Echlin Manufacturing Company

Text (OCR of the scan at left; may contain errors)

IN THE MATTER OF THE ECHLIN MANUFACTURING COMPANY, ET AL.

DISMISSAL ORDER, ETC. , IN REGARD TO ALLEGED VIOLATION OF SEC. 5 OF THE FEDERAL TRADE COMMISSION ACT AND SEC. 7 OF THE CLAYTON ACT Docket 9157. Complaint, July 23, 1981-Final Order, June, 1985 The Federal Trade Commission has dismissed its antitrust challenge to The Echlin Manufacturing Co. s acquisition of Borg-Warner Corp. s automotive-aftermarket operations. The Commission ruled that since there are no barriers to entry into the market for the assembly and sale of carburetor kits there can be no anticompetitive effect from the acquisition, and no violation of the antitrust laws, Appearances For the Commission: Ann B. Malester, Karen E. Chandler and Linda Martin.

and David For the respondents: Basil J. Mezines, Glenn A. Mitchell David U. Fierst, Stein, Mitchell Mezines, Washington, D. C. and Spiller in-house counsel, Branford, Conn., for respondent Echlin Mfg. Kieler and Blake Harrop, Co. Earl E. Pollock, Gary Senner, Louis C. Chicago, Ill. and CherylSonnenschein, Carlin, Nath Rosenthal Longtin, in-house counsel, Chicago, Ill., for respondent Borg-Warner Corp.

COMPLAINT The Federal Trade Commission, having reason to believe that the Respondents, The Echlin Manufacturing Company ("Echlin ) and Borg-Warner Corporation ("Borg-Warner ), subject to the jurisdiction of the Commission, have entered into an agreement which violates Section 5 of the Federal Trade Commission Act, as amended, (15 U.sC. 45); that Echlin has acquired those assets of Borg-Warner described in Paragraph 11 and that such acquisition constitutes a violation of Section 7 of the Clayton Act, as amended, (15 U.s.C. 18) and Section 5 of the Federal Trade Commission Act, as amended; and it appearing that a proceeding in respect thereof would be in the public interest, the Commission hereby issues its Complaint, pursuant to Section 11 of the Clayton Act (15 U. C. 21) and Section 5(b) of the Federal Trade Commission Act (15 U. C. 45(b)), stating its charges as L:A.,Il.LJll lVf\l U.l'ACTURING CO., ET AL. 411 410 Complaint L THE ECHLIN MANUFACTURING COMPANY 1. Echlin is a corporation organized and doing business under the laws of Connecticut, with its principal offce at 175 North Branford Road, Branford, Connecticut.

2. For the fiscal year ending August 31 1980, Echlin s consolidated operating revenues were approximately $301.4 milion and its net income was approximately $8.8 millon. As of August 31 , 1980, Echlin had total assets of approximately $237 milion. 3. Echlin s major area of business is the manufacture and sale of replacement automotive parts. Echlin s products include: carburetor kits; carburetor and emission control parts; ignition system parts; automotive diagnostic equipment; turbochargers; hydraulic and air brake parts and assemblies; automotive wire and cable products; clutch and electrical system components used by remanufacturers; and parts for maintaining lift trucks and small gasoline engines. (2) 4. Echlin is engaged in the sale and shipment of products, including carburetor kits, throughout the United States; and is engaged in or affects commerce within the meaning ofthe Clayton Act, as amended and the Federal Trade Commission Act, as amended. II. BORG-WARNER CORPORATION 5. Borg-Warner is a corporation organized and doing business under the laws of Delaware, with its principal offce at 200 South Michigan Avenue, Chicago, Ilinois.

6. In 1980, Borg-Warner s consolidated manufacturing revenues were approximately $2.7 bilion and its net income was approximately $126 millon. As of December 31 , 1980, Borg-Warner had total manufacturing assets of approximately $1.9 bilion. 7. Borg-Warner s major areas of business are the manufacture and sale of transportation equipment, chemicals and plastics, air conditioning equipment, and industrial products. Borg-Warner also has subsidiaries engaged in the business of financial and protective ser- 8. In 1980, Borg-Warner s Transportation Equipment Group hadvices. sales of approximately $903.6 million and earnings of approximately $31.5 million. The group manufactured and/or supplied products for use in new vehicle production and for replacement use. These products included: carburetor kits; automatic and manual transmissions and transmission components; ignition system parts; individual carburetor and emission control parts; clutches and clutch components; four-wheel drive units; axles; radiators; and automatic slack adjusters. These products are used on passenger cars, trucks, offhighway vehicles, and in farm and marine applications. Complaint 105 F. C: - 9. As of July 13, 1981, all of the divisions of Borg-Warner involved in the manufacture andor sale of replacement automotive parts were included within the Transportation Equipment Group. Included among such divisions that manufactured and! or sold replacement automotive parts were: the Automotive Parts Division ("APD"); the Ballwin-Washington Division ("Ballwin-Washington ); the Ottawa Division ("Ottawa ); APD International; and APD Borg-Warner (Canada) Limited. APD has been engaged in the sale of a variety of automotive products to the replacement channels of distribution, including: carburetor kits; carburetor and emission control parts; fuel pumps; ignition system parts; automotive wire and cable products; and new and remanufactured clutches. Ballwin-Washington s primary business has been the assembly of carburetor kits and the production of the vast majority of parts therefore. Ottawa remanufactured clutches for sale by APD to the replacement market. APD International and APD Borg-Warner (Canada) Limited exported replacement automotive parts.

10. Borg-Warner has been and is engaged in the sale and shipment of products, including carburetor kits, throughout the United States; and is engaged in or affects commerce within the meaning of the Clayton Act, as amended, and the Federal Trade Commission Act, as amended. (3) III. THE AGREEMENT OF THE PARTIES 11. On May 12, 1981 , Echlin and Borg-Warner entered into an agreement, effective February 28, 1981, whereby Echlin would acquire all of the assets ofBorg-Warner s automotive aftermarket operations in exchange for 22% of Echlin s common stock. The Boards of Directors ofEchlin and Borg-Warner gave their final approval to this transaction and Echlin s stockholders ratified the agreement on July , 1981. Pursuant to this agreement, Echlin has acquired the assets of the divisions of Borg-Warner s Transportation Equipment Group set forth specifically in Paragraph 9. In exchange therefor, Borg- Warner acquired approximately 4 500 000 shares of Echlin common stock. The agreement also provides for the election of one Borg-Warner representative to the Echlin Board of Directors so long as Borg- Warner owns at least 10% of Echlin s voting securities. Pursuant to a trademark licensing agreement, Echlin and Borg-Warner have agreed that Echlin will receive an exclusive license to use certain Borg-Warner trademarks in connection with the marketing and sale of various automotive products to the "automotive aftermarket" in the United States and Canada and a non-exclusive license for the rest of the world. In addition, Echlin and Borg-Warner have entered into " sUDDlv agreement whereby Borg-Warner wil continue to supply ECHLIN MANUFACTURING CO.. ET AL. 41;j 410 Complaint Echlin for at least fifteen year with the products which Borg-Warner s remaining divisions have historically supplied to the Borg- Warner divisions acquired by Echlin. The transaction was consummated on or about July 14, 1981.

IV. NATURE OF TRADE AND COMMERCE 12. The relevant geographic market is the United States as a whole. 13. The relevant product market is the assembly and sale of carburetor kits.

14. A carburetor kit is a prepackaged assemblage ofthe parts most often replaced on a carburetor. In order to facilitate the use of a carburetor kit, each carburetor kit also contains gauges and an instruction sheet. There is no practical alternative to the use of a carburetor kit.

15. Carburetor kits are used by professional mechanics and, to a limited extent, by vehicle owners themselves. These installers have found that it is easier and cheaper to buy in the form of a carburetor kit all the items which generally need to be replaced on a carburetor. By purchasing a kit, the installer can be assured of obtaining the parts that should be replaced rather than going through the process of deciding whether or not to replace each different part. The installer also saves the time and effort that would be needed to procure all the individual parts. Therefore, installers purchase carburetor kits in preference to the components thereof in the vast majority of instances.

16. Several hundred different carburetor kits are required to provide coverage for virtually all of the vehicles serviced in the United States. Wholesalers usually stock a line of carburetor kits but only limited amounts of some of the individual components included in a kit. This enables the wholesalers to avoid the extra expense of inventorying, warehousing, pulling and biling numerous items for a single repair rather than one carburetor kit.

17. An assembler of carburetor kits performs both an assembly function in creating the carburetor kit and a sales function in sellng the carburetor kit directly or indirectly to the replacement channels of distribution. (4) 18. The assembly of carburetor kits consists of determining which parts should be included in each kit, obtaining a source for the thousands of individual carburetor parts, and packaging the parts together with appropriate instructions and gauges. 19. The assembler of carburetor kits must have the necessary skils to determine which items to include in each kit in order to achieve the appropriate level of consolidation that will minimize the number of __ g., Complaint 105 F.

kits in a line but still not result in significantly higher per unit prices due to the inclusion of extra parts.

20. All carburetor kit assemblers, including Echlin and Borg-Warner, have supplied carburetor kits directly to the replacement channels of distribution. Prior to July 13, 1981, Borg-Warner also sold carburetor kits to firms which resell such kits to the replacement channels of distribution under their own brand names. 21. There are three channels of replacement parts distribution. Most carburetor kits are sold to warehouse distributors who in turn sell them to jobbers for their resale to installers. A second channel consists of sales to vehicle dealers. A third channel consists of sales to direct buying retailers, mass merchandisers. 22. New vehicle manufacturers do not purchase carburetor kits for incorporation in new vehicles. Carburetor remanufacturers rarely purchase carburetor kits.

V. MARKET STRUCTURE 23. Sales of carburetor kits by assemblers in 1980 totalled approximately 12.6 milion units, having a value of approximately $53 million at the level of sales to the replacement channels of distribution. 24. In 1980, Echlin assembled at least 1.2 million carburetor kits which it sold to the replacement channels of distribution. Echlin accounted for approximately 9.8% of all carburetor kits assembled and sold in the United States in 1980.

25. In 1980, Borg-Warner assembled approximately 4.8 milion carburetor kits, making it the largest assembler of carburetor kits and accounting for approximately 38.4% of all carburetor kits assembled and sold in the United States. Borg-Warner sold approximately 1.86 million ofthese carburetor kits directly to the replacement channels of distribution and sold the remainder to other firms which resold these carburetor kits under their own brand names to the replacement channels of distribution.

26. Concentration in the assembly and sale of carburetor kits is extremely high. In 1980, there were only seven domestic companies that assembled carburetor kits and virtually no imports of carburetor kits. In 1980, two-firm concentration in the assembly and sale of concentra- carburetor kits was approximately 63.7% and four-firm tion was approximately 82.6%.

VI. BARRIERS TO ENTRY 27. The barriers to entry into the assembly and sale of carburetor l-- .lL- L.L_ ECHLIN MANUFACTURING CO. , ET AL. 415 410 Complainl VII. ACTUAL COMPETITION 28. Prior to July 13, 1981, Echlin and Borg-Warner were actual competitors in the assembly and sale of carburetor kits in the United States.

VIII. EFFECTS 29. The effect of the aforesaid acquisition may be to substantially lessen competition in the relevant market in violation of Section 7 of the Clayton Act, as amended, 15 U. C. 18, and Section 5 of the Federal Trade Commission Act, as amended, 15 U. C. 45, in the following ways, among others:

(1) it has eliminated actual competition between Echlin and Borg- Warner and between Borg-Warner and others in the assembly and sale of carburetor kits;

(2) it has created one firm accounting for approximately 48% of the relevant market, whose market power in the assembly and sale of carburetor kits vis-a-vis its competitors is greater than that formerly held by either Borg-Warner or Echlin due to the combined firm s role in the assembly of carburetor kits, its degree of vertical integration in parts production for carburetor kits, and its degree of direct sales to the replacement channels of distribution; (3) it has significantly increased the already high levels of concenfirm concentration increas-tration in the relevant market, with two ing from approximately 63.7% to 73. 5%, pro forma and four-firm concentration increasing from approximately 82.6% to 89. 2%, pro forma based on 1980 figures; and (4) it may disadvantage firms supplying carburetor kits directly to the replacement channels of distribution by increasing the likelihood that such firms would be deprived of the access to carburetor kits which such firms currently enjoy.

IX. VIOLATIONS CHARGED 30. The steps taken by Echlin and Borg-Warner to consummate the aforesaid transaction constitute a violation of Section 5 ofthe Federal Trade Commission Act, as amended, 15 U. C. 45. 31. The acquisition of assets set forth in Paragraph 11 constitutes a violation of Section 7 of the Clayton Act, as amended, 15 U. C. 18 and Section 5 of the Federal Trade Commission Act, as amended, 15 C. 45.

..

Initial Decision 105 F.T. INITIAL DECISION BY MONTGOMEBY K. HYUN, ADMINISTRATIVE LAW JUDGE SEPTEMBER 14, 1984 PRELIMINARY STATEMENT On July 23, 1981, the Federal Trade Commission ("Commission issued a Complaint challenging the 1981 acquisition of certain automotive aftermarket assets of Borg-Warner Corporation ("Borg- Warner ) by Echlin Inc. ("Echlin ) as a violation of Section 5 of the Federal Trade Commission Act, as amended (15 U. C. 45), and of Section 7 of the Clayton Act, as amended (15 U.s.C. 18). The Complaint alleges that the relevant product market in which to assess the competitive effect of the acquisition is the "assembly and sale of carburetor kits " and that the acquisition combined two firms with 8% (Echlin) and 38.4% (Borg-Warner) ofthe market. The Complaint alleges that prior to the acquisition, Echlin and Borg-Warner were direct competitors, ranking third and first, respectively, in the relevant product market and that, as a result of the acquisition, Echlin became the dominant firm in the assembly and sale of carburetor kits. The Complaint further charges that the effect of the acquisition may be to substantially lessen competition in the assembly and sale of carburetor kits. Through its acquisition of Borg-Warner s automotive aftermarket assets, Echlin: (1) eliminated actual competition between itself and Borg-Warner and between Borg-Warner and other carburetor kit assemblers; (2) significantly increased concentration levels in an already highly concentrated market; and (3) created a dominant firm accounting for almost 50% of the market. Respondents fied their Answers to the Complaint on August 27 1981, admitting in part and denying in part the various allegations of the Complaint.

Under the supervision and control of my predecessor, Administrative Law Judge John J. Mathias, the parties were allowed an extensive discovery and all pre-trial steps were concluded by June 1983. On August 16, 1983, the instant case was reassigned to me. Echlin August 31, 1983 motion to disqualify me pursuant to Section 3.42(g)(2) of the Commission s Rules of Practice, which I certified to the Commission with a statement of reasons for my decision not to disqualify myself, was dismissed by the Commission by Order of September 13 1983.

ArlillrE(' tivp hp rind"O; llpon nn pntcHn hc... 1') 1 QQ'1 ".. .r ron""E'l.. ECHLIN MANUFACTURING CO., E'1 AL.

410 Initial Decision on April 19, 1984, with 41 days of trial. Testimony was heard.frem a total of 41 witnesses. (Complaint counsel called 12 witnesses and respondents called 30 witnesses, with one witness called to testify by both complaint counsel and respondents.) Presentation of the case-inchiefbegan in Washington, D.C. on September 12, 1983 and continued through October 6, 1983. Respondents' defense began in St. Louis, (3) Missouri on October 17 1983 and continued through October 21 1983. Respondents' defense resumed on November 7, 1983 in Washington C. and concluded on December 6, 1983. Complaint counsel presented a rebuttal case with hearings held on April 18, 1984 through April 19, 1984. Respondents offered several surrebuttal exhibits that were received in evidence on May 1, 1984. The record, which includes a transcript of 5 327 pages and over 750 exhibits, was closed on May 1 1984,1 References to the record are made in parentheses and the following abbreviations are used:

CPF - Complaint Counsel's Proposed Findings; CRB - Complaint Counsel's Reply to RPF and RB; CX - Commission s Exhibit;

RPF - Respondents' Proposed Findings;

RB - Respondents' Memoranda of Law in Support of RPF; RX - Respondents' Exhibits;

RRB - Respondents' Reply to CPF.

SB - Supplemental Brief of Borg-Warner Corporation SRB - Supplemental Reply Memorandum of Borg-Warner Corporation The transcript citation of a witness' testimony is referred to with the last name of the witness and the page number(s) upon which the testimony appears. Other transcript citations are referred to as "Tr. followed by the page number(s).

Definitions For the purposes of this Initial Decision, the following definitions apply:

(a) The replacement market includes all sales by manufacturers of automotive parts for use as replacement of original equipment parts or of previously replaced parts.

(b) The automotive aftermarket is used synonymously with the term replacement market. (4) and carbu. (c) The terms automotive carburetor kits, carburetor kits retor tune-up kits include kits used to repair carburetors on domesti, 1 By order dated May 31, 1984, the Commissioll extended the due date of this Initial Derision to September l' 1984.

Initial Decision 105 F. and foreign automobiles and light trucks, industrial and agricultural equipment and inboard marine engines.

The proposed findings and conclusions submitted by the parties and their arguments in support thereof have been given careful consideration by me and to the extent not adopted by this Initial Decision, in the form proposed or in substance, are rejected as not supported by the evidence or as immaterial. Any motion appearing on therecord not heretofore or hereby specifically ruled upon either directly or by the necessary effect of the conclusions in this Initial Decision are hereby denied.

Upon consideration of the entire record in this proceeding and having considered the demeanor of the witnesses, I make the following findings of fact and conclusions of law and order based on the record considered as a whole:

FINDINGS OF FACT I. DESCRIPTION OF RESPONDENTS A. Respondent Echlin Inc.

1. Echlin Inc. C'Echlin ), formerly The Echlin Manufacturing Company, is a Connecticut corporation with its headquarters in Branford Connecticut (CX 235A). It has plants and offces in Connecticut, Illinois, Florida, California, Kansas, Michigan, Missouri and several foreign countries (CX 235G-H). For purposes of this case, the two relevant plants are Branford, Connecticut, where all design, consolidation and sourcing decisions for carburetor parts and kits are made, and Litchfield, Ilinois, where the kits are assembled and warehoused (Timberlake 3080- , 3072). The sellng functions are carried out in Branford (Timberlake 3080).

2. Echlin sells numerous automotive aftermarket products, including condensers, contacts, complete distributors, distributor caps, ignition coils, rotors, control modules, pickups, sensors, electronic voltage regulators, ignition wire, automotive testing equipment, hydraulic )rake master cylinders, wheel cylinders for drum brake systems, roors and calipers for disk brake systems, brake repair kits, air brakes, urbochargers, fuel pumps, PCV valves, small engine parts, fork lift truck (5) replacement parts, carburetor parts, and carburetor kits (CX 35B; Timberlake 3127; Schultz 3057). Echlin does not produce or sell ew or rebuil replacement carburetors (CX's 235, 534M). 3. Some of these products are manufactured by Echlin; others are ,Id by Echlin but manufactured by other firms. For example, Echlin Jes not manufacture the carburetor parts that it sells. eithpr ; Initial Decision dividually packaged or in the f-erm of.kits (Smith 3178-79). Fora period of some six to eight months in 1980 and 1981 , Echlin manufactured, more or less on an experimental basis, some pump plungers and diaphragms that were included in some carburetor kits but found its costs to be too high and phased out the venture sometime in late 1981 (Smith 3233 35).

4. Approximately 72% ofEchlin s automotive aftermarket sales are to the traditional channel of aftermarket distribution, independent warehouse distributors ("WDs ) (CX 235C). The remaining 28.2% of the aftermarket sales are to distribution center ("DCs ) affliates of the National Automotive Parts Association ("NAPA") (CX 235C). 5. All products identified with the "Echlin" trade name are sold only to NAPA DCs under the "NAPA-Echlin" name (CX 235m. Among these products are carburetor kits, which are identified as NAPA-Echlin kits and sold only to NAPA (Timberlake 3130). 6. NAPA is a program distribution group (McKenna 3904). It sells automotive parts through 79 distribution centers ("DCs ), owned by Genuine Parts (71 DCs), Quaker City Motor Parts (5 DCs), Britain Brothers (2 DCs), and NAPA-Hawaii (1 DC) (Waters 3298; McKenna 3907). A DC serves essentially the same role as does a WD in the independent aftermarket (Waters 3297).

7. The NAPA DCs distribute parts through about 6 200 jobber stores, of which about 300 are owned by NAP A member companies and about 5 900 are independently owned (McKenna 3907, 3921; Waters 3298).

8. Echlin s carburetor kits bear the dual brand name "NAPA-Echlin" (CX 238, 250; Schultz 3058). These kits are sold only to NAPA (Timberlake 3130; CX 238A). Echlin has never sold, nor tried to sell carburetor kits to any customer other than NAPA (Waters 3302). Echlin has been the sole source of carburetor kits to NAPA at least since 1970 (McKenna 3911).

9. For the fiscal year ending August 31 , 1980, Echlin s consolidated operating revenues were approximately $301.4 milion and its net income was approximately $8.8 million. As of August 31 , 1980, Echlin had total assets of approximately $237 milion (Complaint and Echlin s Answer 2; CX 231 at 16 , 18). (6) 10. Echlin is one of the leading manufacturers supplying replacement parts and supplies to the automotive aftermarket in the United States (CX 235D).

11. Because of the emphasis in recent years on fuel economy, pollution control and safety, Echlin s products "are among the fastest growing lines of parts in the replacement market" (CX 231 at 10). 12. Assuming an average carburetor kit price of $3.397 in 1979 (CX , \ .

Initial Decision 105 F. s total618), Echlin s sales of kits it assembled were (' " J' of Echlin sales (CX 466" 0" in camera). Assuming an average kit price 01'"' in fiscal 1980 (CX 619B in camera), Echlin s sales of kits it assembled in camera).were ("') of Echlin s total sales revenues (CX 466Z-5 1980, Echlin was the third largest domestic assembler and seller of carburetor kits and accounted for approximately 9.8% of all carburetor kits sold in the United States that year (CX 530A). 13. Echlin is engaged in the sale and shipment of products, including carburetor kits, throughout the United States. Echlin is engaged in commerce, and its acts and practices are in or affecting commerce within the meaning of the Clayton Act, as amended, and the Federal Trade Commission Act, as amended (Complaint and Echlin s Answer H).

B. Respondent Borg- Warner Corporation 14. Borg-Warner Corporation ("Borg-Warner ) is a corporation organized and doing business under the laws of Delaware, with its principal offce at 200 South Michigan A venue, Chicago, Ilinois. (Complaint and Borg-Warner s Answer TI 5). 15. In 1980, Borg-Warner and its consolidated subsidiaries had net sales and other revenues totallng approximately $2 689 milion and net income of approximately $126 milion. As of December 31 1980 Borg-Warner had total assets of approximately $1.9 bilion (Complaint and Borg-Warner s Answer n 6; CX 3Z-5 through Z-6). 16. Borg-Warner is a diversified manufacturing corporation. Its principal product lines include transportation equipment, chemicals and plastics, air conditioning equipment, and other industrial products and services. Borg-Warner s subsidiaries are engaged in the business of providing financial and protective services (Complaint and Borg-Warner s Answer n 7; CX 3Z-22, 4D).

17. In 1980 . the Transportation Equipment Group (" E.G. ) was Borg-Warner s largest business group in terms of sales (CX 41). E.G.'s divisions were engaged in the manufacture (7) and! or supply of automotive products to original equipment manufacturers as well as to the replacement market (Complaint and Borg-Warner s Answer TI 8). T. G. had 1980 sales of approximately $903.6 million and earnings of approximately $31.5 millon (Complaint and Borg-Warner Answer n 8; CX 3Z-13).

18. Prior to the acquisition challenged herein, all Borg-Warner divisions involved in the manufacture and! or sale of replacement automotive parts were included within T. G. (Complaint and Borg- Warner s Answer n 9).

19. The products manufactured andor supplied by Borg-Warner . Throughout. this document; refers lo incamero material that has been excised. ECHLIN MANUFACTURING CO., ET AL.

410 Initial - Decision G. in 1980 included: automatic and manual transmissions and transmission components; ignition system parts; carburetor and emission control parts; carburetor kits; clutches and clutch components; four-wheel drive units; axles; radiators; and automatic slack adjusters. These products are used on passenger cars, trucks, off highway vehicles, and on farm and marine applications (Complaint and Borg- Warner s Answer n 8; CX 4I) 20. Borg-Warner s sales and earnings by product areas for the period 1976-1980 were as follows (CX 3X):

Sales and earnings by product area (millions of dollars) 1980 1979 1978 1977 1976 Sales Air Conditioning 587. 570. 516. 439-2 352. Chemicals & Plastics 630. 595.4 472. 421. 399. Industrial Products 516. 467. 422.4 365. 343. Transportation Equipment 903. 986. 844.5 718.5 652.4 Sales of present operations 637. 619.5 255. 944.5 747. Discontinued operations 97. 70. 87.4 114. 673. 717.4 $2,326. 031. 862.4 Earnings Air Conditioning 13. 12. 19. 18. Chemicals & Plastics 28. 27. 18.5 13. Financial and Protective Services 35. 27. 18. 12. 10. Industrial Products 52. 32. 30. 26. 16. T ransport tion Equipment (591 42.4 32. 26. Earnings of present operations 123. 145. 129.4 95. 76. Discontinued operations 126. 155. 133. 104. 81. 21. ("'J (CX's 161A, 3Z-1 , 24C in camera). E.G.'s automotivf aftermarket operations performed better than its original equipmen operations in 1980 and showed an increase in sales over 1979 (C: 3Z-2), although the sales of all (8) transportation equipment, incluo ing original equipment sales, showed a substantial decline (CX 3)1 22. In 1980, the year preceding the acquisition challenged herei Borg-Warner assembled and sold approximately 4.8 milion carbu tor kits, accounting for about 38.4% of all carburetor kits sold in t United States and ranked first in that market (CX 530A). 23. Borg-Warner has been and is engaged in the sale and shipm ofautomot:ve products throughout the United States, and is enga in commerce) or its acts and practices are in or affecting comme within the meaning of the Clayton Act, as amended, and the Fed Trade Commission Act, as amended (Complaint and Borg-Waf! Answer 11 10).

Initial Dec sion 105 F. II. THE CHALLENGED ACQUISITION A. Agreement For Sale Of Assets 24. On May 12, 1981 , Echlin and Borg-Warner entered into an Agreement for Sale of Assets " effective February 28, 1981, whereby Echlin would acquire all of the assets of Borg-Warner s automotive aftermarket divisions of the Transportation Equipment Group ). In exchange, Borg-Warner received 4.5 milion shares of Echlin s common stock, representing approximately 22% of Echlin outstanding shares. The transaction was valued at about $62.4 million (Complaint and Echlin and Borg-Warner s Answers TI 11; CX's 500 at page " , 239A through Z-66).

