SERP's membership commitments will be expended in the form of incentive payments to the successful manufacturer, and for SERP administration and operations.
The SERP program would invite R/F manufacturers to submit proposals on a new CFC-free energy efficient R/F design. SERP will require that the proposed R/F have an automatic defrosting capability and an interior capacity of between 15 and 24 cubic feet. Any R/F manufacturer could submit a proposal. However, bidders would be limited to those firms that currently own, lease, or control, or propose to develop a facility in North America for the production of full-size R/Fs. Proposals from firms that historically have not demonstrated the ability to produce and market 100,000 R/Fs per year, would be required to demonstrate to the satisfaction of SERP the ability to produce and market R/Fs nationwide. The two bidders with the best proposals would be required to submit prototypes of their proposed energy efficient non-CFC R/F. SERP would select the winning R/F bidder after analyzing the prototypes. The winning bidder must propose the number of R/Fs that it will produce and a schedule that sets out when those R/Fs will be delivered to retail stores. The bidders' proposals that will receive the highest marks are those reflecting the most improvement in energy efficiency relative to 1993 DOE standards and the volume of R/Fs the bidder proposes to produce and market.
The Honorable Charles A. James
June 17, 1992
Page 4
SERP would require that each proposal contain the number of energy efficient non-CFC R/Fs that the bidder proposes to produce and tender for sale into the service areas of SERP's participating members. SERP expects the bidders to propose to sell approximately 250,000 units in total for the relevant period. In order to qualify for SERP's incentive payment, a sale of such unit would be made to a retail outlet in the service territory of a participating member of SERP. The number of qualifying sales eligible for incentive payments would be based on the amount that members will have invested in the SERP program. SERP would require that the winner provide to SERP information identifying the volume of sales its makes of the new R/Fs to each retail outlet in each of its members' service areas. In addition, SERP would require that the winner also identify at least 25 percent of the final customers who purchased the non-CFC R/F within each of its members' service areas. SERP would use this information to estimate the number of energy efficient non-CFC R/Fs that are sold in each participating member's utility service area. Nothing would limit the manufacturer or any entity in its chain of distribution from making other sales of the new R/F into other regions of the country.
SERP would require that the winning bidder begin delivery of the new R/Fs by 1995, and final deliveries must be made by June 30, 1997. Bidders' proposals will be graded higher for earlier delivery schedules within that time window. The winning bidder would receive a prorata share of the award as R/Fs are delivered into the service areas of SERP's members.
SERP also would require that the winning bidder tender the new products to its customers, i.e., R/F distributors and retailers, at a price level that is no higher than the price of an existing comparable traditional model R/F that it sells. This initial manufacturer's price cap, however, would not apply to the wholesale prices or retail prices determined by distributors or retailers respectively for the new energy efficient non-CFC R/F. They would remain free to price the products as they may choose. The manufacturer's price cap provision would be intended to promote rather than hinder sales of the new R/F. Furthermore, the manufacturer may ignore the price cap and, of course, forego SERP's incentive payments.
This section analyzes the SERP program in terms of its likely competitive impact on the relevant economic markets it implicates. From an antitrust perspective, SERP is benign in that the program has no horizontal or vertical market effects. SERP members have no competitive relationship with R/F manufactures, i.e., they do not compete with R/F manufacturers in the sale of R/Fs. Likewise, SERP members as a general proposition do not compete with each other in the sale of electricity to residential
The Honorable Charles A. James
June 17, 1992
Page 5
consumers. SERP members also have no vertical relationship with R/F manufacturers, i.e., they do not supply any components to R/F manufacturers required in the production of R/Fs.
The contemplated SERP program would entail the following conduct, which is analyzed below.
In the downstream market for R/Fs:
1. SERP contemplates a "winner takes all" award structure. The winner's initial costs of producing and marketing the energy efficient non-CFC R/F into SERP members' service territories potentially will be reduced by the amount of the award.
2. SERP contemplates requiring that the winning bidder tender the new products to its customers, i.e., R/F distributors and retailers, at prices no higher than that same manufacturer's prices on existing comparable traditional models. The winning manufacturer would not be bound by SERP's contemplated price cap, but its observance would be condition for it to receive SERP incentive payments.
3. Hypothetically, if the initial cost (even after taking into account the SERP award) of the new energy efficient non-CFC R/F exceeds the price of an existing comparable CFC R/F, the winning bidder may not be able profitably to market its newly designed R/F. If the winning bidder is unable to market the required minimum number units, it would forego the award. Thus, SERP does not eliminate marketing risks associated with the sale of the new products.