25. The Boards of Directors of Echlin and Borg-Warner gave their final approval to the transaction and Echlin s stockholders ratified the agreement on July 7 1981 (Complaint and Echlin and Borg-Warner s Answers TI 11).

26. Subject to a hold separate agreement entered into by respondents and the Federal Trade Commission, the transaction was consummated on July 14, 1981 (Complaint and Echlin and Borg-Warner Answers TI 11).

27. The "Agreement for Sale of Assets" between Echlin and Borg- Warner also contained certain additional agreements including: (1) a upply agreement wherein Borg-Warner agreed to continue to supply chlin with various automotive parts and products which Borg-Wars manufacturing divisions formerly supplied to the acquired divions; (2) a trademark license agreement permitting Echlin to market oducts ofthe acquired divisions under the Borg-Warner name; and I an agreement to place a director designated by Borg-Warner on hlin s Board (CX's 500 at 3, 239Z-67 through Z-87, Z-119, Z-204 ough Z-208). (9) :8. During 1977, Borg-Warner s Aftermarket Task Force, an ad hoc up, compiled a study on Echlin focusing on the approaches Echlin taken in marketing its products to the aftermarket. (' " ) (CX' through Z-5, 39L in camera ). It concluded that Echlin s success based primarily on quality, warranty, customer service, and ," (CX (''' J 14N). ("' ) (CX 14C in camera). . Borg-Warner s interest in pursuing a transaction with Echlin Jased on three factors: (1) Echlin was regarded as a "good invest- ; (2) Echlin "specialized" in the automotive aftermarket busind (3) the combination ofthe aftermarket businesses of Echlin org-Warner would result in a "stronger replacement parts oper- . (CX 240C).

Borg-Warner offcial testified that the company agreed to the ed transaction, which included supply and trademark agree- ECHLIN MANUFACTURING CO., ET AL. "40 410 Initial Decision ments, because Borg-Warner was obtaining a large equity position in Echlin as well as representation on Echlin s Board of Directors (Trauscht 3894-95).

B. The Acquired Divisions 31. Through the May 12, 1981 agreement, Echlin acquired all of Borg-Warner s so-called "Automotive Aftermarket Operations " comprising five Transportation Equipment Group ("T. ) divisions. They included two manufacturing divisions (Ballwin/Washington and Ottawa Divisions) and three distribution divisions (the Automotive Parts Division "APD " APD International, and APD Borg-Warner (Canada Limited) (CX's 8A- , 3Z-12, 239N through Z-66, 500 at 2 25).

32. The Borg-Warner "automotive aftermarket operations" a quired by Echlin had 1980 sales of about (*"J including carburetor kit sales of about ( J (CX' s 3Z-12, 29X in camera). 33. The two divisions of Borg-Warner s automotive aftermarket operations involved in the production and sale of automotive carburetor kits in the United States were the Ballwin/Washington Division and APD. (1OJ 1. Ballwin/Washington Division (Ballwin/Washington) 34. In October 1966, Borg-Warner acquired Precision Automotive Components Company C'PACCO"), a company that manufactured carburetor parts and assembled automotive carburetor kits (CX 535E- F; Carlson 2525-27), and incorporated PACCO into Borg-Warner Marvel-Schebler Division (Merz 2713). Sometime during the 1970' Borg-Warner s two facilities involved in the manufacture of carburet. or parts and the assembly of carburetor kits became known as th, Ball win/Washington Division" (Merz 2718; Carlson 2572). The Ball win/Washington Division C'Ballwin/Washington ) includes Borg Warner facilities located in Ballwin and Washington, Missouri (C; 500 at 26; Timberlake 3070-71).

35. ( J (Complaint and Borg-Warner s Answer n 9; CX's 21J i camera 27C in camera 539Z-13 through Z-14). ( J (CX's 21J ; camera 24D in camera 25D in camera, 26C in camera 27C in came' 39P in camera 36. The Ballwin plant, located in Ballwin, Missouri, consists oft. buildings, with a total of 100 500 square feet. One building is devo1 to the assembly of carburetor kits and to the warehousing of products produced at both the Ballwin and the Washington pia that are ready to be shipped. This building occupies approximal 500 square feet. The remaining space of22 000 square feet com of the offce building where all the administrative functions for ( Initial Decision 105F. entire Ballwin/Washington Division are handled. Some ofthese functions include sales, finance, marketing and customer service (CX' 239Z-141, 500 at 28, 167E; Timberlake 3070-71). 37. The Washington facility, located about 35 miles from the Ballwin faciliy in Washington, Missouri, is devoted to the manufacture of carburetor parts, including those used in the assembly of carburetor kits. The facility occupies 69 300 square feet (Timberlake 3071; CX' s 239Z-138, 167C, 500 at 28).

38. (" 'J (CX' s 5C , 21J in camera 24D in camera 25D in camera 26C in camera, 27C in camera 39P in camera 129Z-8, 167B, F, 169Z-8 Z-29, 500 at 26, 535R). (l1J 39. ('''J (CX' s 1621 in camera 9Z-3).

40. (" ' J (CX 162K- in camera). J (CX' s 24E in camera 25E camera 26D , G in camera, 27F in camera 2. Automotive Parts Division (APD) 41. Prior to the transaction, APD was Borg-Warner s aftermarket distribution organization, headquartered in Franklin, Ilinois (CX 500 at 25, 28). ("' J (Complaint and Borg-Warner s Answer n 9; CX's 5A 21F in camera 33D , L in camera 34B, G- , N in camera 35H oamera 39" 0" in camera 500 at 26, 535S). 42. ("'J (CX' s 21F in camera 33D in camera 34B in camera 35B n camera). These facilities were leased and located in: (1) Pico Rivera alifornia; (2) Elkridge, Maryland; (3) Marietta, Georgia; (4) Mesquite exas; (5) Milwaukee, Oregon; (6) Philadelphia, Pennsylvania;; and (7) orfolk, Virginia (CX's 5A, 239Z-135 through Z-138, 500 at 28). 43. ("'J (CX' s 21F in camera 22E- in camera 33D, L, T in camera , V in camera 35H, N in camera 39(( 0" in camera 44. ("'J (CX' s 21F in camera 33P in camera 34Q in camera ). (12J 15. APD obtained its line of carburetor kits from the Ballwin/ Ishington Division (Merz 2715; McCurry 3837). APD was a "nationlccount" of Ballwin/Washington.

6. Immediately following the consummation of the challenged 1saction, Echlin renamed APD (now a subsidiary of Echlin) BWD ("BWD") (Merz 2707, 2747; Martin 2788-89). . Prior to the challenged acquisition, Borg-Warner (through its win/Washington and APD divisions) was the nation s leading ufacturer and supplier of replacement carburetor parts of all s to the replacement carburetor industry as well as the largest abler and seller of carburetor kits in the United States (CX' , 163Z-9, 165V, 167B, 169Z-8, Z-29 171M ..., ECHLIN MA1 Ulf\vlUHHH.

Initial Decision C. Trademark License Agreement 48. As part ofthe Agreement for Sale of Assets, Echlin obtained an exclusive license to use certain Borg-Warner trademarks in connection with the marketing and sale of various automotive replacement products, including carburetor kits, in the United States and Canada and a nonexclusive license for the rest of the world (CX 239Z-67 through Z- , 500 at 3).

49. The trademark license gives Echlin the right to use the Borg- Warner name and other Borg-Warner trademarks on aU products sold to the automotive aftermarket, including those sold through its BWD (formerly APD) and Ballwin/Washington (CX 239Z-67 through 68).

50. The trademark agreement between Borg-Warner and Echlin is etlective for a term of 15 years and continues thereafter unless terminated by either party upon written notice five years prior to termination (CX's 239Z-79, 500 at 3).

51. The licensing ofthe Borg-Warner trademarks was an important feature of the transaction. ('*') (CX 33D, H in camera). Echlin believed that the Borg-Warner trademark was "of importance to the financial condition and results of operations" of Borg-Warner s aftermarket operations and that the name was "identified with quality and dependability" and created "favorable consumer response" (CX 500 at 27).

52. The evidence shows that in the sale of carburetor kits today brand recognition or loyalty is no longer an important factor, primarily due to improved quality of aU carburetor kits, and that as a practical matter neither WDs nor jobbers, much less instaUers, are brand conscious as far as carburetor kits are concerned (e. , Nelson 1622; Jursek 826; Tehansky 861; Carlson (13) 2569; Waters 3301; Brown 3373, 3382-83; Kotcher 3558; Foley 3790; Milford 3771). D. Supply Agreement 53. As part ofthe Agreement for Sale of Assets, Echlin and Borg- Warner also entered into a supply agreement whereby Borg-Warner agreed to supply Echlin with automotive replacement products historically supplied to its aftermarket divisions sold to Echlin by the manufacturing divisions retained by Borg-Warner (CX's 239Z-204 through Z-208, 500 at 3). Under the terms ofthis supply agreement sales of automotive replacement products wil be at prices at least as favorable as those offered to other purchasers that compete with Echlin (CX's 239Z-205, 240" , 500 at 3). 54. The supply agreement also provides that Echlin s BWD wii continue as an aftermarket distributor in the United States for al) Initial Decisi 105 FTC. Borg-Warner manufactured automotive products (CX's 239Z-206, 500 at 3). Echlin agreed to purchase from Borg-Warner its requirements of those automotive replacement products which are manufactured by Borg-Warner divisions and presently sold (but not manufactured) by the divisions acquired by Echlin, provided that Borg-Warner prices remain competitive with those offered by other manufacturers and that Borg-Warner is able to meet Echlin s delivery requirements (CX' s 239Z-206 through Z-207, 500 at 3).

55. The supply agreement between Borg-Warner and Echlin is effective for an initial term of15 years and continues thereafter unless terminated by either party upon written notice five years prior to termination (CX's 239Z-208, 240P, 500 at 3). 56. The Agreement for Sale of Assets also provides for the election of one Borg-Warner representative to the Echlin Board as long as Borg-Warner owns at least 10% ofEchlin s voting securities (CX's 500 at 3, 239Z- 119, 240Q). Frank E. Pilling, a Vice President of Borg- Warner, became a director of Echlin in November 1981 (CX's 240Q, 500 at 3, 536D, 569Y).

E. Hold Separate Agreement 57. On July 7 1981, respondents and the Federal Trade Commission signed a hold separate agreement pending resolution of this proceeding. The Commission agreed not to seek to enjoin consummation of the proposed transaction. The hold separate agreement also establishes certain requirements relating to Echlin s management of its own and the acquired assets involved in the assembly and sale of carburetor kits. The agreement provides that the Borg-Warner assets wil be operated by Echlin as a viable and separate business entity and wil continue to use (14) the Borg-Warner trade name and trademarks in the sale of automotive replacement products (Agreement, File No. 811-0094 (July 7, 1981)).

III. THE PRODUCT MARKET - CARBURETOR KITS A. Carburetor Kits Their Evolution Physical Characteristics And Uses 58. Carburetor kits are used to tune-up or repair malfunctioning carburetors that do not require carburetor replacement. 59. An automotive carburetor is a mechanical apparatus designed to supply a specific mixture of vaporized gasoline and air to the com- 1ustion chamber of automotive engines and is a crucial component of :he automotive fuel system. It usually consists of a core body and hard )arts assemblies (such as needle and seat, pump plunger or dia- )hragm, economizer or power valve), fuel metering and airing de- CIILlN MAN U1"AC'lUHINtl CU. T AL.

410 Initial Decision vices, gaskets and other small parts (clips; retainers, balls, plugs and springs).

60. The typical operation of a carburetor may be described as follows. See generally, RX' s 164W-Z-3, 165" T; Smith 3250- , 3239- 42. Gasoline enters the carburetor via the fuel inlet system consisting ofa fuel bowl, inlet fitting, fuel inlet valve (needle and seat) and a float assembly. Gasoline flows through the seat unti the level of fuel in the carburetor raises the float suffciently to close off the needle and seat. As the level oftuel in the carburetor falls, the float is lowered, which separates the needlc from the seat, thus allowing gasoline to flow into the carburetor. The fuel inlet valvc (needle and seat) and the float must maintain the fuel level at the prescribed specification. The fuel then passes through a series of valves and jets, with its flow controlled by differences in air pressure, the opening and closing of mechanical valves, springs, diaphragms, pumps and check balls. During this process, both the fuel and the air pass through blend holes and the fuel-air mixture is ultimately drawn into the engine cylinder by a vacuum created by the movement ofthe piston. Added fuel for power operations is provided by various types of power valves actuated by a vacuum diaphragm and controlled by a power valve spring. 61. Carburetor "overhaul kits" (sometimes called "major repair kits ) were introduced by Carter Carburetor Co., a carburetor parts manufacturer, during the late 1940's and became widely used during the early 1950's. Overhaul kits essentially contained all carburetor parts (excluding the core body) that may be needed for a complete overhaul of a particular carburetor model and included parts used in the fuel metering and airing (15) mechanisms, passage plugs and linkage for the outside of the carburetor (Thompson 2443; Carlson 2528-29).

62. "Consolidated kits" or carburetor "tune-up" kits were introduced in or about 1949 by Precision Automotive Components Company ("P ACCO"), a carburetor components manufacturer and parts supplier to carburetor rebuilders and were designed to accommodate more than one carburetor model. Carburetor tune-up kits thus contained gaskets, float level gauge and other "high mortality rate mechanical parts which were common to several carburetor models and an instruction sheet, but excluded metering parts and passage plugs. Thus tune-up kits contained less parts than did overhaul kits were less expensive and more convenient to the automobile mechanic and gradually replaced overhaul kits during the 1950s (Thompson 2444, 2446; Carlson 2525-29; CX 169D).

63. The record shows that Berton Carlson ofP ACCO first conceived the idea of packaging carburetor parts most frequently needed for carburetor tune-ups into a kit form in or about 1949. Essentially, ..,, Initial Decision 105 F. Carlson (then in sales) and his partner and plant superintendent Ernie Nieman, hired a draftsman, collected current carburetors, took measurements of component pieces to determine tolerances and introduced a line of about 20 tune-up kits (Carlson 2525-29). 64. After 1949, P ACCO acquired carburetor component manufacturing facilities and, in addition to sellng carburetor parts to carburetor rebuilders, gradually expanded its own line of carburetor kits and when PACCO was sold to Borg-Warner in 1966, its full carburetor kit line contained some 175 kit numbers (Carlson 2526, 2530). 65. Currently, carburetor kits generally include those carburetor parts that wear most rapidly or often ("high mortality rate parts needle and seat assemblies, pump plungers, and components with diaphragms plus gaskets to replace those worn or damaged during carburetor disassembly and cleaning, and gauges for making adjustments and instruction sheets (CX's 129D, 163C). The "hard parts" in a kit, such as a needle and seat assembly, a pump plunger or pump diaphragm assembly and economizer or power valve, perform the same function they perform in a new carburetor (Thompson 297-300). Tomco kits contain a rotary disk fuel inlet valve instead of a needle and seat assembly to control the flow of fuel into the float bowl (Thompson 295 , 297; also see, CX's 126C, 3112- , 436A). 66. The instruction sheet included in each carburetor kit provides illustrated, step-by-step instructions for performing a tune-up job using that kit. Detailed specifications for the adjustments required before and after the carburetor kit is installed are also included in each instruction sheet (CX's 169T- , 431C, 129D; Sheehan 459-60). (16) 67. Normal wear on internal carburetor parts or deterioration of parts caused by gasoline additives may impair the proper functioning of a vehicle s carburetor resulting in poor engine performance and/or poor fuel effciency. In such instances, a carburetor tune-up using the parts in a kit to replace the worn parts can return the carburetor to optimal performance (Martin 2794; Hawkins 1173; CX's 202F, 75B 129D). Specific conditions which might indicate that a carburetor tune-up using a kit is necessary include hard starting, poor acceleration, rough idle, engine stalling, poor gas mileage, repeated flooding, or excessive gas leakage (CX's 328B, 202B- , 122, 165Y). 68. A carburetor tune-up job using a carburetor kit involves the following steps:

(1) the carburetor is removed from the vehicle and disassembled; (2) the core, or major body of the carburetor, and other re-usable (generally nonmoving) parts are cleaned by soaking in a cleaning 111t1nn !:nri ,.h.jp.r! to ll"'p. th ll lnh'rn9 :J,"':m(n:"' 0;,.", ,.10';''' ECHLIN MANUFACTURING CO., ET AL. 429 Initial- Decision (3) the carburetor is reassembled and adjusted using the parts contained in the carburetor kit to replace worn parts and gaskets; and (4) the carburetor is reinstalled on the vehicle using the flange gasket (the major gasket included in a carburetor kit) to insure proper sealing between the base of the carburetor and the engine, and final adjustments are made (Schultz 3020-24; Smith 3168; CX 202C-E). 69. The time required for a carburetor tune-up job using a kit varies according to the complexity ofthe specific carburetor involved. Based on the six examples of carburetor repair jobs using a kit that Ballwin/ Washington considered "typical" (see Secrest 1087-88), the labor time varies between two and three hours, with an average of2.3 hours (CX 170J-0). The labor time used in the Ballwin/Washington sales presentations are derived from the Mitchell Parts and Labor Estimating Guide a standard industry manual (e. , CS's 170I, 164H). An automotive mechanic testified that the labor times shown in this manual provided a ceiling for the amount ofJabor time he charged the vehicle owner. On occasion, he charged for less time than provided for by the standard manual (Milford 3776-77). The evidence shows that the time required for a carburetor tune-up varies from 1.5 to 3.5 hours depending on the nature and complexity of the (17) carburetor and the skil of the mechanic (Smith 3169; Milford 3376). 70- A proper carburetor tune-up and installation of a kit wil "improve mileage, give steadier, smoother performance, and allow easier starting" by restoring the carburetor performance closer to its OE specifications (CX 129D).

71. Carburetor kits used on industrial and agricultural equipment and inboard marine engines are substantially different from those used on automobiles but are often included in suppliers' line of car bu. retor kits as a convenience to their customers, especially in coastal industrial and agricultural areas where a demand for these special ized applications exists (Fife 622; Sheehan 468-69; Thompson 31C CX' s 195- , 198). (''' ) (CX' s 465A in camera 541R-S; Sheehan 46i Thompson 309-10). For example, sales of kits for these specialize applications represented less than half of 1 % of Standard' s total k sales.

72. Various types of aircraft are also equipped with carburetor However, kits for the repair of aircraft carburetors are distinct fro automotive carburetor kits and have never been included in the Ii of automotive carburetor kits (Fife 623; Sheehan 470; Thompson Waters 3314; CX 541Y).

73. Motorcycles are also equipped with carburetors but kits motorcycle applications are not currently included in any line nt;ve carburetor kits (Fife 623; Sheehan 470; Thompson 313- '*"u FEDERAL TRADE COMMISSION DECISIONS Initial Decision 105 F. Waters 3314). Borg-Warner s Ballwin/Washington Division offered motorcycle carburetor kits unti 1976 but they were not included in Borg-Warner s line of automotive carburetor kits and were discontinued (CX 541 W-X).

74. ("'J (Thompson 313-14; Sheehan 469-70; Fife 622-23; CX' 471B, 468E in camera ). Small engine carburetor kits are generally sold through specialized distributors and repair outlets. Although Borg-Warner and Echlin both offered small engine carburetor. kits, those kits were sold under a different brand name and by a different corporate division than their respective automotive carburetor kit lines (CX's 534T- , 238A, 538W- , 541U-V). 75. During the 1980' , a throttle body fuel injection system ("TEl" was introduced in several vehicle models in order to deliver higher air-to-fuel ratio mixture necessary to insure higher fuel effciency and lower air pollution. Simply put, a TBI system electronically controls the carburetion process by the use of a high pressure fuel pump, with a pressure regulator, (18) which forces the fuel through an injection nozzle. A TBI system is an electronically controlled fuel metering system, and as such, it does not include the mechanical parts found in a conventional carburetor, such as seat and needle, float, pump plunger or piston, accelerator pump and economizer valve. Thus, a rel is substantially different physically from a carburetor, although hey both perform the carburetion function in a vehicle s fuel system Fife 699-700; Smith 3256-59).

76. In response to the introduction of the TBI system on a number f late model automobiles, Borg-Warner introduced TBI repair kits nd included them in its July 1980 Supplement to Automotive and arburetor Tune-Up Kit Catalog (CX 40 at 12; see CX' s 189A, F !9Z-3 through Z-, 537Z-27). TEl repair kits are or will be offered , most major carburetor kit suppliers as a demand therefor materilzes (Fife 698-700; Smith 3284; Thompson 381; Jursek 797-98; CX' lA, 189A, F).

77. The evidence indicates the conventional carburetor and the TBI 'I both be used in new automobiles for several years, that carbures wil continue to play the dominant role through the 1980's until TEl is cost-refined to be competitive or give superior mileage ond their present capabilities, and that this evolution wil happen m the demand for conventional carburetors wil diminish (Fife 99; Smith 3281-83; CX's 129Z-3 through Z- 163Z-). B. The Production Of Carburetor Kits . The carburetor kit production process is essentially an assembly ation and does not encompass the fabrication or manufacture of "retor components that are included in carburetor kits. ECHLIN MANUFACTURING CO.. ET AL. 401 410 Initial Decision 79. The physical carburetor kit assembly operation is a simple -arid largely manual process and is typically performed on short, moving packaging lines or at rotating tables, generally by semi-skilled female employees. When a line is used, it is first "set up" with bins or boxes each containing different parts that had been pulled from inventory using the "bil of materials" (F. 138 infra) for the specific kit being assembled. Many assembly operation begins with a "Kliklok" machine which forms a cardboard tray to hold the different parts that go into the kit. As the trays move down the conveyer belt, workers place parts (or plastic bags containing several small parts) (jut of the bins into each tray. At the end of the line, the trays are wrapped with plastic, heat-sealed and placed into shelf cartons. 80. The number of lines operated varies from one to four. (" * (Hawkins 1179; Thompson 337- , 2499; Fife 647; Sheehan 574; CX' 533K, 542Z). (19) 81. When only a small quantity of a given kit is needed, the assembly operation is performed at a rotating table with a large lazy-Susanlike tray at the center bearing different parts. Workers sit around the table and place the needed parts into a bag or tray (Timberlake 3116 3136).

82. In mid-October 1983, the administrative law judge had occasion to view both the line and table type carburetor kit assembly operations ofTomco at its St. Louis plant and was impressed by the simplicity of the process which appeared neither elaborate nor complex (see RX' s 398--04 (photographs showing a packaging line at Echlin s Ballwin/Washington plant)).

C. Substitute Products And Price Sensitivity Replacement Carburetors And Carburetor Kit Components 83. The parties agree that new and rebuilt replacement carburetors are functional substitutes of carburetor kits in the sense that any carburetor malfunction can be resolved by replacing the carburetor rather than by repairing it. However, the two product groups are not true substitutes for each other to the extent that a carburetor failure or malfunction which is due to structural damages to the carburetor casing or core, or to worn linkage or throttle shaft, cannot be "repaired" by installng a kit and requires replacement of the entire carburetor (Thompson 2494-95; Martin 2795; Nelson 1400-04; CX' 328B, 202G; F. 88-91 infra 84. New and rebuilt replacement carburetors are substantially more expensive than carburetor kits. Common sense and daily experience show that automobile mechanics and vehicle owners wil not use replacement carburetors for carburetor malfunctions which can be resolved by a carburetor tune-up using a kit. In this sense, replace- ; ..

Initial Decision 105 FTC. ment carburetors do not offer an economically reasonable alternative to carburetor kits and are not practical substitutes for carburetor kits. This holds true when the total cost to the consumer, including labor cost, is considered (F. 92- infra).

85. The record shows that replacement carburetors and carburetor kits are intended and employed for different uses, that there is no price sensitivity between the two product groups, and that ..replacement carburetors do not offer an economically reasonable alternative to carburetor kits. Therefore, replacement carburetors are not closeenough substitutes to be included in the same market with carburetor kits.

86. The parties also agree that individually packaged carburetor parts which are generally included in carburetor kits (20) today can be used interchangeably with kits. However, the record is clear that the availability of each ofthe numerous kit parts, individually packaged or in bulk form, to the automobile mechanic or vehicle owner is rather limited and that, in any event, the sales of such individual parts are de minimis (F. 108- infra). This comports with reason and common sense. The raison d'etre of kits is economy and convenience. Thus, individual carburetor parts included in carburetor kits either separately or collectively, are not practical substitutes for carburetor kits. Therefore, they are not close-enough substitutes to be included in the same product market with carburetor kits. 87. However, the record also shows that replacement carburetors carburetor kits and kit parts are competitive products in a broad sense and this important fact wil be duly taken into account in evaluating the competitive effects of the challenged acquisition. 88. A carburetor tune-up using a carburetor kit is appropriate when cleaning and installing the parts included in a kit can return a carburetor to optimal performance. In contrast, when wear or damage is so extensive that it cannot be repaired by using the parts included in a kit, it is necessary to replace the carburetor with a new or rebuilt carburetor. Conditions which require replacement include a cracked or warped carburetor casing, a broken core, worn linkage or a worn throtte shaft (Martin 2794; Thompson 2494-95; Nelson 1400-04; CX' 328B, 202G).