4. While the manufacturer's price of the energy efficient non-CFC R/Fs may be constrained by the price of existing comparable CFC-using R/Fs, if the manufacturer elects the SERP incentive option, nothing constrains the wholesale or retail prices of the new R/Fs. SERP would not maintain or dictate wholesale or retail prices. It is possible that a temporary wealth transfer from the winning manufacturer to its distributors or retailers may take place due to the price cap, but ultimately the competitive market will adjust for that.
The Honorable Charles A. James
June 17, 1992
Page 6
In the upstream market for the provision of electricity:
1. An industry consortium would attempt to encourage through a financial incentive the future direction of the downstream market for R/Fs.
2. The industry consortium may receive a return on its investment in SERP from its respective public utility commissions if they could demonstrate that the program yielded benefits in terms of reduced electricity consumption.
As discussed below, the SERP program would not cause any anticompetitive effects in any relevant market, and in all likelihood would have procompetitive, consumer welfare enhancing effects through its stimulation of innovation in the R/F market.
A. The Relevant Product and Geographic Markets.
1. Relevant downstream product market.
The appropriate relevant product market in which competition is implicated by the SERP program is full-size R/Fs, defined as those R/Fs between approximately 15-24 cubic feet. These R/Fs are most often produced for residential use.
From the demand side, although purchasing decisions are made based on price, features and aesthetics, consumers likely would consider most full size R/Fs functional and economic substitutes. Small, portable refrigerators, e.g., those found in dormitory rooms and recreation rooms, however, probably should be excluded from this market definition because, except at the margin, they in all likelihood are neither functional nor economic substitutes for full-size residential R/Fs. Likewise, for similar reasons, large commercial R/Fs should be excluded from this market definition.
From the supply side, while certain differences in R/F format are reflected in product offerings by incumbent producers, as a general proposition, all producers are equally capable of producing R/Fs with the various formats and features. Therefore, from the supply side, R/Fs in the 15-24 cubic feet range properly should be considered in the same relevant market.
The Honorable Charles A. James
June 17, 1992
Page 7
The appropriate relevant geographic market in which to analyze the competitive effects of the SERP program on the R/F market is the United States. Sales of all domestic producers of R/Fs are nationwide in scope. Because R/Fs are typically sold through national distribution networks and through national retail chains, e.g., Sears, and the sales are supported by high visibility advertising programs, it is highly unlikely that geographic price discrimination is possible. Even if geographic price discrimination were possible, because the R/F producers sell their products nationally, the firms necessary to accomplish the price discrimination would be the same as those participating in a national market.
In 1990, approximately four million units of compressor-type CFC R/Fs between 15.5 - 21.4 cubic feet were produced in the United States.2/ The value of these shipments was approximately $1.45 billion.3/ Five U.S. manufacturers (GE, Whirlpool, Electrolux, Maytag, and Raytheon) account for approximately 96 percent of all R/Fs sold in the United States.4/ Basic refrigeration technology is well known and not covered by patents. Among the five major competitors, market shares fluctuate over time in response to innovations such as automatic ice makers, refrigerator/freezer configurations and other add-on features, e.g., indoor access panels. In addition, R/Fs are sold in both branded and private label configurations. The product is differentiated by functional capability, e.g., size, automatic defrost, automatic ice-maker, electronic diagnostic, in-door convenience access panel, and maintenance requirements.
Several of the most significant, large buying chains, such as Sears, own the brand names they sell. Their inventory is produced by various of the manufacturers to the chain's specifications. R/Fs are sold by manufacturers primarily to distributors, large retailers, and governmental entities. Distributors in turn may sell to appliance stores and retailers. R/Fs are purchased by consumers -- as replacements for home use,
2/ "Current Industrial Reports: Major Household Appliances," U.S. Department of Commerce, Bureau of the Census, Economic and Statistics Administration (Washington, D.C., U.S. Government Printing Office, August 1991) at 3.
3/ Id., p. 3.
4/ 1990 market shares for U.S. R/F producers were as follows: GE, 34%; Whirlpool, 23%; Electrolux, 20%; Maytag, 12%; Raytheon, 7%; Others, 4%.
The Honorable Charles A. James
June 17, 1992
Page 8
builder/developers for residential and commercial uses. Commercial uses include: offices, restaurants, hotels/motels, hospitals, military and other governmental entities.