89. Documents generated by both respondents reflect their clear understanding of the different uses for which carburetor kits and replacement carburetors are intended. For example, counter literature prepared by Borg-Warner to explain the use of carburetor kits to vehicle owners states as follows: "If your mechanic has decided not to overhaul your carburetor, it is probably because it was missing parts or crash-damaged-in which case the necessary parts would not be in a standard tune-UD kit or rp.n::ir nft,hp (' hllrpt,",, ;,, ;-no.rlu hl", UclLIN MANUFACTURING CO. . ET AL. 433 410 Initial Decision In this case, we advise yo follo your mechanic s recommendation for a new or rebuilt carburetor-because he is in the best position to judge the quality of rebuilt carbs available in your area" (Nelson 1403-04; CX 202G). A similar document prepared by Echlin explains the different purposes and uses for carburetor kits and replacement carburetors: "Carburetors should be replaced whenever there is wear or damage to parts that will not be replaced in overhauling (i. repair with a kit) (Nelson 1400-02; CX 328B). 90. Mr. Milford, an automobile mechanic called as. a witness by respondents, testified that he always used a carburetor kit to perform a carburetor repair so long as repair with the kit was suffcient to remedy the problem and the carburetor kit was available, which it virtually always was (Milford 3777). As a general rule, Mr. Milford would replace the carburetor when repair with a kit was not possible (Milford 3777-78). (21) 91. While either a new or a rebuilt replacement carburetor may be used when repair with a kit is impossible, some aftermarket suppliers and mechanics prefer the new to the rebuilt because of quality considerations (Milford 3780-82; Foley 3801-02). Echlin also discouraged the use of rebuilt carburetors because oftheir poor quality (CX 328A). 92. A replacement carburetor can always be used when repair with a carburetor kit would be suffcient. However, for functional substitutes to be included in the same product market, it is essential that they be economically reasonable alternatives (Glassman 4370-71; Nelson 1399-400). The large price differentials prevailing at every level of distribution, including prices charged the ultimate consumer (the vehicle owner), between replacement of the carburetor and repair using a kit suggest that these products are not economically reasonable alternatives. In letters written to its carburetor kit customers BalJwin/Washington showed ilustrations of the large price differentials involved in the use of a kit as compared to replacement of the entire carburetor. In 1976, the vehicle owner s total cost, including labor, for replacement of a "typical" carburetor was about $110 if a rebuilt carburetor was used, and about $170 if a new replacement carburetor was installed. Barring a structural damage or worn linkage or shaft, that carburetor could be "tuned-up" using a kit for only $60 (CX 128C). In 1980, these costs had increased to about $154, $206 and $65, respectively (CX 128C). New replacement carburetors for larger and more complicated engines cost as much as $400 to $600 by early 1982 (CX 180A). As a Ballwin/Washington offcial stated: "Consumers with carburetor malfunction cannot and wil not choose new carburetors as a solution to their problems" (CX 180A). 93. Sales presentations prepared by Ballwin/Washington for its national account (private label) kit customers also include cost comg., Initial Decision 105 F. parisons between using a replacement carburetor and a kit (see CX' 129E- , 163D- , 164I- , 165D-K, 166J-Q, 168H-0, 170H-0). In preparing these sales presentations, Ballwin/Washington selected typical, popular applications for various types of carburetors (Secrest 1087-88). An analysis of the costs to the vehicle owner reflected in these documents reveals that the cost of repair with a carburetor kit . is significantly lower than replacement of the carburetor (see Nelson 140fH7; CX's 165D- , 170H-0, 128C, 129Z-12). For example in a 1981 sales presentation, Ballwin/Washington compared the costs of repair with a kit versus replacement of the carburetor for six common applications (CX 164I-N). While the total cost to the customer for repair with a carburetor kit averaged $91. , the customer cost to have the carburetor replaced averaged $162. 14 for replacement with a rebuilt carburetor, and $198.77 with a new replacement carburetor (CX 164I-N). The average list price that the customer would be charged for the carburetor kit itself was $19. 12 (CX 164I-N). For these applications, the price of the (22) carburetor kit would have to increase between 280 and 700% before the cost of repair to the consumer would equal that ofrepJacement with a rebuilt carburetor (Nelson 1420). The comparison to new replacement carburetors is more striking: the price of the carburetor kit would have to increase 415 to 000% before the repair bill to the consumer would equal replacement with a new carburetor (Nelson 1421) 94. Mr. Milford provided additional examples from his business experience ofthe price differential to his customer (the vehicle owner) between repair with a carburetor kit and replacement of the carburetor. For a standard carburetor model that has been available for many years, replacement of the carburetor would cost the vehicle owner between $156 and $176. Repair with a carburetor kit for the same application would cost $60-$65 (Milford 3783-84). The price differential on newer vehicles is much larger. On certain GM Citations, for example, replacement ofthe carburetor would cost the vehicle owner between $620 and $760, while tuning-up the carburetor with the use of a kit would cost the consumer $130-$140 (Milford 3778-79). 95. The disparity between the profit potential to the mechanic of repair versus replacement is also large (Brown 3409-10; Tehansky 862; CX's 121, 1292-12). For example, a 1981 sales presentation by Ballwin-Washington reveals that an installer retained an average of 88.2% of each repair dollar when a carburetor kit was used as compared to 37.9% or 39.8%, respectively, for replacement with a new or rebuilt carburetor (CX 168H- I; see also CX' s 164H, 170H- , 166J- 128C, G).

96. In early 1984, the average price a mechanic paid a jobber for a .J ." onnmvimAtplv $20.00 (Milford g., .

L..u . U H''-'' '.L''''''''U''' VV., -'.L .0... 410 Initial Decision 3779-80). Even a 10% increase ih that Price would not have any effect on the mechanic s decision whether to use a carburetor kit or replace the carburetor since the price differential between these products would remain large (Milford 3780).

97. Mr. Carlson, a witness called by respondents, whose company sells private label carburetor kits produced by Ballwin/Washington agreed that a 5% increase in the prices his company charged for carburetor kits "would not affect our business in any way" and would not cause a loss in sales due to any shift in demand towards replacement carburetors (Carlson 2645-46).

98. Carburetor repair using a kit has always been significantly less expensive than replacement ofthe carburetor (Carlson 2639; CX 537Z -f8). The price differential between repair using a kit and replacement of the carburetor has widened in recent years, making replacement in lieu of repair even more uneconomical to the vehicle owner (Merz 2776, 2780; Thompson 419; Hawkins 1200; CX 128C). (23) 99. Thus, the record evidence shows that kit assemblers can raise kit prices by 5% without causing a substantial shift towards replacement carburetor and suggests that these two products do not belong in the same product market. See U.S. Department of Justice Merger Guidelines Section 2. 11 (June 14 , 1984); Statement of Federal Trade Commission Concerning Horizontal Mergers at 12 (June 14, 1982). 100. Respondents argue that an increase in the price of replacement carburetors has caused an increase in the demand for kits and that therefore, the opposite must be true (see RPFs 107-18). However respondents' evidence linking the increase in replacement carburetor prices to an increased demand for carburetor kits is un persuasive. 101. Respondents' own documents list some half a dozen factors that might be responsible for any increase in the demand for carburetor kits (e. CX' s 164D, 166F, 168D, 169E; see also Fife 695-96). It would be diffcult, if not impossible, to determine the significance of anyone particular factor. Moreover, the Hunter Service Job Analysis C'SJA" ) statistics relied on by respondents' experts in asserting that increased carburetor prices have resulted in an increased demand for carburetor kits (RPFs 113-16) arguably demonstrate just the opposite.

102. Between 1978 and 1981, during the period when testimony by respondents' own employees and evidence from respondents ' documents show that the prices of replacement carburetors increased the most (see, Merz 2776; CX's 537Z-66 through Z-8, 128C, 168D; compare CX 129F- , which contains 1979 replacement carburetor prices, to CX 168I- , which contains August 1981 replacement carburetor prices), the SJA statistics show a decrease in the demand for carburetor kits as compared to the demand for replacement carburet- ), g., Initial Decision 105 F. ors (Wilig 4922-23; CX 567). Furthermore, the data base for RXs 407-08 relied on by respondents is so mixed that their reliability for the proposition that there is a statistically significant relationship between increased prices for replacement carburetors and increased demands for kits is questionable (see Glassman 4397-472; Wilig 4917- 26).

103. Although there is evidence which suggests that extremely large increases in the prices of replacement carburetors (see, Hawkins 1200; CX's 128C, 180A) caused some increase in the demand for carburetor kits, this evidence would not support the proposition that a 5% increase in the prices of carburetor kits, which would amount to an increase of approximately $1.00 (see F. 93-94), would bring about a significant increase in the demand for replacement carburetors. Indeed, the weight of record evidence is to the contrary. 104. On the other hand, the record is clear that the prices of carburetor kits are not raised or lowered in response to a (24) movement in the prices of replacement carburetors (Sheehan 559-60; Hawkins 1198-99; Jursek 787; Thompson 378; Fife 675-76; Carlson 2608-09; Baumann 1028-29). Conversely, movements in the prices of carburetor kits have no effect on the prices of replacement carburetors (Cardone 3624; Sheehan 560; Hawkins 1199; Baumann 1029). 105. The primary factors considered by carburetor kit assemblers in establishing the prices of carburetor kits are: (1) cost; (2) desired profi margin; and (3) competitors' prices for comparable kits (Carlson 2581; Eaton 2942-43; Fife 674-75; Hawkins 1193; Jursek 786; Secrest 1112; Smith 3230; Thompson 377; CX's 533Q, 535Z-39 througb Z-40 539Z-31, 80). Also, respondents analyzed and compared prices and return on investment only with reference to other lines of carburetor kits (Carlson 2582; Secrest 1105; Sheehan 557; Timberlake 3093-94; CX' s 80, 107-08, 113, 129Z, 209A, 211, 570C, 541Z-39). Testimony relied on by respondents in RPFs 130-0 primarily reflects kit suppliers' recognition that carburetor kits and replacement carburetors are competitive products and, to the extent it can be said to suggest that replacement carburetor prices are among the determining factors for kit prices, it is contrary to the weight of record evidence and, in any event, is insuffcient to establish price sensitivity between kits and replacement carburetors.

106. Price sensitivity is determined by the presence or absence of price changes in one product in response to movements in the prices of another product. See, Avnet, Inc. v. FTC, 511 F.2d 70, 77 (7th Cir. cert. denied, 423 U.S. 833 (1975). While carburetor kit assemblers may be generally aware of the prices of replacement carburetors, the evidence clearly shows that there is no price sensitivity hatuT""o.. "'o-rhl1'ta +..r 1-';+-", ".. .. -ranl",.",rnp,nt ,-arhl1r",tr'\' C' P..i,.""", ofl"o:r- \. , n.llll lVl.fl Uli\CTU.tlNtJ CO., ET AL. 437 410 Initial Decision buretor kits are not raised orIowered inresponse to a movement in the prices of replacement carburetors or vice-versa. 107. There is also some evidence tending to show that replacement carburetors are viewed as a distinct product market. Neither Holley nor Carter considered their respective sales of replacement carburetors as in any way affecting their sales of carburetor kits (Jursek 785; Sheehan 559). Similarly, l"') (Thompson 379). (''' J (Nelson 1379-82; CX' s 24C in camera 25C in camera l40C, 162Z-26 in camera 169Z-17 through Z-18, 20Z-2 through Z- , Z-18 through Z- 19 in camera 136A, 331AJ.

108. The record evidence regarding the availabilty of carburetor parts generally included in kits as individually packaged parts is somewhat mixed. It appears that Echlin offered in late 1970's a number of kit components, including some diaphragms, needle and seat assemblers, power valves, pump (25) plungers and some small parts to NAPA (CX's 249, 250 at 139- , 145, 148 254 at 31-35). There are other kit suppliers which offer some individual kit components for sale (Thompson 269, 2471; Martin 2804-05; Timberlake 3138). Some kit suppliers which once offered kit components for sale no longer do so (Sheehan 495, 516-18; Fife 616-17). After the challenged acquisition, both Ballwin/Washington and APD began to sell individual parts (Timberlake 3125- , 3138; Eaton 2981-82; Martin 2804--5). The record also shows service guides containing instructions for the use of parts for various makes of carburetors are available (RX' 162-65). The record as a whole is clear that at the time of the challenged acquisition, the availability of each of the parts included in kits, either individually or together, was rather limited (see CPF 188- 89).

109. In any event, there is no dispute about the fact that sales of carburetor parts generally included in kits as individually packaged parts are de minimis (Glassman 4491; Fife 672-73; Thompson 369; Castagna 3751-52; CX's 148 , 57 DB). This fact underscores the raison etre of kits, namely, economy and convenience. Since a kit includes those parts of a carburetor most subject to wear or deteriorate, when one of the parts needs to be replaced, it is likely that the other parts included in a kit also need to be replaced. And the cost of handling, inventorying and biling the thousands of different parts included in a line of kits is prohibitive at the WD, jobber and installer levels of aftermarket distribution (Martin 2793; Fife 673; Insalaco 948-49; Eaton 2982-83; Sheehan 495, 516-18, Castagna 3752; Nelson 1391- 94).

110. The record is also clear that carburetor parts and emission control parts that are not included in kits, such as floats, dash pots and choke pull-offs, are not functional substitutes for kits and do not . .

Initial Decision 105 F. belong in the same product market with carburetor kits (Fife 674; Sheehan 523-24; Insalaco 949-50; Glassman 4396; CX 106B). 111. From the foregoing findings, it is determined that a preponderance of credible evidence in the record as a whole supports complaint counsel' s position that the assembly and sale of carburetor kits constitutes an appropriate product market for the purposes of this case. IV. THE GEOGRAPHIC MARKET 112. The parties agree that the relevant geographic market in which to assess the effect of the challenged acquisition is the United States as a whole (Complaint and Answer TI 12). (26) V. THE NATURE OF COMPETITION IN THE CARBURETOR KIT MARKET A. Sourcing 113. The term t'sourcing" in the automotive aftermarket means purchasing various components or other sundry parts used in producing or assembling a finished product, including replacement carburetors and carburetor kits. The term "make-or-buy decision" is related to !!sourcing" and means a decision either to make or to buy from aD outside supplier a needed component part.

114. A make-or-buy decision is based on such factors as the comparative costs of manufacturing versus purchasing, the anticipated volume of usage of a given part, in-house manufacturing capabilities and the availabilty of outside suppliers (Schultz 3045; Fife 643; Sheehan 494; Thompson 321; CX's 331A, 537Z-46, Z-9). An initial "buy decision regarding a part may later be changed to "make" decision if there is an increase in usage of the part beyond the original projection (Fife 643-44; CX 533Z).

115. If a buy decision is made, suppliers must be located. It is generally preferable to have two sources for a part or component of a part to assure better quality, price and availability (Schultz 3043; Timberlake 3081-82; CX's 538Z-66, 539Z-9). The criteria in selecting a parts or component supplier include price, quality and reliability (Bush 2908-09, 2868; Thompson 328-32; CX's 537Z-38, 539Z-8). In soliciting bids, the assembler sometimes provides design and material specifications to potential suppliers (Bush 2868, 2909- , 2912; Timberlake 3082). (" ' J (Bush 2909; Thompson 403-04; Fife 646; CX' 544V- Y in camera 457B). In some cases, parts are rejected because the configuration or material is unacceptable (CX's 378B- , 492A). 116. ("'J (Thompson 331; CX's 497A- , 366-67, 377B- , 372B 382A, 544Z-63 through Z-96 in camera 56A- , 201). Assemblers may not be able to take full advantage of available volume discounts on the (',1 - 1__-- 1_.__ l".. r.. rr.; ",:,,'Y lrit tn., ('()"t, ... ,.. .... ... ,..... 410 Initial Decision of increased parts inventory, ' and the required inventory space (Thompson 334; Fife 646-47).

117. Whether made or bought, all parts undergo visual or other tests on a sampling basis before they are used in kit assembly (Thompson 333, 405-07; Sheehan 495; Timberlake 3083-84; CX's 539Z-20 through Z-21, 384B- , 412A-Q, 408A-K, 410A- , 167m. (27) 118. Most carburetor kit assemblers purchase some kits from other kit assemblers. Kits most often purchased are generally those needed to fill out a line or those with lower demand and include kits for foreign car applications, tractor or industrial applications ('Ihompson 335-36; Sheehan 473; CX 538Z-2 through Z-3). Borg-Warner did not purchase any carburetor kits for resale (CX 16). Although selfsuffciency is desired by all kit assemblers (Jursek 729; Sheehan 527- 28), the number and availabilty of new parts required for a kit are often important factors in making a buy-or-assemble decision (Fife 626; J ursek 728).

119. In-house production of parts has important advantages over purchases. Most importantly, necessary parts may be unavailable or in short supply from outside suppliers (Fife 644 , 669; Nelson 1561; see Jursek 728-29). For example, Echlin s preacquisition decision to begin manufacturing certain pump plungers and pump diaphragms for inby the fact that noclusion in its carburetor kit was necessitated supplier would manufacture these parts with the type of material Echlin specified (CX 331A-B; also see CX 456). (''' J (Nelson 1562-65; Jursek 729; CX's 455 , 26I in camera; see CX' s 33T in camera, 232 at 233 at 4 , 20E in camera ). In a pre-acquisition letter to Sherman, one of its leading parts suppliers, Echlin exhibited a keen awareness of these advantages. After stating that it was wiling to continue to purchase parts from Sherman as long as Sherman was supplying top quality products, meeting its delivery performance and pricing the parts in a way that enabled Echlin to meet its profit goals, Echlin went on to state: "If we deviate from this sound business principle then Echlin would have to tool products that it would prefer to buy from Sherman" (CX 455; see Nelson 1563).

120. That dependence on outside suppliers for carburetor kit components can create a serious problem for an assembler was driven home to Echlin in 1979 when, because of problems with its parts suppliers it faced a serious "back order" problem with a number of carburetor kits and feared the loss of its NAPA carburetor kit business (Echlin premier account) to a competitor (CX 456). 121. The evidence shows that aU ofthe leading assembler-sellers of carburetor kits, with the notable exception ofEchlin, were integrated backward to a significant degree and fabricate ("make ), either inhouse or through contracts, many ofthe more important components .j .. .. .. , 440 FEDERAL TRADE COJyMISSION DECISIONS Initial Decision 105 F. used in carburetor kit assembly (Borg-Warner, F. 35- supra; Standard, F. 178 infra; Echlin, F. 3 supra; , F. 182 infra; ACF, F. 184 infra; Tomco, F. 186 infra; Holley, F. 188 infra ). The evidence also shows that carburetor parts production ofGM and Carter were limited to parts with application on carburetors of their own make and that Borg-Warner and Sherman Carburetor Company were the two primary (28) sources of carburetor and kit components for the kit assemblers. However, the record shows that there are a large number of suppliers for the various components that make up carburetor kits (e. , Bush 2861-921; Thompson 326- , 2454-55; Sherman 585; Jursek 732- , 811- , 823-26; InsaJaco 910-14, 983-84; Lehman 5046; Schultz 3040-1, 3043; Timberlake 3081, 3096; Smith 3179-80; RX 276; CX 535).

122. (' " ) (CX 39P in camera 123. In sales presentations to its carburetor kit "national accounts (private brand sellers) customers, Borg-Warner emphasized its carburetor components manufacturing capabilities as an important element and touted itself as a "basic manufacturer of carburetor components of all makes " (e. , CX's 129Z-8, 169Z-29, 163Z-9, 190A; also see, CX's 145A, 127E).

124. Prior to the challenged acquisition, Echlin was the only leading assembler of carburetor kits which did not make any carburetor parts it sold or used in kit assembly, except on an experimental basis during 1980 and 1981 (Schultz 3057; Timberlake 3127; Smith 3178- , 3233- 35; CX 235B). The importance of carburetor parts or carburetor kit components manufacturing capabilities (via backward integration) was driven home to Echlin more than once through problems with some parts suppliers (see F. 120 supra 125. By reason ofthe July 14 1981 consummation of its acquisition of Borg-Warner aftermarket divisions (including Ballwin/Washington), Echlin has become a leading supplier of carburetor parts of all makes and carburetor kits as well as the largest assembler of carburetor kits in the United States.

B. The Development And Maintenance Or Carburetor Kit Lines Consolidation And Full Line 126. The automotive repair industry is a service industry and time is an important factor in the success of an automotive service station or mechanic. Therefore, it is essential throughout the channels of distribution in the aftermarket that replacement parts be readily available to service the vast majority of vehicles on the road (Fife 622 625; CX 233 at 8). (" ' J (CX 33S in camera). (29) 127. In 1981, there were well over 15 000 distinct automotive carbu- 1", T.. ",....n... t-", ..nnt- +-h"" nnt.a..tio:lhr r11Unit"I;U). C!PTno;n rl frrr ECHLIN MANUFACTURING CO., ET AL. 441 410 Initial-Decision specific carburetor tune-up applications, Borg-Warner s Ballwinl Washington Division offered a line of 429 carburetor kits which could service 15 099 different carburetor models; the Carter Automotive Division of ACF Industries, Inc. ("Carter ) offered a line of 526 kits to service 12 000 carburetor models; Echlin offered 623 kits to service 13,098 carburetors; General Motors Corporation ("GM") (under the AC-Delco name) offered 533 kits for 12 923 carburetors; and Standard Motor Products, Inc. ("Standard") (under the Hygrade name) offered 462 kits for 11 535 carburetors (CX's 168P, 170P). 128. Because of the wholesale (warehouse) distributors' demand for broad coverage, a kit supplier must be able to oiler a full line of carburetor kits in order to compete effectively (Fife 625; Sheehan 471-72; Hawkins 1177-78; Tehansky 853; Eaton 2978; Thompson 314; Wilis 3450; Lingle 3516-17, 3527; CX's 541Z-23, 129Z-12, 546B). A supplier offering a "short-line," comprised of only the most popular kits cannot compete effectively against a full-line seller (Eaton 2978; Sheehan 471, 582; CX's 535Z-7, 541Z-23, 546B). For this reason, most kit assemblers, with the exception of Borg-Warner, purchase various amounts of kits (mostly from Borg-Warner) in order to fill out their lines (see F. 150-51 , 176 infra). The coverage or breadth of carburetor applications included in each supplier s kit line is stressed in trade journal advertising and other sales material and is an important aspect of competition among suppliers (Jursek 780; CX's 120- 123C, 125 at 8, 126B, 138B, 129N, 236). For example, Borg-Warner advertised that its kit line provided the "(mJost recent and broadest vehicle application coverage available" (CX's 120, 169IJ. 129. Tune-up kits used to repair automobile and light truck carburetors account for the vast majority of kits being offered today (Nelson 1346; Fife 621-22).

130. A supplier of carburetor kits must initially determine which kits to include in order to provide the broad coverage required in the aftermarket. Since new carburetors are introduced each year, kit assemblers must review their kit line annually (Fife 638-40; Hawkim 1181; CX 537Z-0 through Z-41). The number of new kits that mus' be added each year has increased dramatically in recent years i) response to the proliferation of new carburetor models and the ir creased demand for carburetor kits with application on foreign veh cles (Hawkins 1180-81; Thompson 301--03; Fife 624; Sheehan 48 Nelson 1557-59). For example, in 1977, Standard' s kit line provid, coverage for 6,540 different carburetors whereas in 1981, Standarc kit line covered 11 535 distinct carburetors (Compare CX 165P wi CX 166R; Nelson 1558-59). A certain number of kits are also drop! from the line annually when decreased demand, consolidation w other kits or replacement by a kit containing (30) slightly differenl g., Initial- Decision 105 F. improved parts occurs (Fife 640; Hawkins 1184; Thompson 320; Schultz 3036; Sheehan 458-59).

131. The annual review ofthe line genera1Jy entails an analysis of: (1) all new carburetor models for the current year; (2) the potential demand for a carburetor kit for a given application, which includes consideration of projections of new vehicle sales; and (3) the applications included in competitive kit lines (Fife 626-27; Sheehan 474-75; Smith 3174; CX 539Z). ('**j (Thompson 2453; Rivet 2824-26; Schultz 3035; Smith 3176; CX's 570C, 215B, 162Z-27 through Z-28 in camera 115B, 538Z-8 through Z-50).

132. The degree of consolidation is another way in which the various kit lines may be promoted (e. CX' s 84A, 546B). Consolidation is accomplished by including in a single kit more parts than are needed for the repair of one carburetor so that the kit can be used to repair a number of similar models of carburetors. By consolidation, the kit assembler can reduce the number of different kits included in its line without reducing the coverage offered (Hawkins 1181; Thompson 302; CX' s 537U- , 538Z-4, 539Z-2, 105C, 129Z-11). On the other hand, kit assemblers must avoid over-consolidation which leads to unacceptably high costs per kit and requires the kit user to discard too many unused parts (Rivet 2833-34; Fife 636-38; Jursek 737; CX 216G-F). 133. Kit assemblers review annual1Jy the consolidation of the kits which they se1J (Rivet 2833; Hawkins 1182; Sheehan 488; Thompson 319, 2479-80; CX's 533S, 539Z-2). The focus of this review is a comoarison by the assembler ofthe consolidation of its kit line with that ,f its competitors (CX 541Z-12 through Z-13). 134. The degree of consolidation of carburetor kit lines is a major eJlng tool which is emphasized in advertising, promotional activities nd sales calls (Tehansky 850-51; Secrest 1092-94; Fife 637; CX's 69 15A, 116A, 127G). For example, Ballwin/Washington includes, in its Itional account sales presentations, detailed consolidation com parins of its kit line with that of other kit suppliers, including an analis of inventory cost, inventory turnover and return on investment s 163N through Z-l; 168Q through Z- , 170Q through Z- , 129 Z; see Secrest 1089-94).

35. As an integral part of the decision as to which kits to include its line, an assembler also must decide which specific parts to ude in each kit (CX 539W-X; see also Thompson 316-17, 397-98; s 537Z-40 through Z-41 , 5381). (31) 36. Changes in the material used in existing model carburetors require a review and a determination of the parts to include in ;ing kits (Sheehan 458-59, 476-77; Hawkins 1185). Carburetor kit 11blers also emphasize the development of new materials and ns for the carburetor parts included in their kits as a way to gain ( , ECIILIN MANUFACTURING CU., 1'1 A.L.

410 Initial Decision m8rket advantage (CX's 128D, 162Z- , 98P, S, T, X, 99U, X 101B , 105D, F, N, Q, S, 129Z-5 through Z- , 187G-H). In addition, the quality of those parts included in kits is emphasized in promotional activities (CX's 98-105, 426 at 86, 431B). 137. The instruction sheet contained in every carburetor kit is prepared from service manuals available from vehicle and carburetor manufacturers. The leading kit assemblers, however, design some of their instruction sheets and Borg-Warner and Carter obtained copyrights for their instruction sheets (Thompson 339; Sheehan 459-63, 579-80; Fife 648; Hawkins 1188; Rivet 2834-35; CX's 126C, 169T 218A, 434, 431C, 537Z-51 through Z-53).

138. ('''J (Rivet 2851; CX' s 402- , 220, 421A- in camera). These bils of materials are considered proprietary by the kit assemblers (Sheehan 481-83; CX 220).

139. The kit coverage offered for carburetors on domestic vehicles is generally broader than that for import vehicle carburetors (i. , 95 to 100% coverage for domestic vehicles and 75 to 95% coverage for foreign applications) (Fife 624-25; Jursek 716-17; Sheehan 463-64; CX' s 126B, 167B, 538Z-9 through Z-10). Carburetor kits are being produced for carburetors first installed as original equipment 20 to 30 years ago as well as for more recent carburetor models (Hawkins 1177; Jursek 716-17; Thompson 308; CX 538Z-9 through Z-10). Although kits can be obtaioed for the current model year s carburetors through vehicle dealers (Sheehan 464-65; Baumann 1012-13; Hawkins 1180; Thompson 398), there is a "lag time" of about 12 months between the introduction of a new carburetor on a new model vehicle and inclusion of a corresponding kit in a kit supplier s line (Merz 2723-24; Rivet 2824; CX's 541Z, 538Z-10).

140. ("'J (Fife 640-41; CX' s 30U in camera 24D in camera). l"' (CX 1691) ("' J (CX 21J in camera; see CX' s 24D in camera 25D 25D in cam-camera, 26C in camera). J (32J (CX' s 24D in camera era, 26C in camera C. Marketing And Distribution Of Carburetor Kits CX 15N in camera). The same holds 141. l''' J (Nelson 1424-26; with respect to private brand carburetor kits sold by the so-called resellers " which do not package kits they sell. 142. The vast majority of carburetor kits are sold through the traditional warehouse dealer/jobber channel (Fife 655; Jursek 778; Shee. han 537-39; CX's 570A, 535Z-17 through Z- , 26G). (''' J (Fife 655 Tehansky 841, Bull 3648-50; CX 14Z-3 through Z- in camera 404DJ 143. There are C"' ) to l"' J WDs throughout the United State: which warehouse numerous aftermarket product lines in order t, assure quick delivery of specific replacement parts ordered by their ( ( ( Initial Decision 105 F. jobber customers (Tehansky 836; Merz 2785; Bull 3693; Castagna 3754 55; CX's 33L , V in camera 34N in camera ). They may carry in stock J to (''' J different replacement parts, including a full line of(''' to ("' J carburetor kits (Bull 3642; Brown 3369; Waters 3314; Kotcher 3571; Castagna 3759; CX 15J , X in camera). 144. About ("'J jobbers stock automotive replacement parts for sale to over (" ' 1 installers (CX 33L in camera ). Jobbers generally stock 95 to 120 different carburetor kits and rely on their WDs for other kits not in stock (Bull 3654, 3692-93). Jobbers expect the WDs to carry a full line of carburetor kits at all times (Foley 3799) and try to give immediate delivery of parts to installers since installers do not generally inventory replacement parts (Tehansky 842-43). 145. Many WDs and jobbers are members of program I" ' J distribution groups. (CX 33Q in camera j. Program distribution groups may be "controlled" by common ownership of WD and jobber members (such as NAPA, APS, Republic and ITT) or " noncontrolled" (such as Bumper-To-Bumper, Carquest and Pronto) (Brown 3383- , 3407- 08; Bull 3635- , 3642-43; Fife 657-58; Secrest 1065; Waters 3294 3297; CX's 535Z-20 through Z- , 26F- (33) G, l3P, 15J, 534Z-8). Program distribution groups are formed to enhance the ability of their members to compete with direct-buying retailers as well as other WDs and jobbers, to tie jobbers to their WDs and to undertake advertising and promotional programs which individual WDs or jobbers cannot afford on their own (Fife 657; Baumann 999-1000; Bull s 33Q3639; CX's 535Z-20 through Z- , 566). ("' J (Jursek 789; CX' camera, 566).

s 14Z- in camera 15N in camera). 146. ("*J (Sheehan 537-39; CX' For example, American Motors Corporation ("AMC") purchases carburetor kits from Ballwin/Washington which it sells to its dealerships for use by mechanics in the dealers' service facilities (Merz 2756 2766-67). Similarly, GM distributes carburetor kits it assembles to its car dealerships through the service channel in addition to selling kits through the traditional WD/jobber channel of distribution (Hawkins 1192).

147. (''' J (CX' s 163Z- , 464A- in camera 537Z-33 through Z-34). 148. Foreign vehicle producers' sales of carburetor kits in the Unitd States are limited to sales through their car dealerships and are enerally limited to kits with application on the vehicles they manulcture (Secrest 1098-99; CX 535Z-11 through Z-12). 149. ("' J (CX' s 14Z- , 15N in camera 26G in camera 27F in cam- 33L in camera j (CX's 535Z-17 through Z- , 541Z- , 162Y , camera, 26G in camera 27F in camera). Tomco is now the only major carburetor kit assembler selling through this channel (Thomp- 'n 343-44).

( , ECHLIN MANUFACTURING CO., ET At. 445 410 Initial Decision 150. C"' J (34) (Secrest 1060- , 1065-70; McCurry 3836-38; CX 464A- in camera 151. Ballwin/Washington s national account customers included virtually all of the domestic and foreign automobile manufacturers kit assemblers, and numerous private brand aftermarket product sellers. National account sales to private brand sellers accounted for about r" ' ) of Ballwin/Washington s total kit sales in 1979 and for about C''' J in 1980. National account sales to other kit assemblers accounted for about ('" J of' Ballwin/Washington s total kit sales in 1979 and for about (''' J in 1980. National account sales to carburetor rebuilders accounted for less than J in both years (Merz 2749-58; McCurry 3837-40; CX's 464A- in camera 404A- , 169H, 129Z-8 165V, 163Z-9; RX's 292-305, 307, 309-30).

152. Although only GM assembles and sells kits in the aftermarket at the present time, other automobile manufacturers are potential assemblers of kits. For example, Ford manufactures Ford carburetors and, unti about 1972, assembled and sold kits to the aftermarket. In 1972, although Ford had the capabilities to assemble and offer a full-line of kits, it made a management decision to purchase all kits from outside and cease kit assembly of its own, because buying kits was less expensive than assembling them (see Lingle 3511-18; Baumann 1001-12).

153. In 1967, Borg-Warner shifted responsibility for carburetor kit sales to WDs under the Borg-Warner name from Ballwin/Washington to APD (Merz 2715). APD, as a national account ofBallwin/Washington, served as Borg-Warner s primary distribution arm to the automotive aftermarket (Secrest 1066). s 24C in camera 143B). In l "' 1 (CX' 1979, APD accounted for (" ' ) of Ballwin/Washington s total carburetor kit sales and in 1980, for (" ' J (CX 464E in camera 154. Borg-Warner began selling carburetor components and kits to national accounts mainly for two reasons. First the economies of scale at the manufacturing plant in Washington were such that the more parts you could sell, the more profis you could make" and it did not matter whether the carburetor components were sold in bulk form or in kit form (McCurry 3840-41). Second, (" ' ) (CX 33E camera) s 26G in camera 27F l"' J (35) (''' J (Merz 2785, 2787; CX' camera 169G). In computing its share of the carburetor kit market Borg-Warner included its sales through APD under the Borg-Warner name as well as Ballwin/Washington s national account sales ofprivate-branded carburetor kits to resellers (CX's 140C 169Z- 17 through 18). Borg-Warner viewed its kit sales to reseller customers as part of its base market share of carburetor tune-up kits" (emphasis in original) (CX 140C).

155. In recent years, Carter was the only other carburetor kit assembler that sold carburetor kits to reseller national accounts (Se- 105 F.T. Initial Decision crest 1068-69; Fife 656-57; Jursek 743). However, Carter lost its last carburetor kit reseller customer, Guaranteed Parts, to Ball win/ Washington in 1981 (Sheehan 538; Eaton 2979-80; CX's 142, 184-85). D. Some Important Marketing Aids Provided By Kit Assemblers 156. The major carburetor kit assemblers provide important services to their customers, including a full line of well-consolidated carburetor kits offering broad coverage for the vehicles on the road, and certain marketing and sales support services. 157. Among the marketing and sales support services provided by assemblers are cataloging service and inventory guidance, sales personnel work with the customers' sales forces, warranty, obsolescence and stock adjustment programs, training clinics, advertising and other promotional materials and programs, and a fast delivery. 158. Carburetor kit catalogs are used throughout the distribution channel to locate the appropriate kit for a specific carburetor repair job (Sheehan 465; Fife 620-21; Thompson 304-05). Accurate, complete and up-to-date catalogs are, therefore, important in the sale of carburetor kits (Brown 3397-99; Jursek 780; Thompson 349-50; CX's 217 125 at 2, 182A, 419, 169I, 170Z- , 168Z-5 546B). Carburetor kit catalogs (or supplements) are prepared annually by all ofthe major assemblers and distributed to their WD customers (Schultz 3056; Sheehan 466-67; Thompson 351-52; CX's 40- , 249- , 416- , 426, 429- 439- , 445). Ballwin/Washington also annually provides catalogs supplements and other cataloging information to its national account kit customers (Secrest 1102-03; Kruse 2922- , 2926, 2929; CX' 169K, 141 , 187N, 217).

159. Carburetor kit catalogs generally contain sections cross-referfencing the carburetor kit part numbers to vehicle models and to carburetor model numbers and a numerical listing of (36) tlce assembler s kits (CX's 40-41 , 125 at 2, 416- , 439- , 445, 537Z-60). Some catalogs also cross-reference the assembler s carburetor kits to competitors' carburetor kits (CX's 429-30). Cataloging information is developed primarily from vehicle and carburetor manufacturers manuals (Fife 621; CX 537Z-60). Comparisons to competitors' kit catalogs are also undertaken to improve the format and contents of an assembler s catalog (Sheehan 466; Thompson 306). 160. Carburetor kit assemblers also prepare and distribute interchange lists which cross-reference their kit numbers to competitors kit numbers (CX's 169M, 170Z- 168Z-5 177 A- , 267, 427 A through , 419 535Z-49 , 85- , 88-97). (''' J (Waters 3316-17; Bull 3656-57; CX's 535Z- , 182A , 245I in camera). r*..*, ,n, 1 r- 'n "-.T' C\nn ...,.T' ( , ( ECHL1.N MA.NUl''AL" l'U1HNlj I.U. T AL. ,,,it 410 Initial Decision 419, 442A- , 452, 535Z-31). Ea h kit in the assembler s line is given a !!popularity designation" or !!popularjty code" based on its sales history or projected sales potential. (''' J (Jursek 784; Tehansky 846- 47; Thompson 347; Fife 659; CX's 44, 442A- in camera, 13P in camera). Inventory guidance is also provided to Ballwin/Washington national account customers (Secrest 1101 , 1105-06). 162. Suggested carburetor kit price lists are provided by the assemblers usually on an annual basis, to both WD and national account reseller customers (Eaton 2942; Timberlake 3093; Secrest 1101, 1104- 05; Sheehan 548; Thompson 346; Merz 2772-73; CX's 188A, 192). These suggested price lists generally include suggested prices at the jobber, mechanic Cdealer" or Htrade ) and consumer C'list") resale levels (CX's 257- , 45- , 192 449- , 420A-I; RX's 173-75). 163. In addition J (Kotcher 3575-76; Merz 2782; Waters 3317- 19; CX' s 33S in camera 13P in camera 15J in camera, 34U in camera 534Z-5).

164. ("' J (37) ("'J (Sheehan 545, 548; Tehansky 839-40, 843-44; Nelson 1618-19; Hawkins 1170, 1192; Brown 3377; CX' s 117 A in camera 534Z-5 , 244S-T in camera 419). To provide assistance to its WD customers, Borg-Warner employed a sales force of approximately 105 salesmen (CX 535Z-25); Standard employed approximately 350 salesmen (Fife 661-62); Echlin s Branford division employed approximatein camera, 245V in camera 534I, N-ly (" ' J salesmen (CX' s 244Z Waters 3303, 3313); Honey employed approximately 65 salesmen (Jursek 778); and Carter employed approximately 60 salesmen (Sheehan 544; Tehansky 834).

165. Of the major assemblers, Tomco is the only one that does not employ its own direct sales force. In an effort to control its distribution costs, Tomco has elected to use manufacturers' agents, which in turn employ approximately r" ' J commission salesmen, to distribute its carburetor kits (Thompson 288 , 362; CX 570D; see CX 117C V in cameracamera). J (Thompson 361-62; CX's 535Z-25, 33T, 166. Carburetor kit assemblers provide a limited warranty on the kits which they sell as is customary in the automotive aftermarket (Fife 666-67; Waters 3320-21; CX 541Z-25 through Z-26). 167. Carburetor kit assemblers' sales forces attempt to " changeover" WDs from competitive carburetor kit lines to their kit line (Jursek 783-84; Tehansky 840; CX's 125 at 26 270-77). Changeovers however, occur more frequently at the jobber level than at the WD level (Tehansky 840; CX 535Z-30). ("'J (Waters 3320; Tehansky 843- 45; CX' s 244S in camera 125 at 26 , 245B in camera 262A- , 263A- 301-03). Once the assembler s sales force convinces ajobber to changeover kit lines, the salesman will inventory the jobber s existing kit supply, lift those kits and replace them with the assembler s kit line Initial Decision 105 F.T. (Fife 665-66; Tehansky 844-45; Waters 3320; CX's 65- , 125 at 26 263A, 284- , 535Z-29 through Z-33). ("') (CX's 534Z-9 through Z-1O, 73, 244Z through Z- in camera 261A, 263A). In lieu of the changeover" program, Ballwin/Washington provides its fuji-line kit customers with a stock adjustment program to offset some ofthe costs inherent in such changeovers (Secrest 1108-09; CX's 183, 185, 188B 225C, 168. Carburetor kit assemblers also provide their WD customers 541Z-27).with obsolescence protection programs to facilitate the (38) return of slow-moving, obsolete or superseded carburetor kits (Thompson 351- 54; Tehansky 847; Fife 666-67; Jursek 783; Merz 2768-69; CX's 535Z- , 541Z-26 through Z-27). This obsolescence protection is passed down through the channels of distribution to allow inventory adjustments at the jobber level (Waters 3320-21). Ballwin/Washington provides a similar obsolescence program to its rescUer national accounts (Secrest 1106-07; CX 193B-C).

169. ("'J (Brown 3377; Tehansky 850, 858; Jursek 784-85; Hawkins 1193-95; CX's 429 419 314, 534N, 307, 309A, 313, 570D , 15J in camera).

170. ("'J (Fife 639, 660; CX's 169M , 244Z-8 in camera). For example, Echlin periodically prepares comparisons of its kit prices to those of the other major assemblers for use by its customers' sales forces in selling kits at the jobber and installer level (CX's 259A, 260A-P). Similarly, Ballwin/Washington has prepared detailed comparisons of its kit line with the other major assemblers' kit lines for use by its national accounts in their kit sales efforts (CX's 105A through Z- 537Z-64 through Z- , 83, 84A, 99A through Z- , 129Z-1). 171. ("') (Thompson 347-48; Sheehan 551-52; Jursek 780; CX' 119A- , 432- , 244Z- in camera 245J- in camera 117 A in camera 121, 123C, 125 at 2, 431A- , 419). The purpose of such advertising is to build brand name recognition, thereby making it easier for the assembler s customers to sell that brand of carburetor kits (Fife 660- 61; CX's 570D, 123C).

172 (" ' J (CX' s 190F, 167D, 184 232 at 3, 164Z , Z- , 167D, 182A 170Z- , 168Z-5, 14C , N in camera 26G , J in camera). For example Borg-Warner used displays to compare its kits to those of Standard (Hygrade), stressing those items "which markedly show a quality difference" (CX 83). Borg-Warner also supplied its national accounts with detailed, kit by kit comparisons ofthe contents of competing kits (CX' s 98- , 101- , 105A through Z-4). An assembler s reputation for providing high quality kits is, in fact, one of the critical factors WDs and resellers evaluate in selecting their carburetor kit supplier (Thompson 349-51) 17:1. r (1' phRnRkv R4R-49: Rmlm:;nn 10? F;:;tnn ?QRn. p.rrp ECHLIN MANUFACTURING CO.. ET AL.

Initial Decision 1101; CX's 535Z- , 541Z-35 , 33S in camera 34W in camera 133A Thus, a carburetor kit assembler s ability (39) to supply product, quickly after receiving an order (i. a high "order lil rate ) is impor tant (Thompson 349-50; Timberlake 3105-06; CX's 133A- , 76B-C). The carburetor kit order fill rates ofleading assemblers are very high and range between high 80% and high 90% (Thompson 354; Sheehan 546-47; Fife 667; Jursek 781; CX's 539Z- , 534Z-12 through Z- 535Z-33). Also, in order to insure prompt delivery, leading carburetor kit assemblers maintain at considerable cost a large inventory of kits they offer.

VI. THE EFFECTS OF THE CHALLENGED ACQUISITION A. The Market Structure Evidence 174. At the time ofthe acquisition, there were seven carburetor kit assemblers or packagers offering a full line of carburetor kits and three small special1ized kit assemblers that operated only in a segment ofthe market. And a limited number of carburetor kits were imported for foreign vehicle applications. Total sales of carburetor kits amounted to approximately 13 321 390 units in 1980 and 13 562 030 in 1979 (CX 530A). In 1980, the dollar sales of carburetor kits were about $53 milion at the level of sales to WDs (Nelson 1487; Jursek 816). This would represent some $200 million spent on carburetor kits by the ultimate consumers (i. vehicle owners) (Glassman 4386-94). 1. Firms Which Assemble And Sell Carburetor Kits a. Borg- Warner Corporation 175. Prior to the acquisition Borg-Warner was the nation s largest carburetor kit assembler (CX's 81C, 530A). In 1979, Borg-Warner assembled and sold 5 106 564 carburetor kits which accounted for 37.7% of all carburetor kits sold in the United States. Its 1980 sales amounted to 4 846,396 kits or 36.4% of the market (CX's 530A-CJ. Borg-Warner entered into the assembly and sale of carburetor kits through its acquisition of Precision Automotive Components Company (PACCQJ in 1966 (Carlson 2525-26; CX 535HJ. Borg-Warner sold its kits into the aftermarket both under the Borg-Warner brand name through APD and on a private label basis through sales by Ballwin/ Washington to its national accounts (F. 150- supra 176. Borg-Warner assembled carburetor kits at the Ballwin facility in Ballwin, Missouri. Carburetor parts used in its kit assembly operation were manufactured at its Washington, Missouri (40) plant (F. 37 supra; CX 539Z-14 through Z-15; Bush 2908). Borg-Warner manufactured the majority ofthe carburetor hard parts included in its kit line while purchasing various small parts and subcomponents such as FEDERAL TRADE COMMISSION UM"WM"

105 F. Initial Decision eings, rubber cups and clips from outside suppliers (CX's 126B 9Z-14 through Z-16, 533Z through Z- , Z-15 through Z-16; Bush , 2915-17). At the time of the acquisition, Borg-Warner assemed a full line of carburetor kits consisting of429 kits (CX's 164" i7R, 168P, 170P). Borg-Warner advertised that its kit line offered ;he broadest coverage available" in the industry (CX's 120 , 1691). org-Warner did not purchase any kits from other suppliers (CX 41R; Bush 2908). Most of its competitors purchased some kits from lorg-Warner s Ballwin/Washington Division in order to fill out their ines (McCurry 3836; Fife 648; Hawkins 1188-89; Thompson 334-35). n addition, Borg-Warner, through its Ballwin/Washington Divisio llso supplied carburetor parts to other carburetor kit assemblers :Thompson 326-27; Sheehan 496-98; Insalaco 919; McCurry 3836) and was considered the major full line carburetor parts supplier in the industry (Insalaco 974-76).

b. Standard Motor Products, Inc.

177. Standard Motor Products, Inc. ("Standard") is the nation second largest assembler of carburetor kits. In 1979 and 1980, Standard had sales of approximately 2 889,600 and 3 169 664 kits, respectively, and accounted for about 21.3% and 23.8% of the market (CX 530A-C). Standard also offers four other products to the automotive replacement market: ignition parts, wire and cable, general service parts, and automotive air conditioning parts (Fife 610). Standard entered into the assembly and sale of carburetor kits in 1947 when it purchased Hygrade Motor Products, an assembler of major overhaul kits (Fife 615-16). During the late 1950s and early 1960's, major overhaul kits were replaced by carburetor tune-up kits (Fife 616). 178. Standard assembles carburetor kits at a plant in Edwardsvile Kansas and distributes them nationally under the "Hygrade" brand name (Fife 612-13; CX 440). Standard offers a full line of 400 to 500 carburetor kits which provide coverage for approximately 95% of all the vehicles on the road, including domestic and imported passenger cars, light and medium trucks, some heavy trucks, light tractors and other farm equipment (Fife 621-22). Standard manufactures almost all of the hard parts included in its line of kits (Fife 641-43), but generally does not offer any carburetor parts or components for sale s 465Ato other kit assemblers (Fife 645). (' " J (Fife 648-49; CX' camera 464B in camera ). (41J c. Echlin Inc.

179. Before acquiring the carburetor kit business of Borg-Warner Echlin was the third largest kit assembler. In 1979, Echlin assembled and sold 1 365 598 carburetor kits accounting for 10. 1 % of the mar- ( ECHLIN MANUFACTURING CO. , ET AL. 451 410 Initial Decision keto In 1980, its share ofthe market was 1 390 121 kits or 10.4% (CX' 530A-C). Echlin entered into the assembly of carburetor kits in the early 1960's (CX 538X; Schultz 3026, 3033-34). Echlin s kit line is marketed to the distribution centers ("DCs ) (WD members) of the NAPA program distribution group by its Automotive Controls Corporation C'ACC" division located in Branford, Connecticut. Research and development work for Echlin s carburetor kits is also handled by ACC at Branford (CX 538F , 1).

180. Echlin s carburetor kits are assembled at and shipped from the Litchfield, Ilinois facility of its Brake Parts Company division (Schultz 3047-48; Timberlake 3129-30). Parts for its kit assembly operation are also purchased and inventoried by Brake Parts Company (CX 538N). Historically, Echlin purchased its requirements of carburetor parts for inclusion in kits from outside suppliers (Schultz 3040-42; Sheehan 496). At the time of the acquisition, Echlin was in the process of developing manufacturing capability for some pump plungers and diaphragms to be used in its kit assembly operation. Echlin s plan was to offer 30-40 premium carburetor kits as part of its line featuring parts made with a premium grade of fluorocarbon rubber (CX 538Z-75 through Z-77). This project was placed "on hold" in late 1981, after Echlin purchased Ballwin/Washington (CX 538Z- 76 through Z-77; Smith 3235-37). ('''J (Schultz 3039; CX's 538Zin camera). Historically,through Z- , 464A- in camera 467 A- Baliwin/Washington has been Echlin s major source of purchased kits, supplying a majority of Echlin s kits for import applications (Secrest 1070; Schultz 3061). Echlin s purchased kits accounted for less than 10% of its kit sales (Schultz 3036). d. General Motors Corporation 181. General Motors Corporation (GM), through its Rochester Division, was the fourth largest producer of carburetor kits in the United States (CX 530A-C). GM's sales of kits it assembled were approximately 1 240,616 in 1979 or 9. 1 % of the market, and 1 072 538 kits or 8% in 1980 (CX 530A-C). GM, which is also an original equipment carburetor manufacturer, has been assembling carburetor tune-up kits since at least the early 1960's (Hawkins 1174-76). GM currently assembles kits for application on all of its own Rochester-manufactured carburetors. ) (42) (" ' ) (Hawkins 1175-78; CX 468B in camera).

182. GM assembles carburetor kits at a plant in Tuscaloosa, Alabama (Hawkins 1178). In its kit assembly operation, GM uses parts manufactured for its OE carburetor production but does not sell any kit components to other kit assemblers (Hawkins 1187). GM purchases kits for non-Rochester carburetor applications from Ballwinl Washington (Hawkins 1188-89). Its purchases from Ballwin/Wash- ..

Initial Decision 105 F. ngton accounted for approximately ("'J of its kit sales for the period 468E in camera). 1979-1980 (CX' s 464A in camera, e. ACF Industries, Inc.

183. ACF Industries, Inc. ("ACF"), through its Carter Automotive Division ("Carter ), was the fifth largest assembler of carburetor kits . in the United States (CX 530A-C). In 1979, Carter s sales of kits it assembled represented 7.7% ofthe market and its 1980 share was 7% (CX 530A-C). Carter is a supplier ofOE carburetors to the automobile industry (Sheehan 464-65). Carter began assembling tune-up or "zipkits" in the late 1950's or early 1960's (Sheehan 449, 452). Carter offers a fuUline of carburetor kits for sale in the aftermarket (CX' 416-18, 469B; Sheehan 463- , 466-72).

184. Carter assembled its carburetor kits at a facility in Corning, Arkansas (Sheehan 443, 452-53), for both Carter and non-Carter carburetor applications and purchased about (" ' J of its annual kit sales from others, including Ballwin/Washington (Sheehan 473-75; CX 469A in camera ). In its kit assembly operation, Carter used carburetor parts it manufactured in-house, parts manufactured by outside suppliers on a contract basis using tooling owned by Carter, and parts offered for sale by carburetor parts manufacturers such as Ballwinl Washington (Sheehan 491- , 496-97). Carter sold carburetor parts only for Carter carburetor applications to other kit assemblers (Sheehan 496; Insalaco 985).

f. Tomco, Inc.

185. Tomco was the sixth largest carburetor kit assembler at the time of the acquisition. Its share of the market was approximately 6.4% in both 1979 and 1980 (CX 530A-C). Tomco is a privately-owned supplier of fuel system products to the automotive aftermarket. Its products include: carburetor kits, choke thermostats, choke pull-offs chokes, fittings and adaptors (Thompson 285 296). Tomco began packaging carburetor tune-up kits in 1963 (Thompson 297) and offers a full line of some 300 to 400 carburetor kits (Thompson 314; CX's 445, 447). (43) 186. Tomco is located in St. Louis, Missouri (Thompson 284). Its kits differ slightly from others in that they include a rotary disc fuel inlet valve, a proprietary product ofTomco, rather than a needle and seat assembly (Thompson 295-96). Tomco has its fuel inlet valve manufactured on a contract basis by an outside supplier using Tomco owned tooling (Thompson 324-25). Other carburetor parts are purchased by Tomco from carburetor parts manufacturers such as Ballwin/Washington (Thompson 326-27). Tomco also purchases certain slow-moving 1 1 - L'_ 1=", llmin IW Rshincton to fill out its line g.

vrlLm MANUFACTURING CO.. ET AL. 453 Initial Decision (Thompson 334-35). Purchases of such kits account for less than 10% of Tomeo s total kit sales (Thompson 2457-58). Holley Replacement Parts Division Of Colt Industries Operating Corporation 187. At the time of the acquisition, the Holley Replacement Parts Division of Colt Industries Operating Corporation ("Holley ) ranked seventh in the carburetor kit market in the United States, accounting for 3.3% of the market in 1979 and 3.4% in 1980 (CX 530A). A sister division, Holley Carburetor Company, manufactures inter alia, and replacement carburetors for the automotive industry (Jursek 707). Holley Carburetor Company's aftermarket products are distributed by the Holley Replacement Parts Division (Jursek 708-9). Holley first began to offer carburetor tune-up kits in the late 1950' (Jursek 739). Holley offers a full line of some 600 kits under the Renew Kit name as well as two other specialized lines of carburetor kits (J ursek 715-20).