As noted above, the purpose behind the SERP program is to reduce electricity consumption. From the utilities' perspective it is advantageous to them over time to reduce consumption rather than increase the availability of electricity capacity. The benefits of reducing electricity consumption allows for reducing the need for new power plant construction and the elimination of marginal power plants now in existence. In addition, public utility commissions throughout the country offer incentives to utilities to undertake programs that reduce demand for electricity. Moreover, federal and state laws mandate that new appliances meet certain energy efficiency standards. Finally, long term reductions in electricity demand reduce negative externalities, e.g., air, water, and thermal pollution that results from electricity generation.
R/Fs account for approximately 22 percent of total residential electricity consumption. This large amount of electricity consumption on the part of R/Fs, coupled with the fact that an R/F's useful life expectancy is approximately 20 years are reasons why SERP targeted R/Fs as the focus of its initiative. Equally important, however, is the fact that CFCs will be banned in R/Fs by the turn of the century, if not sooner. Thus, future R/Fs will have to be redesigned anyway. Given these facts, SERP would encourage R/F manufacturers to give significant consideration to electrical efficiency in a new design non-CFC R/F, and would like to accelerate production of that product.
The SERP program does not alter the market's equilibrium in any material way. The winning bidder will be required to produce approximately 250,000 R/Fs over a three year period, or approximately 83,500 per year. In comparison, in 1990 approximately four million R/Fs were produced in the 15.5-21.4 cubic foot range alone.5/ Thus, the impact of the SERP program on the product market will be minimal in terms of the impact on relative market shares. The same number of producers and sellers will exist if the SERP program is implemented and is successful as would exist in the absence of the SERP program. SERP will simply accelerate the development of super efficient, CFC-free R/Fs. In terms of the existing market and demand characteristics for R/Fs, SERP is unlikely to alter the types of configurations and features incorporated in
5/ "Current Industrial Reports: Major Household Appliances," at 3.
The Honorable Charles A. James
June 17, 1992
Page 9
existing models nor change R/F functional characteristics. Although it is possible, that over time, market shares may fluctuate as a result of the introduction of the SERP prompted developments, this should not be considered an antitrust concern. Market share fluctuations in response to innovation are in most cases procompetitive and at worst competitively neutral.
No change would result in the relevant geographic market. The ability of incumbent firms to sell existing R/F models or super efficient, CFC-free models through existing distributors and retail networks would be unaffected by the SERP program. While SERP would require the winning bidder to sell a specified number of R/Fs in the geographic proximity of SERP's members, nothing prohibits the winner also from marketing such R/F's elsewhere. Given the production requirements of the SERP program (approximately 83,500 R/Fs per year) relative to the total U.S. production of full-size R/Fs (in excess of four million),6/ as well as the fact that there are no requirements for exclusive use of the new energy efficient non-CFC R/F, the SERP program would have no effect on the scope of the geographic market.
With regard to the downstream market for the production of R/Fs, the net effect of SERP would be to accelerate the development of the next generation of R/Fs by creating an economic incentive for one firm to develop and put into production the product sooner than it otherwise would. Due to the fact that the SERP program does not have a significant effect on competition from either a horizontal or vertical perspective, the SERP program is competitively benign. However, because of the complementary nature of the two products, electricity and R/Fs, SERP wants the Division's approval of its program before its members undertaken such a large investment. Set forth below are aspects of the program which require analysis.
While SERP contemplates a "winner takes all" award, the structure of SERP is intended to move not just one firm but the entire R/F industry toward greater energy efficiency. Because SERP members receive benefits from reduced electricity consumption, the more new R/Fs that are produced and sold, the greater the benefit
6/ The number of U.S. shipments of R/F over 13.5 cubic inches was 6.9 million in 1990. (See "Current Industrial Reports: Major Household Appliances," at 3).
The Honorable Charles A. James
June 17, 1992
Page 10
to SERP and its members. Thus, there is no incentive to SERP to offer a program that would lead to collusive behavior (i.e., supply restricting behavior) on the part of R/F manufacturers. This is a reason why the SERP program has been structured so that SERP members have no direct ownership interest in the winning design (i.e., no joint venture relationship to the winning bidder).
In addition, nothing would prevent others from designing a program similar to SERP, and SERP's effort may well encourage other firms and/or consortiums to finance similar or other technological developments in the industry. Thus, SERP is only one of possibly numerous other investment ventures that may be created to stimulate energy efficiency innovation.
The SERP bidding process requires the bidders with the best two proposals to each produce a prototype energy efficient non-CFC R/F. SERP will choose the winning bidder based on its evaluation of the prototypes. Thus, the company with the second-best bid will be a prime candidate to complete development and market a CFC-free R/F. In addition, by encouraging all industry members to participate in the program, SERP believes that the program will spur more R&D expenditures, collectively, on energy efficient CFC-free R/Fs than the $30 million it will award to the winning bidder.
An additional factor relating to the winner take all aspect of the program is that by providing the financial incentive for one firm to develop the next generation R/F now, the winner could have an initial production cost advantage vis-a-vis competitors which also are interested in developing the new R/Fs, but do not receive the incentive. But, because the award is so small, compared to the overall market for R/Fs ($30 million vs. $1.45 billion in sales in 1990), the conditions are not present for predation to occur.
Finally, nothing in SERP's proposed program suggests that SERP's involvement could be construed as creating an entry barrier into the production of energy efficient non-CFC R/Fs. First, SERP is encouraging rather than discouraging competition and innovation (an essential component of competition). Moreover, the significance of large chain retailers who own their own valuable brand names (Sears/Kenmore) and buy from a variety of suppliers, would serve as a buffer to any attempt by manufacturers to exercise market power or deter entry. Finally, in addition, to the extent that patentable technology is developed, SERP will require that the winning bidder license its patents to any other manufacturer after two years.7/
7/ Based on presently available information, SERP does not believe that the winning design would result in patentable technology.
The Honorable Charles A. James
June 17, 1992
Page 11
SERP's most likely ancillary competitive effect (as opposed to other potential goals and effects), will be to stimulate competition in the R/F industry to facilitate the production of the next generation of energy efficient, CFC-free R/Fs.
SERP would require that the winning bidder maintain a manufacturing price which is not more than the comparable price of similar CFC R/Fs of the same manufacturer, if the manufacturer wants to avail itself of the SERP incentive payment. Nothing would prevent a winning manufacturer from ignoring the incentive offer. If the manufacturer accepts the incentive, this provision would require that the winning bidder design a CFC-free R/F that effectively competes with its own existing CFC R/Fs. The price provision, if accepted, would help assure that consumers will have an additional R/F to choose from among those R/Fs within specific price ranges. In this context, this type of maximum price provision yields a procompetitive outcome.8/ The price cap was decided on because it is SERP's goal to make available the new efficient, CFC-free R/F to a large number of consumers, not just to those customers who are willing to pay more for the opportunity to own a more environmentally friendly, energy efficient R/F. It also provides some comfort to SERP members that a manufacturer will not both take their incentive money and reap extra returns from high prices to their customers.
The price cap would not affect the price of comparable existing CFC-using R/Fs, even for the winning bidder. Competition between all R/F manufacturers would determine the price of existing R/Fs. In a hypothetical case where demand outstrips the available supply of the new energy efficient non-CFC R/F, SERP would expect distributors and retailers to increase their price of the new non-CFC R/F.9/ If such a scenario were to occur, the distributors and retailers would receive a temporary increase in their profits, i.e., they would receive economic rents due to SERP's price requirement. This outcome, however, would create strong incentives for other R/F manufacturers to produce and market their own energy efficient non-CFC R/Fs and for the winner to increase its own output. As those R/Fs entered the marketplace, one would expect the price of energy efficient non-CFC R/Fs to fall.
8/ The price ceiling raises none of the possible adverse effects identified by the Supreme Court in the vertical maximum price fixing case Atlantic Richfield Co. v. USA Petroleum Co., 110 S. Ct. 1884, 1891 (1990).
9/ Recall that these firms fall outside of SERP's price cap provision.
The Honorable Charles A. James
June 17, 1992
Page 12
3. No Anticompetitive Effects Result from the SERP Program in the Markets for the Provision of Electricity.
No adverse competitive impact would result in markets for electricity as a result of the SERP program. As a practical matter, SERP's members, while industry participants, do not compete with each other for the provision of electricity to residential consumers.
Their goal, through participation in SERP, is to reduce the economic costs of providing electricity to consumers by reducing the requirement for new investment in high-cost electricity production facilities. In addition, the resulting reduction of electricity production that flows from reduced usage also reduces the external costs that are borne by consumers. These external costs include, among others, the air, water, and thermal pollution that are created in the production of electricity. SERP believes that these goals will result in both long term and short term lower electricity prices to consumers, a result that is procompetitive and consumer welfare enhancing.
For all of the above reasons, the proposed SERP program raises no significant antitrust concerns and, in fact, would likely result in procompetitive benefits. Therefore, SERP requests that the Division provide it with a statement that given the above-described circumstances, the Antitrust Division would not take any enforcement action against the proposed program.
Thank you very much.
Sincerely,
James R. Loftis, III
Robert M. Huber