188. Holley assembles carburetor kits in Goodlettsvile, Tennessee (Jursek 714). Parts for inclusion in kits for both Holley and non- Holley carburetor applications are supplied to the Replacement Parts Division by its sister division, Holley Carburetor Company (Jursek 733 34), as well as by other suppliers (Jursek 731). Holley Carburetor Company also sells carburetor parts for Holley carburetor applications to other assemblers (Insalaco 985; Secrest 1116). Holley has always assembled a majority of its carburetor kits (Jursek 738) and in recent years instituted a program designed to increase the percentage of its line that it assembles (Jursek 741-43; CX's 546B, 547A 548A). Holley purchases kits for low-volume applications, primarily from Ballwin/Washington (Jursek 727-30). Holley s kit purchases accounted for approximately (' "J of its kit sales in 1979 and 1980 (CX' s 471G-D, 464A- in camera). (44) h. Other Carburetor Kit Assemblers 189. At the time of the acquisition, there were three other smaller kit assemblers operating in two market segments (CX's 475A- , 476A , 477 A-C). Royze, located in southern California, assembled and sold kits for import vehicle applications only (Merz 2733-34). Royze kit sales accounted for about 2% of the kit market in 1979 and 2. 1 % in 1980 (CX 530A-C). Royze also purchases some kits from Ballwinl Washington to fill out its import car kit line (Merz 2734). 190. Facet Enterprises, Inc. ("Facet") and Tru-Part Manufacturing Inc. ("Tru-Part") both assemble and sell kits for application on off-theroad vehicles, such as tractors and earth-moving machinery (CX' 476A, 477 A). Facet also assembles a very few kits for inboard marine ( ( FEDERAL TRADE COMMI,SSlUN lJ-MS'tmv u-, Initial Decision 105 F. pplications (CX 476A). Neither compa;'y assembfes any carburetor. jts for passenger cars (CX' s 476A, 477A). In 1979 and 1980, Facet and fru-Part together accounted for less than 1.5% of the carburetor kit narket (CX 530A--). Due to the increasing displacement of gasoline engines by diesel engines in agricultural machinery, this "agricultur- 11 and industrial" segment of the carburetor kit market has been declining steadily (Nelson 1462-63).

i. Foreign Kit Competition 191. Despite the growing popularity of imported cars in the United States, carburetor kits imported in 1979 and 1980 accounted for only about 1.2% of the carburetor kit market (Nelson 1482-83; CX 530A- C). This is due to tpe fact that demand for carburetor kits for imported cars is largely met by domestic kit assemblers (Nelson 1463-66, 1483- 86). Ballwin/Washington is the primary supplier of import car kits to other domestic suppliers (Sheehan 473; Jursek 727-28; Hawkins 1188 90; Secrest 1078; see CX 115A), ("*J (Merz 2762; Secrest 107&-80; Nelson 1464-65, 1483-86; CX 464A- in camera), j (Nelson 1484; Secrest 1075-76; CX's 115A, 537Z-34 through Z-35, 464A- in camera 192. The small number of imported kits brought into the country primarily by foreign car manufacturers, such as Nissan, Toyota, and Honda, are sold through their dealer networks for use on their cars (Secrest 1081, 109&-99; CX's 461A- , 478A- , 479A-E, 480A- , 535Z 11 through Z-12). In addition, some ofthe import kit resellers import a small volume of import car carburetor kits (Nelson 1464-65; CX' 461A- , 535Z-12 through Z-14). Royze also imports a limited number of import car carburetor kits (CX's 474A-C, 461A-B). However, the record shows (45j that no company imports carburetor kits for use on cars of domestic manufacture at the present time (Nelson 1466). 193. It may be expected that, as imported cars continue to gain an increasing share of the U.S. car market, carburetor kit imports for foreign car applications wil gradually increase. However, the record does not indicate that imported carburetor kits will capture a significant share of the U.S. carburetor kit market in the near future (Nelson 1482-84; CX 212A). ( j (CX' s 24Z-3 in camera, 180A). 2. The Measurement Of Market Shares 194. The measurement of market shares in the carburetor kit maret icomplicated by the presence of some J so-called resellers thos, ' rms which do not assemble kits but buy them from Ballwin/ Washmgton (Merz 2749-59; Timberlake 3108; RX's 292-307, 309-30), and sell them under their own private labels to the wholesale channels of distribution.

410 Initial Decision 195. It is settled that, once the product market is determined as is the case here, it is appropriate to include a firm s sales to resellers (private branders) in the firm s market share for the purposes of market share analysis. See United States v. Black Decker Mfg. Co., 430 F.supp. 729, 737-38 (D. Md. 1976); Beatrice Foods Co. 3 Trade Reg. Rep. (CCH) TI 22 035 at 22 624 (May 26 , 1983); Champion Spark Plug Co., Docket No. 9141, Initial Decision at 90 n. 10, adopted by the Commission (Final Order dated June 20, 1984) l103 F. C. at 623). In the instant case, although the assemblers do not "manufacture" or process" kits, they are in a real sense "producers" of kits and the relationship between a kit assembler and a kit reseller is similar to the relationship existing between a manufacturer/processor and a reseller involved in the cases cited hereinabove. 196. Furthermore, in computing its kit market share in internal marketing documents, Borg-Warner itself included its kit sales of both APD and Ballwin/Washington (private label sales) (CX's 140C 169Z-17 through Z-18) and viewed its kit sales to "reseller" customers as a part of its base market share of carburetor tune-up kits" (emphasis in original) (CX 140C).

197. Also, unit sales are more realistic as well as more meaningful because of the distortions in dollar sales that would be introduced by the sales to and by the resellers.

198. The MEMA, an industry trade association, also collects and publishes carburetor kit sales data on a unit basis (CX 436A-B). (46) 3. Market Shares And Concentration Ratios 199. Prior to the acquisition, Borg-Warner s sales ofear retor kits it assembled accounted for about 37.7% of the market j 1979 end 36.4% in 1980. Standard, the second ranked assembler, accounted for about 21.3% in 1979 and 23.8% in 1980 (CX 530A-C). Echlin, the third ranked firm, accounted for about 10.1% in 1979 and 10.4% of the market in 1980 (CX 530A-C). General Motors, the fourth ranked firm accounted for about 9. 1 % in 1979 and for 8% in 1980. The pre-acquisition four-firm concentration ratio in the kit market was about 78. in 1979 and 78.6% in 1980 (CX 530A). The pre-acquisition two-firm concentration ratio in 1980 was about 60.2% (CX 530A). Thus, the kit market was an advanced oligopoly prior to the acquisition. 200. As a result ofthe acquisition, Echlin increased its share of the market to 46. 8%, on a 1980 pro forma basis, thus becoming the nation s leading kit assembler. Echlin s post-acquisition share was almost twice that of its closest rival, Standard, and almost six times that of the third ranked assembler, GM (CX 530A-C). The resulting twofirm concentration ratio was 70.6% and the four-firm concentration ratio rose to 85. 6%, on a 1980 pro forma basis (CX 530A-C). ).

Initial Dccision 105 F.TC. 201. The pre-acquisition Herfindahl-Hirschman Index ("HHI" measured approximately 2 172 in 1980 (CX 530A), in the range generally considered to be "highly concentrated." As a result ofthe acquisition, the HHI in the kit market increased by 757 points to about 2 929 on a pro forma basis (CX 530A).

202. The acquisition' also eliminated competition between the merging firms. In 1980, prior to the challenged transaction, Borg-Warner and Echlin were the first and third largest assemblers of carburetor kits, with 36.4% and 10.4% ofthe market, respectively (CX 530A). By acquiring Borg-Warner s automotive aftermarket assets, Echlin has eliminated this competition.

B. The Challenged Acquisition Constitutes A Probable Violation Of Section On The Basis Of Market Structure Evidence 203. Under established judicial and enforcement criteria, the resulting market shares of the acquisition and the degree of market concentration in the kit market are so high that the acquisition wil be a presumptive violation of the merger law and be prohibited unless it is shown that the market shares are not reliable indicia of the true competitive significance ofthe acquisition or that the market is likely to perform competitively regardless ofthe acquisition. See IV Areeda & Turner, Antitrust (47) Law nn 909-12 at 29-68 (1980); Sullivan Antitrust, Sections 204a and 204b at 613 21 (1977); BASF Wyandotte Corp. 100 F. C. 261 , 392 (Initial Decision dated May 14, 1982). In the instant case, the administrative law judge wil be guided by the 1982 Statement of Federal Trade Commission Concerning Horizontal Mergers dated June 14, 1982 ("FTC Guidelines ) and wil also take into consideration the recently revised Department of Justice Merger Guidelines (June 14, 1984) ("DOJ Guidelines Also see Clanton Focusing the Inquiry: Specificity In The Merger Guidelines and Elsewhere 71 Calif. L. Rev. 430, 433 36 (1983); Greenfield Beyond Herfindahl: Non-Structural Elements of Merger Analysis, 53 Antitrust Law J. 299 (1984).

204. The FTC Guidelines indicated the Commission s view that while market share data remain "an important indicium of the likely competitive effects of a merger " it wil employ "a more refined treatment" of such data in light of "more recent empirical economic research and well over a decade of practical experience in analyzing and evaluating horizontal mergers" and take into account "non-market share considerations " the most important of which being entry barriers. FTC Guidelines at 2- ECHLIN MANUFACTURING CO. , ET AL. 457 410 Initial Decision VII. CERTAIN NON-MARKET STRUCTURE FACTORS WHICH BEAR ON THE COMPETITIVE EFFECT OF THE ACQUISITION 205. The FTC Guidelines thus mandate a further inquiry into those additional factors relevant to the assessment of market power effects" and suggests an examination of "whether any market power conferred by the merger is likely to persist over time" (market power duration factors) and "whether market conditions are conducive either to the exercise of individual firm market power or to collusivetype behavior. FTC Guidelines at 3-4. 206. Such additional factors relevant to the assessment of market power effects ofthe challenged merger include two marketwide conditions in the carburetor kit market, namely, the restraining influence of substitute products and the limited power of kit assemblers to control competition by resellers.

A. Substitute Products Compete With Carburetor Kits 207. Although new and rebuilt carburetors are not close-enough substitutes for carburetor kits to be included in the same product market, rebuilt carburetors wil have a significant restraining influence upon the market power of the merging firms. 208. The record is clear that the firms in the carburetor kit market regard rebuilt carburetors as competitive products, that they are generally aware of the price range of rebuilt (48) carburetors, that there is a general relationship between the prices of the two product groups and that there is a general perception that the price movements of rebuilt carburetors may affect the sale of kits to some degree. B. Kit Assemblers ' Power To Control Competition In The Kit Market Is Limited 209. In measuring market shares, the private brand sales of Ballwin/Washington to some 40 resellers were included in Borg-Warner market share. However, the record shows that Borg-Warner s power to control competition in the kit market is limited in important respects. And this fact serves to diminish the significance ofthe merging firms ' market shares as a surrogate measure of their market power. 210. For example, Borg-Warner had no control over the terms of sale of private brand kits by its reseller customers (Eaton 2959; McCurry 3847; Secrest 1133). The assemblers also have only a limited control over the design and quality of private brand kits. And in many cases, the design, quality and consolidation of a kit is determined by the reseller alone or in consultation with the assembler (see RPF 658 68). The evidence also shows that the contents of kits for each vehicle are more or less standard regardless of the seller (Sheehan initial Decision 105 F. 477; Hawkins 1185). The consolidation decisions, once critical to competition, are no longer a diffcult task and most firms appear to have similar consolidation patterns today (Merz 2786; Tehansky 872). There also appears to be a great deal of design copying among competitors (Fife 690; Tehansky 853; Eaton 2975; Smith 3172). 211. The resellers sold kits not only in competition with APD but also with other kit assemblers. Most kit assemblers regardedresellers as their competitors. The buyers (WDs, jobbers and installers) made no distinction between assemblers and resellers (e. , Thompson 413 2482; Carlson 2648-49; Merz 2740).

212. The evidence shows that most of the national accounts follow Ballwin/Washington-suggested price lists (CX 225B). However, this evidence does not prove that the resellers do not have the freedom to set their own prices. In fact, there is testimony that the national accounts are able to exert some influence on Ballwin/Washington pricing decisions (Eaton 2946-47).

213. Analysis of price changes of some kit resellers, although based on selective and incomplete data, suggests that these resellers appeared to have enjoyed significant freedom to pursue their own pricing strategies, indicating a significant limitation on Borg-Warner power to control kit prices (see RX's 348-54; Glassman 4067- , 4088- 4114- 4142-43). This (49) is not surprising in view of the tact that Ballwin/Washington sells kits to resellers at the same prices it charges APD and that resellers enjoy about the same gross margins (from 33% to 40%) 2 APD (McCurry 3855-56) in the pricing of purchased kits.

214. The power of kit assemblers over kit prices is also significantly limited by competition from substitute products, especially replacement carburetors, which places a significant constraint on the market power of the merging firms. See FTC Guidelines at 12. C. Entry Barriers Are Very Low 215. In the Commission s view, entry barriers "is perhaps the most important qualitative factor" in determining the probable impact of a merger "for if entry barriers are very low it is unlikely that market power, whether individually or collectively exercised, will persist for long. FTC Guidelines at 3- 216. The record is clear that entry into the assembly and sale of carburetor kits is rather easy and that entry barriers are very low. 217. Complaint counsel assert that economies of scale (CPF 286-93), parts production or procurement barrier (CPF 295-308), sunk costs (CPF 308-14), and product differentiation (CPF 314-19) constitute substantial entry barriers into the carburetor kit market, relying Initial Decision 218. The record, however, does not show-the presence of such commonly recognized entry barriers as legal license, patents and other secret processes, control over scarce resources, entrenched buyer preferences or product differentiation and high capital costs (Glassman 4215-38). See generally, II Areeda & Turner, Antitrust Law, TI 409 at 298-306 (1978).

219. The record does not show the presence of any significant technological barriers to entry into the assembly of carburetor kits. Although some assemblers use patented components, patents and secret processes do not hinder anyone from starting a kit assembling operation.

220. The record does not show that kit assembly requires a high degree of technological know-how. Although the development and maintenance of a kit line requires a degree of knowledge about automotive carburetors and some expertise is involved in kit consolidation and component procurement ("sourcing ), there is no evidence to show that such knowledge or skil is scarce or prohibitively expensive. Indeed, the kit industry was started by automobile mechanics and automotive products salesmen with (50) enterpreneurial vision. And the kit assembly itself is performed by semi-skilled workers and requires minimal training.

221. The record does not show any significant production barriers such as control of scarce resources. Although the record is clear that the kit component manufacturing capability, achieved through backward integration, is a distinct competitive advantage in kit assembly and that the leading kit assemblers (now including Echlin) do make most of the major kit components, the record also shows that there are sufficient numbers of alternative sources of supply for kit components (see RPF 42-47, 404-22). At any rate, the record does not show and complaint counsel do not suggest, that entry is possible only with kit component manufacturing capability.

222. The record fails to show that a new entrant must bear heavy promotional costs and overcome deeply entrenched buyer preferences. Brand loyalty appears to be no longer an important factor in the sale of carburetor kits, whatever its importance may be now in the marketing of other aftermarket products. Although kit suppliers (both assemblers and resellers) use brand names, kit installers seldom specify brands when ordering. It is also true that customers at all levels of distribution want to carry quality products and that a reputation for quality products is a valuable marketing tool. However, the record is clear that product quality in the kit market is taken for granted today and that this fact is largely responsible for the diminishing brand loyalty in the kit market (see F. 51- supra). 223. Although there is some evidence showing that a new entrant 105 F.T. Initial Decision will face some buyer resistance until its products become known in the aftermarket channels of distribution (CPF 319), the record also shows that "changeovers" are common occurrences both at the WD and jobber levels (F. 167 supra; CPF 142-43). 224. The only distributional barrier in the kit market appears to be the need to develop and maintain a full line of kits not only for the multitude of old model carburetors but also for numerous new model carburetors as they appear each year (F. 127- supra). It isreasonable to conclude from the evidence that a new entrant would be unable to compete effectively until it attains a full-line or near-fullline capabilty. However, the record also shows that a would-be entrant can enter the kit market with a limited line of conventional or modified kits and gradually expand its line. 225. However, with respect to the so-called national accounts or resellers (of private brand kits) which historically have purchased their kit requirements from Borg-Warner s Ball win/Washington Division, a considerable degree of buyer preference for Ballwin/ Washington appears to exist. This is not surprising in view ofthe fact that for some years Ballwin/Washington has been the sole source of private brand kits for resellers until November 1981, when Sherman Carburetor (51) Company, an established basic manufacturer of carburetor and carburetor kit components, began soliciting the national account business in competition with Ball win/Washington (see F. 243 infra ). The record shows that Sherman has made only a modest gain in this venture in spite of substantial price concessions it offered. Sherman s lack of success, however, may be due largely to its failure to offer a full-line of kits to the national accounts or private brand sellers who offer a full-line to WDs and would rather buy their kit requirements from a single source until Sherman becomes a fullfledged second source of a full line of private brand kits (see F. 244 252 , 254 infra).

226. Capital costs in the sense that entry requires a large absolute expenditure offunds do not constitute a "barrier to entry" unless they are so high as to be prohibitive. A new entrant may of course have to pay higher interests than established firms, but this wil likely be the case in all industries. See II Areeda & Turner, n 40ge. Capital costs for starting a kit assembly operation are rather modest (e. , Carlson 2545-49, 2558-64).

227. Complaint counsel's u economies of scale" argument, simply stated, is that (1) since each of the leading firms either accounted for 10% or more of the total market sales in 1980 or possessed the capacity to assemble that many kits, the minimum etTcient scale operation in the kit market is about 10% ofthe market or 1.3 millon units, and (2) since the minimum effcient scale operation is 10% of the market , --_u. "-'A.' .'A..A"". "H''' VV. 1 n1.. ,-U.l Initial Decision sales, scale economies are substantial ahdwil seriously impede entry (see CPF 281-94).

228. At the outset, the evidence relied on by Dr. Nelson, the government' s expert witness, for his conclusion that the minimum effcient scale for entry into the kit market is 10% of total sales is insuffcient to establish Dr. Nelson s sweeping conclusion and is not persuasive (see CPF 293-94).

229. Not surprisingly, the record shows that some of the familiar elements contributing to scale economies, such as higher volume effciency realizable in machinery and plant cost (Nelson 1592; 1596-97), labor cost (Nelson 1586-87, 1593-95) and materials cost (Nelson 1597- , 2164, 2245-46, 2261), are also present in the kit assembly operation. Thus, it is safe to conclude that scale economies do exist in the kit market. However, there is evidence indicating that scale economies in the kit assembling operation are "inconsequential" and less significant than those present in the manufacture of carburetor and carburetor kit components (see McCurry 3840-45). 230. As Areeda and Turner put it succinctly, however, to the extent that scale economies impede entry, it is not because new entrant is unable to produce at the same cost as incumbents (as it would be usual11y the case in most industries), but because entry at minimumcost scale (or minimum effcient scale) would so expand the supply as to depress the price and profit that entry (52) would not be attractive. This may very well be the case in many small markets such as the kit market (1980 sales of about $55 milion). From an antitrust policy point of view, to call this a "barrier to entry" would be somewhat incongruous. See Areeda & Turner, II Antitrust Law, TI 409. 231. Also, there is a school of thought which holds that economies of scale do not deter entry, especially in cases where capital costs associated with entry are low (Glassman 4273-77). The carburetor kit market is such an industry.

232. In terms of market power assessment, it has been suggested that even where the minimum effcient scale or economies of scale severely limit the number of firms that can operate profitably in the market, the incumbents will have no "meaningful market power" if sunk costs associated with entry are low. See Wentz Mobility Factors in Antitrust Cases: Assessing Market Power in Light of Conditions Affecting Entry and Fringe Expansion," 80 Mich. L. Rev. 1545 1591 (1982), cited in RPF at 214 n. 43.

233. "Sunk costs" is that portion of capital costs associated with entry which cannot be recovered if a would-be entrant decides to exit from the market. High sunk costs can impede entry into a market (see Nelson 1575- 2011-12; Glassman 4240-44; Willig 4776-78). 234. Dr. Nelson testified essentially that, although the absolute ..

Initial Decision 105 F. capital necessary to enter the kit market is not suffciently large as to impede entry, the portion of initial entry costs that constitutes sunk costs is suffciently high to deter entry (CPF 309). Dr. Nelson then enumerated and elaborated on the main components of sunk costs, such as Hdevelopment expenditures/' inventory costs and pro motional costs (CPF 310-13).

235. However, the record shows that entry costs in terms of c",pital costs are rather modest to begin with. "The development expenditures" and promotional costs associated with entry would largely be unrecoverable in most industries. According to Dr. Nelson, as a rule ofthumb, 50% of inventory costs may be unrecoverable (Nelson 2026- 27). From the foregoing, it is fair to conclude that there are sunk costs associated with entry into the kit market, but the evidence does not show that the sunk costs are suffciently high to deter entry whether considered alone or in conjunction with the economies of scale discussed hereinabove.

236. From the foregoing discussions, it is concluded that barriers to entry into the carburetor kit market are very low and are not likely to impede entry that may otherwise be expected to occur in response to noncompetitive performance in the carburetor kit market. (53J 237. The evidence relating to the recent history of new entries is somewhat mixed in that the record shows only one substantial entry during the last decade in addition to a number of small-scale entries during the same period.

238. Former president ofBorg-Warner s APD division testified that Nissan and Toyota of Japan began marketing carburetor kits in this country sometime during 1974 through their dealer organizations and Japanese trading companies. However, they have not yet achieved significant sales volume (Merz 2727-32). 239. Tru-Part Manufacturing Company, a reseller of industrial and agricultural kits, began to assemble a line of kits sometime during the last ten years, but its sales are believed to be small (RPF 928). 240. Sherman Carburetor Company ("Sherman ) is an old manufacturer of carburetor parts which began its business in 1939 (Insalaco 887). As of 1982, Sherman made pump plungers, pump diaphragms and economizer valves, assembled needle and seat using purchased parts and also purchased gaskets and other misce1Janeous small parts to complete its line of carburetor parts, which were sold to carburetor rebuilders and kit assemblers (Insalaco 886, 904-05). 241. When PACCO began to offer carburetor kits in the 1950' , one ofthe firms selling carburetor overhaul kits was Sherman Carburetor Company (Carlson 2535). In the mid 1960' , Sherman stopped se1Jing kits (Carlson 2539). The record does not disclose the reason for Sher- 10lH"",. H'Ytil thp bt.p 70' s. Sherman nei- 410 Initial Decision ther packaged nor sold carburetor kits. It was however, a leading manufacturer of carburetor parts for sale in bulk to carburetor rebuilders and kit packagers.

242. In 1978 or 1979, Sherman was bought by John Roberts (Carlson 2539; Insalaco 887). Mr. Roberts had no prior experience with carburetors (Insalaco 888-89).

243. In early 1981, Sherman made a decision to enter the kit market (Insalaco 920). Sherman had been assembling a small number of kits for export into Mexico since at least 1979 (Insalaco 919; CX 457C) and hired additional personnel to assist with its carburetor kit program (Insalaco 974). As a well-established manufacturer of replacement carburetor parts, it was in a good position to enter the kit market (Nelson 1649). In November of 1981, Sherman began soliciting kit sales (Insalaco 888). Angelo Insalaco, a sales representative for Sher- IDan with many years of experience in the automotive aftermarket was placed in charge of Sherman s kit sales (Insalaco 895). Sherman first approached national account customers hoping to be a second source of carburetor kits to the private brand sellers (Insalaco 928- 30). (54) 244. The original line assembled by Sherman was a short line of some 120 kits limited to fast moving kits (Insalaco 929). The kits were exact copies of the Ballwin/Washington kits, and were priced about 20% lower (Insalaco 929). Sherman did not offer a catalog (Insalaco 930). The line was intended to be used as a free-standing short line or as a second source to fil out a line from another kit supplier (Insalaco 930). Sherman intended to expand to a full line in a few years (Insalaco 932).

245. The Sherman kits are sold only as a private label brand offered to national accounts (Insalaco 932). One reason for this decision was that Sherman decided it was possible to sell to national accounts without ofiering a catalog or other services, because they could provide their own catalogs and services, but sales to individual WDs would require such services (Insalaco 939-40). 246. Sherman s first sale of kits was in February 1982, three months after the decision was made to begin assembling kits and two months after Insalaco began soliciting sales (Insalaco 935). The sale was to Sorensen, which purchased 3 000 to 8 000 kits per month (Insalaco 935-36).

247. In the next few months, Sherman obtained the Can tire account which purchased about 1 000 kits per month (Insalaco 935). American Parts System also purchased about $26 000 of kits (Castagna 3748- 49). Sherman also made a sale of unspecified size to Niehoff (Eaton 2951-53). In fact, within two years of the decision to assemble kits Initial Decision 105 F. Sherman had made some sales of kits to most of Ballwin/Washington s national accounts (Timberlake 3131). 248. Sometime in 1981, around the time that Sherman first decided to assemble kits, it assembled on contract 50 000 Easy Way kits for Allparts (Carlson 2553).

249. Sherman also began soliciting sales to Bumper-to-Bumper BTB") and Pronto, two program distribution groups in June. or July 1982 (Insalaco 937). At least one sale was made to Pronto but the record does not disclose the amount of Pronto s purchases. 250. The BTB program distribution group is ten years old, and has 15 domestic WDs and one Canadian WD member (Brown 3384; Bull 3635-A). BTB contracted with Sherman to buy a short line of 50-0 kits bearing the Rockhill brand name. The 50-60 kits wil provide 80% coverage, domestic and import (Brown 3391-92). It is expected that for the present the BTB members wil carry the Rockhill (i. Sherman) short line of fast moving kits, plus a full line of kits with a different brand name (Bull 3683-85). Sherman has not included catalogs or a return privilege in its sales to BTB, though catalogs wil be provided when Sherman expands to a full line (Bull 3686-87). (55) 251. BTB has annual kit sales of about $1.2 or $1.3 million (Bull 3675). Mr. Bull, its Vice President and General Manager, estimates an average cost of slightly less than $5/kit which would produce his WD an annual kit volume of 30 000 to 35 000 kits (Bull 3665). Using $5/kit as an average, BTB WDs have sales of 240 000 or 260 000 kits. BTB, as a whole, wil probably purchase 10% to 15% of its kits from Sherman (24 000 to 39 000 kits per year) (Bull 3697-98). 252. However, two years after it entered the kit market, Sherman total monthly sales of carburetor kits amounted to no more than $16 000 and the company was losing money (Nelson 2266; Willg 4899 900).

253. During the summer of 1983, ACF Industries ("Carter ), a kit assembler which ranked fifth in the kit market in 1980, decided to cease its kit assembling operation and become a reseller (Sheehan 528-29). After Sherman and Ballwin/Washington made sales presentations, Carter chose Sherman as its supplier of private label kit requirements. A former Carter employee testified that Carter chose Sherman over Ballwin/Washington mainly because Sherman quoted a significantly lower price (9% lower J (Lehman 5016). Carter was expected to complete the transition from a kit assembler to a reseller sometime in 1984 (Sheehan 536-37).

254. There is testimony in the record that Sherman was at one time threatened with parts cut-off by Ballwin/Washington (Insalaco 975- 76) and that both Sherman and Carter entertained some doubt as to whether Ballwin/Washington would continue to supply Sherman DL-llllll lVANU!,' ACTURING CO., ET AL. 465 410 Initial Decision with carburetor parts needed to m'l out the kit line (Lehman 5032-33). There is also testimony that at the time Sherman was buying about 20% of its carburetor parts requirements from Ballwin/Washington (Lehman 5033) and that it would take Sherman "a couple of years to be able to offer a full line of kits on its own (Insalaco 932). The record shows that Sherman s dependence on Ballwin/Washington parts has been reduced to about 10% (Lehman 5034). 255. Ballwin/Washington became aware of Sherman s entry almost immediately (Eaton 2954). Throughout 1982 Ballwi'l/Washington monitored Sherman s prices. When a new price list came out in late 1982, Ballwin/Washington held down its price increases on a number of kits to remain competitive with Sherman s lower prices (Eaton 2954-56). As Sherman became more successful, Ballwin/Washington reduced prices on some kits. In August 1983 (shortly after Carter decided to buy its kits from Sherman), Ballwin/Washington lowered the prices on 19 of its fastest moving kits, ranging from 5% to 30% for each kit (Timberlake 3098-99; Lehman 5040-1). 256. From the foregoing, it appears that Sherman has made a good start towards becoming an alternative source of supply of carburetor kits to the private brand sellers and that its (56) position wjJ become more secure as it attains greater degree of self-suffciency with respect to carburetor parts.

257. Sherman s attempts to sell carburetor kits to program distributors have also made impressive gains. However, it has not been able to compete with the leading firms across the board mainly due to its lack of a full line and its inability to offer marketing services, such as catalogs and promotional aids, to WDs at the present time. 258. Allparts, Inc. C'Allparts ) was formed in 1969 by Merton Carlson, who had founded Precision Automotive Components Company P ACCO") and conceived what is known today as carburetor tune-up kits (Carlson 2525, 2527-28). PACCO was sold to Borg-Warner in 1966 (Carlson 2526).

259. In 1973, Allparts became a reseller of carburetor kits assembled by the Ballwin/Washington Division of Borg-Warner (Carlson 2546). For the past ten years, Allparts has sold Ballwin/Washington line of carburetor kits under the "Auto-Mech" trade name, which was owned by Borg-Warner until the challenged transaction and is now owned by Echlin (Carlson 2602). Allparts' continued use of the "Auto- Mech" trade name is contingent on Echlin s willingness to renew its annual contract with Allparts (Carlson 2603-04). 260. Allparts has never assembled carburetor tune-up kits similar to what it has been buying from Ballwin/Washington since 1973 (Carlson 2573). Mr. Carlson testified that Allparts has never considered assembling a line of kits similar to the one it currently resells :

Initial Decisjon 105 F. in part because Allparts does not have a suffcient volume of sales of the Auto-Mech kits (Carlson 2575-76). In 1980, Allparts resold approximately (''' ) carburetor kits purchased from Borg-Warner (CX 464A in camera). These sales represented approximately 1. 16% of all carburetor kits sold in the United States in 1980 (CX 530A). 261. In 1980 or 1981, Al Stefan, a former car mechanic, designed a new kit line called Easy Way (Carlson 2549, 2551). The Easy Way line has 108 kits in it (Carlson 2551). Because the line is highly consolidat- , it has the same coverage as the 380-kit Auto-Mech line purchased from Ballwin/Washington (Carlson 2551).

262. The Easy Way line was originally purchased from Ballwinl Washington and Sherman (Carlson 2552-53). The Easy Way line was fully designed and consolidated by the reseller, and the assemblers merely carried out the packaging function (Carlson 2549, 2552). The first order was for 50 000 kits from each of Sherman and Ballwinl Washington (Carlson 2553).

263. After the first order, Allparts purchased the Easy Way line kits from Sherman for about one year (Carlson 2555). In late 1982, Allparts decided to assemble the line itself(Carlson (57) 2555). One ofthe factors which induced this decision was that Sherman increased its packaging fee (Carlson 2556).

264. The Easy Way kits are assembled in the basement and first floor of Mr. Carlson s home (Carlson 2558-59). Mr. Carlson s description of the facilities used in this packaging is indicative of the ease low cost and short time span involved in beginning a modest kit assembling operation (Carlson 2562-63). Using rudimentary facilities, two employees earning $4/hour are able to package 4 000 kits per week, or an annual rate of about 200 000 kits (Carlson 2562, 2564). 265. Allparts has no sales employees but relies on independent manufacturer representatives (Carlson 2564) as does Tomco. This is a satisfactory arrangement for Allparts and helps to reduce costs (Carlson 2564).

266. The market acceptance of Easy Way kits appears to have been limited. Easy Way kits contain fewer parts than standard carburetor tune-up kits (Carlson 2551). They contain fewer gaskets and small parts, no economizer or power valves, no flange gaskets, and in some instances, only a replacement rubber cup rather than an entire pump plunger assembly (Carlson 2589-90). There are only 108 Easy Way kits as compared to 350-380 kits in the standard carburetor tune-up kit line that Allparts resells (Carlson 2551). Allparts has attempted to market Easy Way kits primarily to mass merchandisers, chain stores and discount houses (Carlson 2552, 2563). 267. ("' J (CX 162Y in camera).

- -- -r .. "'nn ""F" l"t:u;:.plJp.r of the standard line of carburetor .

.LU' UH .n'-'i.'-J , DJ. .nL1. 410 Initial Decision tune-up kits assembled by Ba.llwin/Washington, inquired whether Ballwin/Washington would be interested in producing kits similar to the Easy Way kits for resale by Sorensen (Eaton 2977). Mr. Eaton Ballwin/Washington s Director of Sales and Marketing, informed Sorensen that it would not be economical for Sorensen to sell this type of kit or for Ballwin/Washington to assemble it (Eaton 2977). Mr. Eaton also testified that in his opinion, the Easy Way kit concept was not a proper marketing approach and that the traditional channel of distribution, through WDs, would not be interested in this type of kit (Eaton 2977-78).

269. In March of 1983, Mr. Merz, then President of the BWD subsidiary of Echlin, testified that BWD had no intention of marketing an Easy Way-type kit because the concept had not been successful and demand for these kits was extremely limited (CX 535Z-15 through Z-17). Mr. Merz explained that the Easy Way (58) kits were directed at mass merchandisers, rather than WDs. Mr. Carlson testified that many mass merchandisers (such as Montgomery Ward, K-Mart Woo\co, Sears and J.C. Penney) did not carry carburetor kits and that in tact, a number of these companies had carried carburetor kits in the past but had discontinued carrying them (Carlson 2586-87). 270. AJlparts began selling Easy Way kits in 1981 but its volume of sales in both 1981 and 1982 was far lower than the company s projections (Carlson 2592). Mr. Carlson agreed that the sales of Easy Way kits were "very bad, compared to what they should have been" (Carlson 2605). In fact, sales of Easy Way kits in 1982 amounted to only 660 kits. Allparts ' most hopeful projections for sales in 1983 and 1984 were 42 000 and 44 000 kits, respectively U-'c fan 2689). Thus even assuming that total U.S. sales of carburetor kits remained steady between 1980 and 1984, Allparts' projected 1984 sales of Easy Way kits, four years after development of the Easy Way kit line began, would account for no more than one-third on % ofall carburetor kits sold in the United States (Glassman 4498). 271. The Allparts/Easy Way experience shows that a small firm on a modest budget can design a line of kits, have another firm package it or take over the packaging function itself and become a viable business, albeit not as a full-fledged competitor. 272. Many of the witnesses who testified in this proceeding have been involved with the packaging and sale of carburetor kits since the inception of the consolidated kit in the 1950's and 1960's. The majority ofthese witnesses testified that entry in the carburetor kit business in the early days was simple. They also expressed their view that entry at the present time would be no more ditlcult, and in some ways easier, than it was 25 years ago (see RPF 974-1001). A manufacturer Initial Decision 105 F. of gaskets used in carburetor kits expressed a contrary opinion (Secrest 1119-20).

273. Thus, the evidence shows that entry barriers to the assembly and sale of kits are very low. Although the recent entry history is somewhat mixed, the relative paucity of substantial entries does not necessarily imply high entry barriers. See The Grand Union Co. Docket 9121 (July 18, 1983) at slip op. 46 (102 F. C. at 1063). Rather the record as a whole is consistent with the view that the mixed history of recent entries at the assembly level may be due to the relatively small size of the kit market ($53 milion) and the presence of some ten assemblers in the market as well as to suffciently attractive profit opportunities enjoyed by the private brand sellers of purchased kits (see F. 152, 213; Sheehan 529-31). (59) D. The Evidence Does Not Show A Trend To Increasing Concentration Or Undue Reduction In The Number Of Sellers 274. From the small number of industry members discussed in VLA. hereinabove, and the market share table in evidence (CX 530A- C), one might conclude that the number of sellers in the kit market is small and that the challenged acquisition brought about a further reduction in the number of sellers, approaching the critical threshold of 10 to 12 (see Areeda & Turner IV Antitrust Law, n 911 a and b at 60-62).

275. However, because of the existence of some 40 to 50 resellers the carburetor kit market has not experienced an undue reduction in the number of sellers (see RX's 292-307, 310- , 317- , 282). 276. In addition, the evidence fails to show a tendency towards increasing concentration. The record is bare of any direct evidence showing the direction of concentration over time one way or the other. However, it is fair to say that over the last decade, the number of carburetor kit assemblers has remained more or less constant, while the ranks ofresellers (or private brand sellers) expanded. Also during the late 1970's and early 1980' , a number of firms have entered, either as assemblers or resellers. The market penetration attained by these entrants, however, appears to be rather modest. 277. The market share evidence in the record is limited to two years and does not permit an informed determination of the market-share stability issue.

E. Competition From Imports 278. The evidence shows that imports of carburetor kits are small and that the demand for carburetor kits for import cars are largely met by domestic kit assemblers (see F. 191 supra). 279. However, the evidence also shows that import cars have ac- :+ ECHLIN MANUFACTURING CO.. ET AL. 469 410 Initial Decision counted for an increasing portion of new cars in recent years and that some Japanese car manufacturers have begun to bring into this Country carburetor kits to be sold through their dealer organizations and Japanese trading companies (see F. 192- supra). Therefore, it is reasonable to conclude that import competition wil increase with respect to kits for import car application. (60) F. The Carburetor Kit Market Is Undergoing A Radical Technological Transformation 280. A rapid technological change is another market factor relevant to the assessment of market power effects of a merger for market power may be harder to exercise or less likely to endure in the face of such a change. FTC Guidelines at 4- 281. Due to adoption and installation of the so-called TEl in the place of conventional carburetors in increasing number of new vehicles in recent years, the carburetor kit market is faced with a gradual diminution and transformation during the coming decades. Although the evidence indicates that the kit assemblers would be able to make the necessary adjustment and remain viable factors in the TEl tuneup kit business, this prospect is unsettJing to kit assemblers and provides significant market dynamics (see F. 75- supra ). To that extent, market power in the kit market is less likely to endure and may be harder to exercise.

VIII. EVIDENCE RELATING TO MARKET PERFORMANCE A. Product Innovation And Improvement 282. The evidence shows that the record of product improvement and innovation in the carburetor kit market has been satisfactory. 283. The evidence shows that the time lag between the appearance of a new carburetor model on new vehicles and the introduction of carburetor tune-up kits for that model in the aftermarket is about 12 months (Merz 2723-24; Rivet 2824).

284. The evidence also shows a series of significant product improvements over the years, including incorporation of viton parts in the needle and seat subassembly and introduction of highly fuel resistant elastomer components in pump plungers and diaphragms (see F. 119, 136 supra ). In recent years, a lew kit assemblers have also introduced TEl tune-up products in response to incorporation ofTBIs in some new vehicles.

285. The evidence also shows that the record of kit consolidation which increases distribution effciency as well as user convenience has been rather remarkable in the assembly and sale of carburetor +-;",.. r :._.. , Initial Decision 105 FTC. B. Price Competition 286. Ballwin/Washington did not publish "list prices" to WDs as such, although it published suggested jobber net prices, from which WD prices could be derived. Other assemblers published price lists which they seldom discounted (Sheehan 601-03; Tehansky 856-57). The record is silent as to price competition among the numerous private brand sellers. However, the record as a whole does not reflect vigorous price competition among competing sellers, and competition is rather waged in non-price terms, such as completeness of the kit line and various marketing aids generally provided by kit assemblers (see F. 156-73 supra 287. ("') (Nelson 1503-04; CX's 28Z- 12 in camera 29Y in camera 30Z-9 in camera 162W in camera C. The Evidence Fails To Show Supra-Competitive Profits In The Kit Market 288. Complaint counsel assert that "profitabiliy" in the assembly and sale of carburetor kits is "high" and "well above competitive benchmarks" (see CPF 243-65). However, the record is devoid of any empirical data directly addressing the long-run profit levels of the carburetor kit assembly market that will permit a determination of whether the relevant product market has enjoyed supra-competitive profits over time, which may be indicative of noncompetitive market performance.

289. Complaint counsel largely rely on the opinion testimony of Dr. Nelson who used CX's 543 and 544 for the purpose of reconstructing the accounting profits of a hypothetical kit assembler at the 1.5 million unit level and concluded that the profits likely to be realized by leading firms" in the assembly and sale of carburetor kits are substantially higher than the QFR figures for "all manufacturing" for any year during the 1978-1982 period. However, Dr. Nelson s analysis of a hypothetical profit model and conclusions derived therefrom essentially lack that degree of probity which will support a finding on such a key issue as the market's long-run profit levels. 290. ("' ) (e. , Secrest 1114; Nelson 1511- , 2303-05, 5143- 5286-87; CX's 162Z- in camera, 618-22 in camera). However, such evidence, while suggestive, is fragmentary and falls far short of what would be required for an informed determination of whether the profit levels of the kit assembly market over time have been (62) consistent with a competitive performance or the contrary is the case. 291. The record also contains testimony of'some knowledgeable L__,,__-- n_ .6u__--- "-L_ l-- 1 ' 1 . ECHLIN MANUYACTUKU ":r LV. 1 1"1..

410 Initial Decision profitable venture compared to some other product lines in the aftermarket (see RPF 860-63).

D. The Smaller Firms Are Profitable And Appear To Be Able To Grow 292. The evidence shows that the smaller firms in the kit market such as Carter, Tomco and Holley, have been profitable and able to grow (see Thompson 310, 369; Sheehan 557; Jursek 738, 777-78; CX 530). And the recent exit of Carter from kit assembly is not related to any adverse market conditions Carter faced in the kit. market (Sheehan 527-31). Thus, the record is consistent with competitive performance of the kit market as a whole.

IX. THE EVIDENCE RELATING TO FACTORS OTHER THAN MARKET SHARES IS SUFFICIENT TO OVERCOME THE SUBSTANTIAL ANTICOMPETITIVE POTENTIAL OF THE ACQUISITION INFERRED FROM MARKET SHARE EVIDENCE 293. The anticompetitive potential of the challenged acquisition is substantial as predicted by the combined market shares ofEchlin and Borg-Warner in the carburetor kit market. In such a case, other non-market share factors may be given less weight. See FTC Guidelines at 6.

294. In the instant case, however, the administrative law judge is persuaded that the nature and quantum of the record evidence related to market factors other than seller concentration, including low entry barriers (F. 215-73, supra ), to factors affecting the significance of market shares (F. 207- supra) and to the radical technological change brought on by the throttle body injection system (TBIJ which promises to make carburetors and carburetor kits obsolete in a decade (F. 75- supra), collectively and cumulatively, are sufficient to overcome the adverse inference based on market share evidence. Cf, United States v. General Dynamics Corp. 415 U.S. 486, 501-02 (1974). 295. Of the evidence generally recited in the preceding finding, some merits particular emphasis. First, over (" ' ) of Borg-Warner 1979 kit market share and almost (" ' J of its 1980 share were "national account" sales to resellers, Borg-Warner s kit sales under the Borg- Warner name accounting for the remainder in both years. As for Echlin, it was the only one, among the top seven kit assemblers, which lacked any parts manufacturing (63) capability; aU the others manufactured some or most of the components that made up the carburetor kits they sold. By acquiring the aftermarket divisions from Borg- Warner, Echlin, among other things, acquired that capability. In this perspective, the substantial anticompetitive potential predicted by Initial Decision 105 F. the combined market shares ofthe merging parties must be discounted accordingly.

296. The second important factor which merits particular emphasis is the condition of entry. The record is clear that entry barriers into the assembly and sale of carburetor kits are virtually nonexistent or very low.

297. The third factor which merits some emphasis is tpat direct competition between Borg-Warner and Echlin was not as large as the pro forma market shares of the merging firms may suggest. Echlin has never sold, nor tried to sell, any carburetor kits to any customer other than NAPA, a large program distributor group (F. 6- supra). While APD sold to the traditional channels of aftermarket distribution of which NAPA is a part, Ballwin/Washington did not sell directly to the WD/jobber channel but only to national accounts (which sold to the WD/jobber channel).

298. From the foregoing, it is safe to conclude that the long-term anticompetitive effect ofthe challenged acquisition is not likely to be substantial. With respect to short-term efiects, the question is a closer one. However, the non-market share factors discussed hereinabove including the significant restraining force emanating from rebuilt carburetors, together with ease of entry coupled with the presence of numerous potential entrants, including Ford, Chrysler, AMC and other large resellers of kits (F. 150-51 supra), are sufficient to check any adverse short-term effect which may be inferred from the market share evidence.

299. Accordingly, on the basis of the record as a whole, it is found that the efiect of the challenged acquisition is not likely to lessen competition substantially in the assembly and sale of carburetor kits. (64) DISCUSSION This case involves the 1981 acquisition by Echlin, a manufacturer and marketer of a wide range of automotive aftermarket products, of the assets and business of Borg-Warner s five aftermarket divisions, including Automotive Parts Division CAPD") and Ballwin/Washington Division ("Ballwin/Washington ). Echlin also obtained from Borg-Warner a license to sell aftermarket products of the acquired divisions under the Borg-Warner name and a supply agreement covering certain Borg-Warner products Echlin may need for the continued operation of the acquired divisions.

The complaint challenges the acquisition because of its alleged anticompetitive effect in the carburetor kit segment of the automotive aftermarket and seeks the divestiture of the acquired divisions hv ECHLIN MANUFACTURING CO., ET AL. 473 initial- Decision- -410 Echlin or, in the alternative, rescission of the entire transaction. The domestic sal s of carburetor kits of the acquired divisions (through APD and Ba\lwin/Washington) amounted to about 13% of the total sales of the five aftermarket divisions acquired by Echlin from Borg- Warner.

A. The Product Market A carburetor kit is a kit or package containing those carburetor components which are most subject to wear and need replacement most often, plus some gaskets and gauges and an instruction sheet. Carburetor kits are used by automotive mechanics to repair or tuneup malfunctioning automotive carburetors which do not require carburetor replacement.

The 1980 total sales of carburetor kits in the United States were about $53 milion at the wholesale level (to warehouse distributors). Ballwin/Washington, one of the acquired Borg-Warner divisions manufactured carburetor parts and assembled and sold carburetor kits to about ( ) national account customers, which sold the kits under their own brand names to the traditional warehouse dealer jobber distribution channel. Ballwin/Washington also sold kits to APD, its sister division serving as Borg-Warner s principal aftermarket product marketing arm. APD sold kits to the warehouse-jobber channel in competition with other national account customers of Ballwin/Washington and kit assemblers.

Echlin purchased carburetor parts and assembled and sold carburetor kits to NAPA distribution centers. Echlin s kit sales accounted for less than 2% of Echlin s total U.S. sales of aftermarket products. The complaint defines the relevant product market as "the assembly and sale of carburetor kits" (Comp. n 3). Respondents' (65) attack on the alleged product market is twofold. First, respondents argue that carburetor parts and new and rebuilt carburetors are competitive products and belong in the same product market with carburetor kits. Second, they argue that carburetor kit "assembly" is a euphemism for a packaging function which lacks any economic significance and results in excluding from the product market competitive sources of kit supply, namely, some l**' ) private brand sellers of kits which do not "assemble" kits but sell purchased kits in competition with assemblers." In my view, the evidence shows that the "assembly and sale of carburetor kits" is a valid product market in which to assess the competitive effect of the challenged acquisition. Individually packaged carburetor parts can of course be used tc tune-up a carburetor instead of using a kit containing these parts They are functional substitutes and, in theory, they are interchanges hlQ nroriucts. However, trade realities show that separately package, Initial Decision 105 F. parts are not practical substitutes for kits. The evidence shows that the availability of individually packaged carburetor parts are limited and their sales are de minimis. This is not surprising because carburetor kits came into their own by offering a convenient package which can be used to tune-up carburetors of different makes and models. Thus, individual parts do not belong in the same product market with kits. Also, sales of carburetor parts to warehouse dealers are de minimis and their inclusion or exclusion will not make any significant difference to the outcome of this case.

Replacement carburetors are functional substitutes for kits in the sense that a sluggish carburetor can always be replaced with a new or rebuilt carburetor instead oftuning up the old carburetor using a kit. However, kits are designed for use in carburetor tune-ups only and cannot be used when the carburetor malfunction is due to structural damages and requires carburetor replacement. Also, to the extent that replacement carburetors can be said to be functional substitutes of carburetor kits, replacement carburetors are much more expensive than kits and do not offer an economically reasonable alternative to kits, even when labor costs to the car owner are taken into consideration.

Most importantly, the evidence shows that there is no price sensitivity between replacement carburetors and carburetor kits, although the two product groups are generally regarded as competitive products. Therefore, new and rebuilt replacement carburetors are not close-enough substitutes for kits to be included in the same product market with carburetor kits.

Finally, whether it is called "assembly" or "packaging, " the economic function that an assembler performs in creating a kit out of bulk components is essentially a production function, however simple. Furthermore, to suggest that a kit assembler does no more than package selected carburetor parts into a kit is to grossly mislead. A kit assembler s roles in developing and maintaining a line of kits from year to year and in providing the (66) many important marketing 3ervices to private brand customers (resellers) and to warehouse dealer/jobber customers are economically significant. These functions colectively set an assembler apart from a mere reseller and confer upon he assembler a degree of market power to which a reseller cannot spare.

The evidence also shows that all of the leading kit assemblers, with Ie notable exception of Echlin (the third-ranking firm)" fabricated house or on contract many of the key kit components, such as aphragms, valves and needle-and-seat subassembly. However, all of e leading kit producers "sourced" other parts needed to make up os and no firm, including the OE carburetor manufacturPN n'..- ECHLIN MANUFACTURING CO., ET AL. 475 Initial-- Decision-410 completely self-suffcient. Rather, the economic significance of backward integration in the kit market appears to be that the more selfsuffcient a firm is in parts, the greater is its control over kit production and product quality. In this sense, parts production capability is an important competitive advantage for a kit assembler. In sum parts-fabrication is not an essential firm function in the production and sale of kits but assembly is. Therefore, the "assembly and sale of carburetor kits" is an appropriate product market for the purposes ofthisB. Market Sharecase.And Concentration Evidence In the measurement of market shares and market structure analysis, the administrative law judge adopted complaint counsel's methodology and included in Borg-Warner s kit sales the unit sales of both APD and Ballwin/Washington divisions. It is true that Borg-Warner sold about ("*j of kits it assembled (at Ballwin/Washington) to private brand sellers and other kit assemblers and sold the remaining (*"j under the Borg-Warner name to the warehouse dealer/jobber channel of distribution through APD. However, it is settled that, once the product market is determined, as is the case here, it is appropriate for purposes of market share analysis to include in a firm s market share its sales to private brand sellers (or resellers). See United States v. Black Decker Mfg. Co., 430 Supp. 729, 737-38 (D. Md. 1976); Beatrice Foods Co., 3 Trade Reg. Rep. (CCH) TI 22 035 at 22 624 (May 26 , 1983); Champion Spark Plug Co., D.9141, Initial Decision at 90 n. 10, adopted by the Commission (Final Order dated June 20, 1984) (103 F. C. at 623). In the instant case, Borg-Warner s kit market share includes all kits it assembled and sold regardless of whether they are sold under the Borg-Warner label or not as long as the kits were marketed, either directly or through middlemen, to the same channels of aftermarket distribution. Borg-Warner likewise regarded its kit sales to resellers as a part of its "base market share" in the kit market. (67) Prior to the acquisition, Borg-Warner was the top-ranking assembler and seller of carburetor kits and accounted for about 37.7% ofthe kit market in 1979 and 36.4% in 1980. Echlin was the third largest firm, accounting for about 10.1 % in 1979 and 10.4% in 1980. As a result of the acquisition, Echlin became the top-ranking firm, with a pro forma 1980 share of about 46.8%.

The pre-acquisition four-firm concentration ratio of about 78.2% in 1979 increased to about 85.6% in 1980 on a pro forma basis. The Herfindahl-Hirschman Index ("HHI") increased from about 2 172 ir 1980 to about 2 929 after the acquisition on a pro forma basis. Thus, as a result of the acquisition, Echlin became by far the larges ).

Initial Decision 105 F. assembler and seller of kits and the acquisition further exacerbated the already high concentration in the carburetor kit market. C. The Effects Of The Acquisition Under established judicial and enforcement criteria, the resulting market shares ofthe acquisition and the degree of market concentration are so high that the acquisition wil be a presumptive violation ofthe merger law and be prohibited unless it is shown that the market shares are not reliable indicia of the true competitive significance of the acquisition or that the market is likely to perform competitively regardless of the acquisition. See IV Areeda & Turner, Antitrust Law nn 909-912 at 29-68 (1980); Sullivan, Antitrust, Sections 204a and 204b at 613-21 (1977); BASF Wyandotte Corp. 100 F. C. 261, 392 (Initial Decision dated May 14, 1982).

In the instant case, the administrative law judge wil be guided by the 1982 Statement of Federal Trade Commission Concerning Horizontal Mergers dated June 14 1982 ("FTC Guidelines ) and will also take into consideration the recently revised Department of Justice Merger Guidelines (June 14, 1984) ("DOJ Guidelines Also see Clanton Focusing the Inquiry: Specificity In The Merger Guidelines and Elsewhere 71 Calif. L. Rev. 430, 433-36 (1983); Greenfield Beyond Herfindahl: Non-Structural Elements of Merger Analysis 53 Antitrust Law J. 299 (1984).

The 1982 FTC Guidelines indicated the Commission s view that while market share data remain "an important indicium ofthe likely competitive effects of a merger " it wil employ "a more refined treatment" of such data in light of "more recent empirical economic research and well over a decade of practical experience in analyzing and evaluating horizontal mergers" and give greater consideration to nonmarket share evidence, the most important of which being that of entry barriers. See FTC Guidelines at 2-3. (68) In the instant case, several important factors diminish the significance of market shares as a surrogate measure of the merging firms market power. First although new and rebuilt carburetors are not close-enough substitutes for carburetor kits to be included in the same market with kits, they are generally competitive products and rebuilt carburetors wil have a significant restraining influence upon the 'larket power of the resulting firm (Echlin). See Greenfield Beyond 'lerfindahl: Non- Structural Elements of Merger Analysis, 53 Antirust Law J. 299 (1984).

Secondly, the power of kit assemblers to control the product quality, rice or output in the kit market is limited. Although the kit sales of n assembler to private brand sellers (or resellers) were included in le assembler s sales for the purpose of market share analysis, the , ( ECHLIN M Initial Decision ::0., ET AL. 477 410 Complaint evidence is clear that the resellers ) are generally able to pursue' substantially independent competitive strategies with respect to product quality and price in competition with kit assemblers. Thirdly, and perhaps most importantly, entry barriers are virtually non-existent or very low. Although there has been only one substantial entry during the past decade, there has been other smaller entries. In any event, the paucity of new entrants do not necessarily show high entry barriers. Grand Union Co., D. 9121 (July 18, 1983) at slip op. 46 (102 F. C. at 1063). Rather, when viewed against the evidence showing low entry barriers, the record is consistent with the view that paucity of new entry into the assembly of kits may be due to the relatively small size of the market (1980 sales of about $53 milion) and the presence of some ten assemblers in the market, as well as to the suffciently attractive profit opportunities enjoyed by private brand sellers of purchased kits.

In my view, these factors clearly show that there are important factors at work to limit the exercise and the duration of the market power by the leading firms in the kit market and that the market shares are not reliable indicators of the true competitive significance of the acquisition.

Furthermore, several elements in the non-market share phase of the kit market suggest that the industry has performed in a fashion consistent with competition and is likely to do so in the future. Although the evidence does not show a vigorous price competition a lively competition is waged in terms of product innovation and improvement as well as in a wide range of customer services. Also, there is no credible evidence to conclude that the leading firms have enjoyed supra-competitive profits in recent years. And the record shows that smaller firms, albeit small in number, are profitable anc able to grow. (69) Finally, due to recent incorporation of the throttle body injectio ("TBI") system (in the place of conventional carburetors) in increasir numbers of new cars, the carburetor kit market is undergoing a ra( cal technological change and faces a gradual decline in the next t years. In the meantime, there is reason to believe that as the numl of imported cars increases so wil competition from imported car retor kits.

Therefore, it is concluded that complaint counsel have faile' show by a preponderance of credible evidence that the effect of challenged acquisition is likely to lessen competition substantial - '-1" nd sale of carburetor kits. .. , Initial Decision 105 F. CONCLUSIONS OF LAW 1. The Federal Trade Commission has jurisdiction over the respondents and the subject matter of this proceeding. 2. Respondent Echlin Inc. ("Echlin ) is a Connecticut corporation with its headquarters in Branford, Connecticut. 3. Respondent Borg-Warner Corporation ("Borg-Warner ) is a Delaware corporation with its principal offce in Chicago, Ilinois. 4. At all times relevant to this proceeding, respondents were engaged in commerce, or their acts and practices were in or affecting commerce, as "commerce" is defined in the Clayton Act, as amended, and in the Federal Trade Commission Act, as amended. 5. The appropriate product market within which to evaluate the competitive effects of the acquisition of Borg-Warner s automotive aftermarket assets by Echlin is the assembly and sale of carburetor kits (carburetor kit market).

6. The appropriate geographic market within which to evaluate the competitive effects ofthat acquisition is the United States as a whole. 7. The carburetor kit market is highly concentrated. 8. In 1980, Borg-Warner accounted for about 36.4% and Echlin, for 10.4%, of the carburetor kit market.

9. The acquisition produced a top-ranking firm with a combined 1980 share of 46. 8% on a pro forma basis, and the anticompetitive potential of the acquisition is substantial. 10. The acquisition eliminated the direct competition between Echin and Borg-Warner. (70) 11. The record evidence concerning various non-market share fac- )rs and market performance is suffcient to overcome the substantial lticompetitive potential of the acquisition inferred from market are evidence alone. Such factors include the following, among lers:

) Substitute products, such as new and rebuilt replacement carbu- )fS, compete with carburetor kits;

.) A large number ((" ' JJ of resellers of kits (private brand sellers urchased kits) compete with kit assemblers; The power of kit assemblers to control competition in the kit ;et is significantly limited;

Entry barriers into the assembly and sale of kits are very low; The kit market is undergoing a radical technological change; he kit market is likely to face increasing import competition. here is lively competition in terms of product innovation and element as well as a wide range of customer servi('p ECHLIN MANUFACTURING CO. , ET AL.

410 Opini by the kit assemblers, although a vigorous price competition is ribt eviden t.

(h) The evidence does not show that the market is characterized by supra-competitive profit levels over time; (i) The smaller kit assemblers are profitable and appear to be ableto grow. 12. Complaint counsel have failed to establish, by a preponderance of credible evidence, that the acquisition by Echlin of Borg-Warner automotive aftermarket assets is a violation of Section 7 o(the Clayton Act or Section 5 of the Federal Trade Commission Act, as amend- , as alleged in the complaint.

Accordingly, the following order wil be entered. (71) ORDER It is ordered That the complaint be, and the same hereby is, dismissed.

OPINION OF THE COMMISSION By DOUGLAS Commissioner:

The complaint in this matter alleges that respondents The Echlin Manufacturing Company and Borg-Warner Corporation have violated Section 7 of the Clayton Act, 15 U. c. 18 (1982), and Section 5 of the Federal Trade Commission Act, 15 U. c. 45 (1982). The alleged offense is Echlin s acquisition of Borg-Warner s automotive aftermarket operations, which manufacture, assemble, and sell automotive parts that are used to replace original equipment on automobiles. This acquisition is said to have posed a likelihood of substantially lessening competition in the assembly and sale of carburetor kits. ALJ Montgomery K. Hyun issued his initial decision on September 1984. He found that the allegations of the complaint had not been proved and therefore ordered that the complaint be dismissed. Complaint counsel appeals that decision; (2) respondent Echlin also ap peals and urges the Commission to "correct" several of the ALJ' 1 In the remainder of this opinion, the (oHowing short forms and abbreviations will be used Initial Decision Finding of Fact No - Complaint Counsel's Exhibit No I.D,J. The Administrative Law.JudgeexAl...Echlin The Echlin Manufacturing Company Borg.Wamer Borg.Warner Corporation Sherman Carburetor Company Sherman - Carter Automotive Division of ACF Industries, loc Carter FTC Statement - ITC Statement on Horizontal Mergers (,June 14, 1982) DO.J Guidelines - Justice Department Merger GuideJhws (June 14, 1984) Traoscript citations wi!! be given as the last name of the witness and the page on which the testimony apt: Opinion - 105 F.T.

findings. No appeal was fied by or with regard to respondent Borg- Warner.

We affrm. The ALJ conducted an exhaustive analysis of the relevant product market, various quantitative measures of concentration and a number of qualitative considerations bearing on the likelihood of anticompetitive em,cts caused by the acquisition. We wil not address all the issues resolved in the initial decision or raised by the parties on appeal because we have determined that there are no barriers to entry into theassembly and sale of carburetor kits. In the absence of entry barriers, there can be no anticompetitive effect from the acquisition, and no violation of the antitrust laws. This Opinion begins with a brief summary of the relevant facts. We will then outline the analytical framework within which these facts must be viewed, including the significance of barriers to entry and a description ofthose industry characteristics that can constitute entry barriers. Finally, we wil apply that framework to the facts ofthis case to determine whether the assembly and sale of carburetor kits is characterized by significant barriers to entry that could permit the exercise of market power.

I. THE ASSEMBLY AND SALE OF CARBURETOR KITS A carburetor kit is a collection of parts that can be used to "tune " a defective carburetor and return it to optimal performance. It consists of those parts that are most likely to be in need of replacement. I.D.F. 58-68. A given kit may be (3) designed for only a single carburetor model, or it may contain alternative parts that enable it to be used for anyone of several models. I.D. F. 127, 130-36. The carburetor kit industry includes manufacturers of the parts chat are used in the kits, assemblers of the kits themselves, and esellers who purchase kits from some assemblers. These functions, "while theoretically distinct, frequently overlap in practice. The inividual carburetor parts that are contained in kits are manufactured many companies and can be obtained with relative ease. I.D.F. 121 11; Sheehan 585; Jursek 732-36, 810-11; Thompson 2454-55; Carln 2554-55; Bush 2863-77; Schultz 3040-45; Timberlake 3080-83; nith 3178-82. Most assemblers of carburetor kits manufacture at 1st some of the parts used in their kits; before its acquisition of rg-Warner, Echlin was the only major exception to this rule. I.D. 121, 124. Nonetheless, no major assembler manufactures all the ts used in its kits. See I.D.F. 118, 175-88. Most assemblers also buy Ie kits from other assemblers to fill out their lines. I.D. F. 118 e ot1ly aq,'UabJy cootrary testimony was given by a witness who admittedly was never involved in the lics of acquiring parts, bllt who IJO!wthBless was able to identify several alternative source;; for various tor parts- Insalaco 910-- 15, 974-88 ECHLIN MANUFACTUltmc. CM. ' hh.

410 Opinion 175-90. At the extreme end of this continuum arb 'about forty re ellers of carburetor kits, who buy all their kits from assemblers, but sell them under their own labels. I.D.F. 150-55, 195. (4) Assembled carburetor kiis are generally sold to wholesaler-distributors, who sell them to jobbers, who in turn sell them to the ultimate users, automobile mechanics. I.D.F. 142-45. Some kits are sold to automobile dealerships for use in their service facilities. I.D. 146-8. A very few kits are sold to the public through mass merchandisers. I.D.F. 149. In 1980 approximately 13 milion carburetor kits were sold to wholesaler-distributors for about $53 millon. I.D.F. 174 230. The parties are in agreement that the market for carburetor kits is likely to remain stable or to decline through the 1980s. See I.D. 77.

At the time of Echlin s acquisition of Borg-Warner, Borg-Warner was easily the nation s largest assembler of carburetor kits. Through its Ballwin/Washington Division, it manufactured carburetor parts some of which were sold to carburetor rebuilders and other kit assemblers, and assembled carburetor kits, which were sold to resellers, to other assemblers, and through its Automotive Parts Division, to wholesaler-distributors. I.D.F. 14- , 31-47, 175-76. Other large kit assemblers included Echlin, Standard Motor Products, Inc., General Motors Corporation (through its Rochester Division and AC-Delco marketing organization), ACF Industries, Inc. (through its Carter Automotive Division), Tomco, Inc., and Colt Industries Operating Corporation (through its Holley Replacement Parts Division). I.D.F. 1- 177-88. Finally, there were a few very small domestic and foreign assemblers. I.D.F. 189-93. (5) Two former assemblers, Ford Motor Company and Carter, have stopped assembling carburetor kits and now function solely as resellers. I. F. 152, 253; Lingle 3511- , 3521, 3525-26. In 1980 Ford accounted for about six percent of kit sales to wholesaler-distributors while Carter, which had not yet ceased assembling kits, assembled about seven percent of all kits; no other reseller sold more than three percent of all kits. I.D.F. 183; CX 464 (in camera); CX 530. About seventy percent of Borg-Warner s sales of carburetor kits were to resellers or other assemblers. I.D.F. 151, 295. Resellers of carburetor kits include Carter, Ford Motor Company, Chrysler Corporation American Motors Corporation, Volkswagen of North America, Suba- , Wells, and divisions ofTRW and Gulf & Western. Merz 2749-61; CX 464 (in camera). Many resellers participate in the design of their line of carburetor kits. See Baumann 1013-23, 1033- , 1041-44; Secrest 1127-28; Carlson 2552-53; Merz 2724-26; Rivet 2836-39; Eaton 2946-49; Lingle 3518-21; Lehman 5016-19. Many also have established distribution systems. Nelson 1963-65. And at least some resell- Opinion 105 F.

ers already possess the physical facilities needed to assemble carburetor kits. See Baumann 1037-39. Others who do not presently assemble kits perceive little diffculty in entering the market. I.D. 272; Brown 3387-90; Bull 3675-77; see Wilig 4737-44. The physical assembly of carburetor kits is a simple manual process. In essence, it involves placing the right carburetor parts into the right boxes. I.D.F. 78-79, 220. This can be facilitated by the use of a conveyor belt or a "Iazy-Susan-like " (6) tray. I.D.F. 79-81. The ALJ visited the production facilities of one assembler and "was impressed by the simplicity of the process which appeared neither elaborate nor complex." LD.F. 82. The capital investment required to begin assembling kits is not large, Nelson 2323, and consists mainly of equipment that could be used for other purposes, Willig 4778-85. One recent small-scale entrant started by assembling kits in his home. His out-of pocket cost was about five hundred dollars, which paid for "a local carpenter (to J come in and put up tables, hang new lights so they could see better. I would say that plus buying shelving from a local hardware store that went out of business." Although his company s market share remains very low, he testified that it was successful. Carlson 2558-63.

In recent years, at least two foreign and two domestic firms have made small-scale entries into the assembly of carburetor kits for sale in the United States. The two foreign firms are Japanese automobile manufacturers; the two domestic firms began as resellers and expanded into kit assembly. I.D.F. 237- , 258-71. The ALJ found that the small kit assemblers "are profitable and appear to be able to grow. LD.F. 292.

In early 1981 Sherman Carburetor Company, a manufacturer of carburetor parts, began to consider assembling kits. Insalaco 920 926- , 965-66. Sherman made the decision to enter the kit market in November 1981, Insalaco 888, and first solicited sales the following month, Insalaco 929. Its first sale came in February 1982, only three months after it made the decision to enter the market. I.D.F. 246. Within two years, Sherman had (7) made sales to most ofthe resellers although its market share continued to be small. I.D.F. 247 , 252. There is no evidence that Sherman s relatively low level of sales resulted from anything other than its potential customers' preference for Echiin at prevailing prices.3 When Carter stopped assembling kits, J The ALJ fol1nd that. S)1Prrnan was handiCOIpped by its lack of.. full line ilnd its "inability " to offer marketing services, sllch as catalogs and promotional aids- I.D.F. 257;see I.D.F. 128, 224. The need to offer a full lineonly applies, frat all, to ,dcci to whnje';fller- djstrihut.ors; the evidence i.5 dear that reseUers and other assemblers could and oftcn did, buy less t.han 3 fnilline. See I.D.F. 118, 128, J77-8l1, 247 , 253, Moreov, the need to otter a full line is not the same as the ne..d to assemble a full jine. Like most assemblers, Shennan could have purchasp.d additional kits to fil out its line- The !):wd t.o offer marketiI1g aids would also not apply to resellers, who are alrcady in the busioess of marketing carburetor kits, and many of whom prepare their own catalog - ID.F. 245; Baurnarm L020-- , 1036; StefatJ 2672; KruAe 2926; Castagna 3744-46. Sherman was fuJjy capahle of preparing catalogs and ECHLIN MANUFACTURING CO., ET AL. 483 410 OI!inion it chose Sherman as its supplier. Sherman was able to expand its operations to supply the additional kits to Carter. Insalaco 983- 987-88. It was expected that Carter would be buying all its kits from Sherman within six to nine months; the delay was designed in part to allow Carter to exhaust its existing inventories. I.D.F. 253; Sheehan 536-37; Eaton 2957; Lehman 5023- , 5037-38. (8) Complaint counsel maintains that Echlin, which had already acquired Borg-Warner s production facilties for carburetor parts and kits, took various actions in retaliation for Sherman s entry into kit assembling. One witness testified that Echlin threatened to cut off the supply of parts to Sherman. I.D.F. 254; Insalaco 932- , 975-76. Another witness claimed that Echlin threatened to withhold its catalog from Carter and stated that Echlin would address its supplying of parts to Sherman at a later date. Lehman 5027--3. Nonetheless Sherman continued to buy parts from Echlin, and Carter continued to receive catalogs. I.D.F. 254; Insalaco 966-7; Lehman 5033- 5042-43. Finally, when Sherman tried to attract Echlin s customers with lower prices, Echlin reduced some of its prices. I.D.F. 255. II. THE ANALYTICAL FRAMEWORK A. The Significance of Barriers to Entry Section 7 of the Clayton Act prohibits acquisitions that may have the effect of substantially lessening competition or tending to create a monopoly. Because Section 7 applies to " incipient" violations, actual anticompetitive effects need not be shown; an acquisition is unlawful if such an effect is reasonably probable. E.g., American Medical International, Inc. No. 9158, slip op. at 17-18 (July 2 1984) (104 F. Traditionally, an analysis under Section 7 begins with the definition of a relevant market and measurement of the concentration in that market. See generally FTC Statement Sections III, VI; DOJ Guidelines Sections 2 11. This approach by itself is unsatisfactory, however, because it fails to reflect many factors (9) that can determine whether a merger is likely to lessen competition substantially by enabling one or more sellers to impose higher prices than would prevail under competitive conditions. See Grand Union Co. 102 F. 812, 1038--1 (1983). Therefore, the Commission also looks to other considerations that bear on the likelihood of anticompetitive effects. See generally FTC Statement Section II; DOJ Guidelines Sectiom 21 to 3.45. These additional considerations often do not lend them other promotion;!) aids if it so desired.See Insalaco 965; Carlson 2560-1; Stefan 2669-72; Men 2737; Krm 2922-29, Rather, Sherman made a conscious decision to offer a short line without promotional1 assistance, at a 10 price, as a competitive gambit.See I.D, F- 243-44. If this strategy failed, it was only because potential buyer preferred Echlin s product, price, and mix of services to those offered by Sherman. See LD. F. 225; cf United Stat 1-'. Waste Management. Inc. 74.3 F.2d 976, 984 (2d Cir. 1984) (goodwil earned by incumbent fitt is not barrier entry).

. \ Opinion 105F:T.

selves to precise mathematical expression, but they can be more important than quantitative measures of concentration. See American Medical International, Inc. No. 9158, slip op. at 27 (July 2, 1984) (104 C. 1).

The most important ofthese considerations is the existence ofbarriers to entry. See Grand Union Co., 102 F. C. 812, 1063 (1983); FTC Statement Section III (A)(l). On the one hand, an absolute barrier to entry, such as a governmental prohibition or an essential but unobtainable natural resource, supports the original market definition by confirming that no new supply would enter the market even if prices were increased significantly above the competitive level. On the other hand, if there are literally no barriers to entry, the original market definition loses any economic significance. An attempt to exercise market power in an industry without entry barriers would cause new competitors to enter the market. This additional supply would drive prices back to the competitive level. Indeed, the threat of new entry can be as potent a procompetitive force as its realization. As the Supreme Court has recognized, the presence (10) of potential entrants on the fringe of a market can prevent the exercise of market power by the incumbent firms even if the potential entrants never actually enter the market. See United States v. Falstaff Brewing Corp. 410 S. 526 , 533 (1973); Ford Motor Co. v. United States 405 U.S. 562 567-68 (1972); FTC v. Procter Gamble Co. 386 U.s. 568, 580-81 (1967); United States v. Penn- Olin Chemical Co., 378 U.s. 158, 173- (1964); see also BAT Industries, Ltd. No. 9135, slip op. at 6-7 (Dec. 1984) (104 F. C. 1154). Thus, in the absence ofoarriers to entry, incumbent firms cannot exercise market power, regardless of the concentration in the nominal Umarket " and indeed even if that mar. ket" has been "monopolized" by a single firm. See United States v. Waste Management, Inc. 743 F.2d 976, 981-84 (2d Cir. 1984); Grand Union Co., 102 F. C. 812, 1063 (1983); FTC Statement Section II- (A)(1); DOJ Guidelines Section 3.3; W. Baumol, J. Panzar & R. Wilig, ':ontestable Markets and the Theory of Industry Structure, 350 1982); F. Scherer Industrial Market Structure and Economic Per- 'ormance 11 (2d ed. 1980); 2 P. Areeda & D. Turner Antitrust Law 05 (1978); J. Bain Barriers to New Competition 3-- (1956); Ordover , Willg, The 1982 Department of Justice Merger Guidelines: An Eco. omic Assessment 71 Calif. L. (11) Rev. 535, 555, 563 (1983); Baumol Wilig, Fixed Costs, Sunk Costs, Entry Barriers and the Sustainabilof Monopoly, 96 Q.J. Econ. 405, 411 (1981); Landes & Posner arket Power in Antitrust Cases 94 Harv. L. Rev. 937, 950 (1980). In American. Brake Shoe Co., 73 F. C. 610 (1968), the Commi sion stated that "the finding that cotry into a ket is diffcult j not indispensable to the finding ofiIegaJity under Id. at 684. In thClt case, however, the mission also noted several industry characteristics that cou.ld delay entry for several years. ld. at 683; see a/so 'Produ.cts Co., 65 F. C. 1163, 1208 (1964) (explaining that ease of entry could not overcome Section 7 j)egality (footnote conl' ), ECHLIN MANUFACTURING CO.. ET AL. 485 410 Opinion B. The Nature of Barriers to Entry This conclusion requires us to address the nature of barriers entry. Under Section 7 of the Clayton Act, the focus must be on industry characteristics that would allow incumbent firms to reap monopoly profits for a significant period of time. Complaint counsel suggests that entry barriers are high whcnever it is unlikely that new firms will decide to enter the market. We cannot agree. Although high barriers indicate that entry is unlikely, reversing that statement goes too far. For example, entry would be most unlikely if all the incumbent firms were losing money, yet this is clearly not the kind of barrier that facilitates the extraction of monopoly profits. See 2 P. Areeda & D. Turner Antitrust Law 11 409b (1978). The likelihood of entry may be influenced by general business conditions, the (12) potential return on alternative uses of capital, or the inherent riskiness ofthe industry. But although these factors may be relevant to an estimate ofthe competitive price level in an industry and thus to the likelihood of entry at any given time, they do not make it more or less likely that incumbent firms will be able to cxceed the competitive price level and charge monopoly prices. Regardless ofthe exact price level that competition should dictate in an industry, which wil depend in part on the factors cited above, prices above that levcl will create a strong inducement to potential entrants and make new entry more likely. Cf DOJ Guidelines Section 3.3 (referring to "likelihood" of entry in response to price increase). Respondents propose that an entry barrier be defined as additional long-run costs that must be incurrcd by an entrant relative to the long-run costs faced by incumbent firms. This definition is now widely accepted in the legal and economic communities. See G. Stigler The Organization of Industry 67 (1968); see also Nelson 1531- , 1654- 1935; Glassman 4214; Wilig 4746-47; 4 E. Kintner Federal Antitrust Law 37.4 (1984); 2 P. Areeda & D. Turner Antitrust Law n 409a (1978); R Posner Antitrust Law: An Economic Perspectivc 59 (1976); Baumol & Wilig, Fixed Costs, Sunk Costs, Entry Barriers and the Sustainability of Monopoly, 96 Q.J. Econ. 405, 408 (1981). The rationale underlying this definition is that low-cost incumbent firms can keep prices above the competitive levcl as long as those prices remain below the level that would provide an incentive to higher-cost potential entrants. Thus, a long-run (13) cost differcntial could erect because new entry "js likely to be "t ber!. a long-term affair af(d :l47 F.2d 745 (7th Cir. 1965). Because our understanding of barriers to entry, set forth below . eOCOmp!l9Se!) significant delays encountered by entrant.s, it does not depart from these earlier cases. To the extent that thegc cares may be relld as implying t.hat a violation of Section 7 can he found in the absence even of a significant dcby in entry, they are inconsistent. with more recent Iega1 and economic developrnentR, whirh are rel1ected in the materials dated above. Opinion 105 F.

a permanent barrier to new entry that would allow the maintenance of supracompetitive profits for an indefinite period of time. This definition offers a good framework for our analysis of the likelihood that an acquisition will substantially lessen competition. If for example, potential cntrants face an absolute governmental prohibition on entry, their costs would obviously exceed the costs of the incumbent firms. On a less extreme level, an entry barrier mjght exist if the incumbent firms possess patents that enable them to operate with substantially lower costs than an entrant who could not duplicate the patents or achieve the same results by any other means. It should be noted, however, that a definition of entry barriers in terms of costs can be misleading. The relevant costs are economic costs measured at the time of entry. No barrier to entry is created, for example, ifinflation increases the cost ofa factor of production so that entrants must pay a higher nominal price than was paid by incumbents when they acquired their productive capacity. The economic cost to incumbents is the opportunity cost of retaining a factor of production, not the original price that was paid for it. Likewise, costs must be viewed from the same temporaJ vantage point for all firms. For example, the act of entering a market may involve a high risk that declines sharply after the successful entry has been made. If one were to compare the risk faced by a potential entrant with the current risk confronting an incumbent firm, it would appear that the incumbent has a decided risk advantage and thus a lower (14) cost of capital. This is not a proper comparison. The incumbent firm s apparently lower costs merely reflect compensation for the risk it incurred in entering the market. The potential entrant' s apparently higher costs will decline to that of the incumbent firm if its attempted entry is successful. The only meaningful way to compare the risks and costs incurred by the two firms is to apply the same yardstick to each by viewing each of them at the time of its own entry. Unless there is a barrier to entry, as defined above, market power cannot be exercised indefinitely. Sooncr or later, new firms wil enter the market and drive prices back down to competitive levels. From the standpoint of the public, however, it makes a great deal of difference whether this occurs sooner or Jater. There may be litte practical difference between an absolute barrier to entry and conditions of entry that delay thc restoration of competitive prices for decades. Therefore, we will also consider a second type of barrier to entry, which might more accurately be caJJed an impediment to entry. An impediment to entry is any condition that necessarily delays entry into a market for a significant period of time and thus allows market power to be exerciscd in the interim. "To be sure, merger analysis nroDP,rlv f()f'11 P.S nrimRrilv on lrmp"- prm rnmnptit,ivp -imnl-i(' tirmo; hilt ECHLIN MANUFACTURING CO.. ET AL. 487 410 Opinion- short-term effects should not be ignored, particularly if they are substantial." FTC Statement Section IIIA)(1); see also 2 P. Areeda & D. Turner Antitrust Law nn 409a, 505 (1978); R. Bork The Antitrust Paradox 311 (1978); R. Posner Antitrust Law: An Economic Perspective 58-59 (1976); cf DOJ Guidelines Section 3.3 (two-year test); Easterbrook Limits of (15) Antitrust 63 Texas L. Rev. 1, 32-33 (1984) (five-year test). For example, if entry into an industry is only possible by constructing a physical plant that cannot be completed in less than a decade, that industry would appear to be characterized by a high barrier to entry for purposes of our analysis under Section 7. Once again, the inquiry must focus on industry characteristics that permit incumbent firms to earn monopoly profits, not on characteristics that increase risk or decrease profitabilty for all firms and thus affect the competitive rate of return for the industry as a whole. 111. THE EFFECT ON COMPETITION As explained above, an acquisition is not likely to have substantial anticompetitive effects if the evidence shows that there are no barriers to entry, regardless ofthe level of concentration that is present in the relevant market. For purposes of discussion, therefore, we will assume that the ALJ was correct in adopting the relevant product market advanced by complaint counsel, which is the assembly and sale of carburetor kits. (16) Complaint counsel identifies four alleged market characteristics as potential barriers to entry: Sunk costs, economies of scale, recent history, and predatory practices. We reject each of these arguments and agree with the ALJ' s conclusion that entry into this industry i8 rather easy" and that entry barriers are !' virtually nonexistent" or very low." I.D.F. 216, 273, 296. The evidence shows that there are n, barriers to entry into the assembly and sale of carbureter kits and n. impediments to entry that would delay entry for a significant tim, Because this finding eliminates any possibility that the acquisitio may have a substantial anticompetitive effect, we affrm the ALJ initial decision dismissing the complaint. Complaint counsel first maintains that entry into the assembly sale of carburetor kits requires a significant investment to design line of kits, purchase equipment, build an inventory of kits and par and introduce the new line. These costs, according to complaint cov sel, are sunk costs because they would be unrecovered or only parti Iy recovered if entry were unsuccessful. The presence of sunk co 5 We also assume without deciding that complaint counsel has satisfied the initial burden of offering evie suffcient to make out a prima facie case of liability. When resolution of one issue may make it urmeccssa undertake a comp!!)" add time-consuming analysis of other issues, the CommissionwiJ often consider that first. See International Telephone Telegraph Corp., No. 9000, slip op. at 29-0 (July 25, 1984) l104 F. In this case, thf! absence of barriers to entry is just such an issue Opinion 105 F.

thus imposes an additional risk on a potential entrant and requires it to demand a higher-than-normal rate of return before making the investment. Because the required rate of return is an economic costs potential entrants would face higher costs than the incumbent firms. This argument fails for several reasons. First, as the AW conclud- , the evidence shows that these costs are insignificant. See I.D. 226 233-35. Second, we note that one (17) class of potential entrants resellers, would not incur many of these costs because they already have kit designs, an inventory of kits, and a distribution system. Moreover, as complaint counsel concedes, an investment in equipment to facilitate the assembly of kits would not be a sunk cost because the same equipment can be used for other purposes. Third, there is some doubt whether sunk costs, standing alone, should be viewed as a barrier to entry at all. See Glassman 4240-1; Willig 4785. Ifsunk costs are considered an entry barrier, it must be because they create a difference in the risk confronting the incumbent firms who have already committed their resources and potential entrants who have yet to make that decision. See Nelson 1575-81; Wilig 4775-78; Baumol & Wilig, Fixed Costs, Sunk Costs, Entry Barriers and the Sustainability of Monopoly, 96 Q.J. Econ. 405, 418-19 (1981). This, however, is a false comparison, because the returns earned by the Incumbent firms reflect in part the risks they faced at the time they nade the decision to enter the market. In any event, the evidence ndicates that a potential assembler of carburetor kits would run risks hat are no higher than those faced by past entrants, who are the resent incumbents. LD.F. 272; Glassman 424 49. Complaint counsel next argues that economies of scale create a 1rrier to entry. This is especially true, complaint counsel maintains, 'cause the market for carburetor kits is small and may decline in the ture. Under this theory, a potential entrant wil recognize that its !ty is only feasible on a scale that would create excess capacity in e industry and thus (18) depress prices. This potential for excess pacity and falling prices increases the risk of en try and the required ;e of return.

TVe agree with the ALJ's conclusion that there are no substantial nomies of scale in this industry. See F. 227-32. In fact, all but ) ofthe firms operating after the merger are doing so at a level that )wer than the "minimum eficient scale" urged by complaint counwhich is ten percent ofthe market; several firms operate with two ,ent ofthe market or less. LD.F. 189-92. Thus, ifthe smaller firms nplaint COlJncJ' s expert witrw8S testified that these companies are able to !lurvive only because the existing ,vel is above the competitive level and thus creaWi! a "price umbrella " sheltering the smaller competitors 1eir alleged cost disadvantage. Nelson 2242-44, 2255-6. This position was properly rejected by lhe ALJ on unci that there is no credibje evidence that kit assemblers have uilrned supracompetitive profits. I.D, Complaint counsel's argumeo.t is also inconsistent with the decisions by Ford Motor ComparlY and Carter (foot""." ..

._ _. ., . ..UU 410 Opinion do suffer a cost disadvantage, it does not reach a level that would raise substantial competitive concerns. Moreover, a reseller intending to enter the market would have a significant head start on the necessary market share, even assuming that such a thing exists, because it could begin by assembling kits for its own use. Finally, we cannot agree that economies of scale and declining markets necessarily create barriers to entry. They may increase the risk of participating in this industry and thus the competitive rate of return required by all participants but they (19) do not impose a risk or cost on potential entrants that was not borne equally by the incumbent firms. See 2 P. Areeda & D. Turner Antitrust Law n 409b (1978); R. Bork The Antitrust Paradox 311 (1978); R. Posner Antitrust Law: An Economic Perspective (1976); G. Stigler The Organization of Industry 67 (1968). But see Nelson 1581- , 1657--0; DOJ Guidelines Section 3.3 n.21. Complaint counsel also looks to the historical record, which is characterized as devoid of successful entry into the market. The paucity of past entry is claimed to evidence high barriers. See FTC Statement Section III(A)(l). The absence of past entry, however, does not prove the existence of entry barriers because it is equally consistent with alternative explanations, such as a declining industry or competitive prices. I.D. F. 273; United States v. Waste Management, Inc. 743 F. 976 983 (2d Cir. 1984); Grand Union Coo, 102 F. C. 812, 1064 (1983); Baumol Contestable Markets: An Uprising in the Theory of Industry Structure 72 Am. Econ. Rev. 1, 14 (1982). Obviously, it would be absurd to infer the presence of entry barriers and a potential for supracompetitive profits from evidence that is equally consistent with competitive or subcompetitive profits.

In fact, the historical record refutes the allegations of barriers to entry in this industry. The critical facts do not concern the frequency of entry so much as the manner, ease, and rapidity of the entry that has occurred. Five companies have (20) entered the market in the last decade. All five companies continue to operate despite the fact that four of them have market shares that are extremely small. One company began operations with an out-of-pocket investment of only five hundred dollars. The largest recent entrant, Sherman, sold its first kit only three months after making the decision to begin assembling kits. When Carter later decided to buy its kits from Sherman instead of assembling them, Sherman was able to expand within six to nine months to meet the substantial new demand, and part of this delay was due to Carter s decision to exhaust its inventories before purchasto cease OIssembling carburetor kits. If price levels were so far above the competitive norm that even very small firms were profitable dcspite the allegedly bigh economics of scale, it is diffcult to see why these two relatively large assernblers found it more profitable to buy their kits from others.See I.D.F. 152 7 In particular, the risk of excess capficity is borne by all firms, whether they entered the market recently or long ago, because any firm could lose market shsre in the event of excess supply or insuffcient demand Opinion 105 F.

ing all its kits from Sherman.8 Thus, the experience of Sherman demonstrates that if prices were ever raised above competitive levels a future entrant could become a major assembler of carburetor kits in less than a year, even if it is assumed that Sherman s entry and expansion were delayed solely by conditions of entry. Moreover, Sherman was in no better position to enter the market, and in many ways was in a worse position, than many resellers of carburetor kits, who are poised at the edge ofthe market in a position to begin assembling kits on short notice with only a minor investment. We (21) conclude that nothing in this record suggests that entry barriers would allow the incumbent firms to maintain supracompetitive prices for any meaningful period of time.

Finally, complaint counsel argues that the risk of retaliatory actions by the incumbent firms against entrants is a substantial deterrent to new entry. This theory postulates that potential entrants unlike the incumbent firms, risk arousing the wrath of the present assemblers, who can undermine an entrant's profits through belowcost pricing and other predatory devices. Recognizing this additional risk, the entrant would require expected profits that exceed the competitive level before committing itself to entry. The existence of this margin between competitive prices and prices sutlcient to trigger entry would allow the incumbent firms to exercise market power. This argument fails for two reasons. Retaliatory price-cutting and other predatory practices are unlikely to deter entry unless there is a significant barrier to entry in addition to the mere threat ofretaliation. Ifthere is not such entry barrier, the incumbent firms will never be able to raise prices above the competitive level without attracting entry, and therefore they wil never be able to recoup the losses they suffered by selling their products below costs. 3 P. Areeda & D. Turn- Antitrust Law n 711b (1978); see International Telephone Telegraph Corp. No. 9000, slip op. at 37, 43-44 (July 25, 1984) (104 F. 359). We have found no such additional barrier, so any threat of retaliation in this industry would be unlikely to be effective and hence would not be credible. (22) Moreover, the evidence does not reveal that retaliatory actions have taken place in the past or will take place in the future. The first alleged retaliatory act was a threatened refusal by Echlin to sell parts to Sherman after Sherman began assembling kits. The only evidence of such a threat is the hearsay testimony of an independent sales 3 Complaint counsel asserts that the prCS€lice ofbClrriers1.0 entry is demonst.rat.ed by Shennan s low market share after twu year of assembling kits. The evidence, however, reveals that Sherman low market share was attributable to its diffculty in selling kits in competition with other assemblers, and not to any impediment that prevented it from producing kil See supm. or course, if other assemblers were charging Rupracompctitivc prices, demand for competitively priced kit.s would in r€ase, and a company like Sherman would be limited in its effectonthemarkctonlytothcext.€!1tthOltentrybClrriershjndered it. efforts to assemble enough kits to meet that demand ECHLIN MANUFACTURING CO.. ET AL. 491 410 Opinion representative of Sherman, who said that Sherman s owner tala him that an unidentified person associated with Echlin threatened to stop selling parts to Sherman. No party offered testimony from anyone with first-hand knowledge of the conversation in question. Unlike a federal court, an administrative agency may receive hearsay into evidence, but hearsay is not thereby entitled to the same evidentiary force as testimony based on the first-hand knowledge of the witness. Under the circumstances of this case, we believe it is entitled to no weight at all. In fact, notwithstanding the alleged threat, Sherman did enter the market and continued to buy parts from Echlin. And even if the alleged threat had materialized, the record is clear that Sherman had access to many other potential suppliers. We cannot conclude from this record that there was a threat, that it would have been credible, or that it had or could have had any em,ct on competition.

The second allegation of retaliation is that an Echlin employee threatened to stop providing Carter with catalogs after Carter decided to buy its kits from Sherman and told Carter that Echlin would address its sales of parts to Sherman at a later date. Withholding a catalog is hardly a threat suffcient to deter entry, since catalogs were widely distributed and readily (23) available from other sources. Lehman 5046--7; see Fife 626; Nelson 1980; Thompson 2452. The statement regarding the supplying of parts is not even an explicit threat and even if it were, a threat to withhold parts would not be credible because of the presence of alternative suppliers. Since Echlin continued to supply catalogs to Carter and parts to Sherman, and since Sherman was not deterred from entering the market, we wil not infer the existence of a barrier to entry from this weak and ambiguous evidence.

Complaint counsel also calls attention to price cuts initiated by Echlin after Sherman entered the market with kits priced at a discount. There is no evidence that these prices were below Echlin average variable cost or that they met any of the other criteria for predatory pricing. See generally International Telephone Telegraph Corp. No. 9000, slip op. at 17 24 (July 25 1984)1104 F. C. 359). Vigorous price competition is encouraged by the antitrust laws, and an increase in competition can be the expected and desirable result of entry by a new competitor. See W. Baumol, J. Panzar & R. Willig, Contestable Markets and the Theory of Industry Structure481 (1982). A price was is evidence of competition, not the absence of competition. Therefore, we .fnd that there is no barrier to entry into the assembly and sale of carburetor kits. The evidence demonstrates that entry into this market is extraordinarily easy and can be quite rapid. There is thus no possibility that supracompetitive prices for carburetor kits Dissenting Opinion 105 F. could be maintained and no likelihood that competition will be lessened substantially by the acquisition. (24) IV. CONCLUSION Because we conclude that Echlin s acquisition of Borg-Warner automotive aftermarket division does not violate Section 7 of the Clayton Act or Section 5 of the Federal Trade Commission Act, we affrm the decision of the ALJ dismissing the complaint in all respects.

DISSENTING OPINION OF COMMISSIONER PATRICIA P. BAILEY This is a merger between competing firms with 36% and 10% of a small and declining market so highly concentrated that six firms account for 95% of sales. The Herfindahl-Hirschman index as a result of this acquisition rises by over 750 points to just under 3000. These figures would suggest that this market is susceptible to collusion. There are few sellers in the market for assembly and sale of carburetor kits, and their market shares have remained stable over the past 15 years. There are large numbers of buyers most of which make relatively small purchases, limiting the ability of buyers to disrupt collusion. Because of the similarity of these buyers' businesses in reselling what are fairly standardized, noncustomized products, there are relatively few issues over which sellers need collude. Substitute products (new and rebuilt carburetors) are considerably more expensive, and demand is alleged to be inelastic, since car repairs create necessity. Industry members use price lists, which facilitates price policing, and discounts off these lists are uncommon. There has been relatively little price competition, according to the ALJ, although he found that non-price competition did exist. There is evidence that the largest respondent exercised price leadership. The question of supracompetitive profits is disputed (the ALJ considered the evidence fragmentary" and the Commission rejects it without discussion), but industry leaders testified that their operations were profitable. (2) Under the 1984 Justice Department Merger Guidelines-the most recent government pronouncement on merger analysis-a merger that looks like this one is so likely to be anticompetitive and therefore unlawful that only the "extraordinary" case wil avoid legal sanction. The Commission has dismissed this case on the sole ground that finds no barriers to entry into the market, holding that this conclusion renders the otherwise strong structural case for ilegality irrelevant) L In fact acknowlerlgrnentofcomplaint counsel's prima facie case is reJegated to a footnote in themaiorjtv , ECHLIN MANUFACTURING CO., ET AL. 49:: 410 Dissenting Opinion I have three primary concerns about the Commission' s decision and its implications for future FTC merger policy. First, I believe the Commission has embraced a particularly narrow definition of barriers to entry that may be ill-suited to merger analysis, and which is moreover, a source of much dispute among industrial economists. Second, I disagree with the conclusion drawn by the Commission, that entry into this market is "extraordinarily easy and can be quite rapid." Finally, as a matter of legal policy, I am concerned over the Commission s single-minded focus on the hotly disputed barriers to entry issue as dispositive ofJegal liability in a horizontal merger case where the prima facie case for antitrust concern about collusion is as strong as it is here. (3) Barriers to entry are clearly of increasing importance to antitrust analysis. From a conceptional point of view, this is not hard to understand. Former Director of the FTC's Bureau of Economics, F. M. Scherer, has stated that " significant entry barriers are the since qua non ofmonopoJy and oligopoly, for. . . sellers have little or no enduring power over price when entry barriers are nonexistent."2 The Commission has recognized the role of barriers as a supplement to consideration of quantitative factors such as market shares and concentration. "The issue of entry barriers is perhaps the most important qualitative factor, for if entry barriers are very low it is unlikely that market power, whether individually or collectively exercised, wil persist for long.'" The Department of Justice has gone even further in stating, "If entry into a market is so easy that existing competitors could not succeed in raising price for any significant period of time the Department is unlikely to challenge mergers in that market." Two recent federal court decisions have hoisted the Justice Department on its own petard (4) by denying government merger challenges on the basis of low barriers to entry.

But to say that barriers to entry are important in antitrust thinking does not lead me to the necessary conclusion that barriers analysis has yet reached the point where it should resolve antitrust disputes as easily as it is being used to do in this and in possible future cases. For one, there is such lack of clear consensus about the nature or 1 Scherer industrial Markel Siructl!rermd Economic Performance 11 (2d cd, 1980) .. Statement of Federal Trade Commission Concerning lIorizontaJ Mergers, S..crion III A(l), p- 5 (J982) (hereafter FTC Merger Statf'ITwnl"

, US. Department of Justice Merger Guidelines, Section 3. 3 (1984) (hereafter DO,J Merger Guides S. v. Waste Management, Inc 743 F. 2d 976 (2dCiT 1981); Us. u. Calmar 1985-1 Trade Cas. (CCH) 588 (D. 1985) F. No matter how appropriate it may be to take ent-ry barriers into account in determining whether or not it is worthwhile to bring divestiture actions against duminant firIns in concentrated industries, the case for modl'- (footnote cont' .. . , , , , . . ) .. . 494 FEDERAL TRADE COMMISSIQN DECISIONS Dissenting Opinion 105 F.T. effect of barriers to entry' that some suggest this issue is elusive, and can confound the resolution of complex antitrust questions.8 One scholar has observed that barriers to entry is "the single most misunderstood topic in the analysis of competition and monopoly," exceeding (5) even the issue of market definition in this regard. The majority cements its agreement with respondent' s definition of entry barriers ("additional long run costs that may be incurred by an entrant relative to the long-run costs faced by incumbent firms ), by a citation to Professor George Stigler, among others. The majority declares this position is "now widely accepted in the legal and economic communities. " Stigler s formulation (1968) defines barriers to entry "as a cost of producing (at some or every rate of output) which must be borne by a firm which seeks to enter an industry but is not borne by firms already in the industry. l0 This view, as carried forward in the majority s analysis, is that entry barriers block new competition from the market; access to the market is closed to outsiders on account of the barrier. Examples of such barriers given by the majority are governmental entry restrictions and patents. Other examples might be control of scarce resources, such as essential raw materials, or unique management or labor resources. Conversely, under a "Stiglerian" approach, if some factor simply imposes risks and costs on new entrants resulting in possible delay or deferral of entry, that factor is not really a barrier to entry, because access to the market imposes or has already imposed the same costs or risks on all firms, at one time or another. All firms have equal access to the market, even (6) given the need to undertake certain prescribed steps to accomplish entry.

Perhaps at the other end of the scale from Stigler s view is the neo-classical" view of Joe S. Bain (1956), which would measure the prospect of entry by the "extent to which, in the long run, established firms can elevate their selling prices above the minimal average costs of production and distribution (those costs associated with operation at optimal scales) without inducing potential entrants to enter the industry. !! The condition of entry is thus defined "as the 'disadvantage' of potential entrant firms as compared to established firms or rating presumptive merger rules where entry barriers are low is much more tenuous. IV Areeda and Turner Anlirusl Law 917(b) at 86 (1980) 7 Demsetz Barriers to F:ntry" 72 Am. Iocon. Rev. 47 (1982). Determining the existence height,' and effect;; of entry barriers is besel with some theuretical diffculties and with empirical problems of seemingly formidable proportions." IV Areeda and Turner. supra . 1917(b) at 87. See also- Rowe The Decline uf Antitrust and the Delusions of Models: The Fau tian Pact of Law and Economics 72 Geo 1. J. 1511 (1984). (Barriers to entry an "elusive " concept. Wherever the market, some enterprising rivals unless kept out by legal fiat, can in time dimb in. So, barriers to entry stand only as high as time waxes long and rivals grow ta!!."Id. at 1543.

"Fisher Diagnosing Monopoly, " Q. Rev. Econ. & Bus- 23 (Summer, 1979). I" Stigler The Organization of Industry 67 (I96A) , , ECHLIN MANUFACTURING CO.. ET AL. 495 410 Dissenting- Op!niqQ conversely, the 'advantage' of established over potential entrant 12 In essence, an entry barrier is whatever allows incumbentfirms. firms to charge supra-competitive prices yet not attract new entry. The Bain view, while subject to almost thirty years of criticism by its opponents and revisionism by its friends, remains an alternative approach which provides a different perspective on entry questions. The Commission distinguishes between "absolute" barriers to entry -which are barriers touched by the Commission s Stiglerian Philosopher s Stone-and mere "impediments" to entry, evidepce of which are taken up by the record in this case. (Some ofthese resemble Bain barriers to entry sent to the back of the classroom). The Commission finds no absolute (7) barriers to entry in this case at all, but it insists on an extended treatment of the subject, perhaps to accomplish the result of fixing in the caselaw its particular entry barrier definition. As to entry-delaying "impediments " the Commission rules that none of these would permit any exercise of market power by incumbent firms because ofthe ease with which the impediments could be kicked aside.

The Commission, then, in this opinion embraces the current "Chicago School" economic "State Religion" approach to barriers to entry, a view which simply is not generally "accepted 13 in the legal and economic communities." In both communities, though this view has gained some ground recently in a few cases 14 the battle rages fiercely, and is as empirically unresolved as ever.!5 (8) II.

But is this point important, or do I belabor an all-too technic, issue? It seems to me the point is important if barriers to entry, , particularly defined, become the easy way to resolve complex an trust cases. Section 7 of the Clayton Act makes jJegal mergers th have the probability of substantially lessening competition. The st ute does not add the word "forever . A merger can lessen competiti and therefore violate the statute, in my view, if market condition structural features, or behavioral patterns make entry an unath tive investment risk for a significant period oftime. Ifsuch a situat exists so as to permit supracompetitive pricing within an industry 13 The Commission uses the term "widely accepted. " Alas, that may be so, depending however on the ci which one travels.

14 Those cases include the Commiggion s decision inGeneral Food Corp. 103 F. C. 204, 354 n. 54 (1984) I expressly disassociated myself from the Commis.ion s notational embrace ofthisStiglerian view. (103 f 312).

15 See for example, Demsetz supra. Waterson On the Definition and Meaning of Barriers to Er Antitrusl Bulletin521 (1981), and Went7. Mobilty Factors in Antitrust Cases; Assessing Market Power of Conditions Affecting Entry and Fringe Expansion " 80 Mich. L. Rev. 1545 (1982). Dissenting Opinion 105 F. yet prevent for a substantial period new entry or the expansion of marginal fringe competitors, then it is possible that a barrier to entry exists. The defect in the Stiglerian alternative is that it does not account for the time, scale and cost necessary for a successful entry that is a meaningful threat to incumbent firms. This situation, as I understand it, is essentially what complaint counsel is arguing. They do not claim that there are any of Stigler barriers to entry into this market, but rather they assert that entry is unlikely due to the fundamental unattractiveness of the market. New entrants are deterred from making investments in entry because they cannot expect to make acceptable profits. At the very minimum the argument seems to be, entry would be delayed for a significant period of time (9) during which there would be injury to competition constituting a violation of Section 7.

The majority admits complaint counsel's pragmatic point about conditions that delay or impede entry. The Commission states: "There may be little practical difference between an absolute barrier to entry md conditions of entry that delay the restoration of competitive ,rices for decades." Delayed entry "for decades" fits an almost-Stilerian scenario of an industry where potential entrants must await 1e expiration of patents or overcome similar dramatic entry prob- 'is. However, decade-long delays should not be and are not, the only rcumstance of concern under the antitrust law,16 Most temporal easures of new entry speak of diffculty of entry in terms of two to e years.

In the end, the majority concludes that where entry is not blocked its analysis), it is easy and can also be rapid-with citation to the ry over the past ten years of about five firms at the fringe of the rket. The majority assumes that any of these firms could expand 1perations virtually at will. (10) Jmplaint counsel buttress their statistical case by descriptions of ket conditions that permit the exercise of market power without lting in the expansion of fringe entrants or the entry of new Jetitors. They view the market as conducive to collusion and y profitable, but shielded by barriers that deter entry at a signifiscale.

,murky issue of profits cannot be finally resolved on this record. ,complaint counsel argue that this market enjoys supracompetirofits and therefore is attractive to entrants, neither the ALJ e Commission accepted this analysis. The Commission consid- (merger Statement Section IIl(A)(1) (1982). ("To be sure, merger analysis properly focuses primarily on ompetitive implications, but short term effects should not be ignored, particularly if they are substan. ninistrative law judge found that the evidence on profitability was fragmentary and hypothetical. since 19 run profitability data on carburetor kits wilsavailable for individual firms. In addison to teslimn"" ECHLIN MANUFACTURING CO., ET AL.

410 Dissellting Opinion ers it possible that the (11) industry may be unattractive to entrants because prices are competitive or simply because the market is declining,18 It is noteworthy that the ALJ, while finding some non-price competition, concluded that "the record as a whole does not reflect vigorous price competition. 19 And, if the declining market simply does not interest outsiders in and of itself, there would seem to be at least modest room for collusion, which this merger might facilitate. Accepting for the moment that the market is at least somewhat attractive for entry from the initial profitability assessment standpoint, there are alleged to be four factors that complaint counsel say mitigate against entry, but which the Commission rejects even as impediments.

First, complaint counsel assert there are sunk costs associated with entry that cannot be recovered if a firm decides to exit the market. These sunk costs are not large in terms of dollars, but they are large relative to the expected return in this small ($53 milion) and declining market, thereby increasing (12) the risk and decreasing the likelihood of entry given alternative investment opportunities. Second, complaint counsel contend that entry is deterred by the need to achieve an economy of scale that is quite high. Like the arguments surrounding supracompetitive profits in this record, arguments about economies of scale are a subject of bitter dispute. Complaint counsel's expert witness, using exhibits prepared by respondents, estimated that about 10% of the market represented minimum effcient scale,2o The ALJ rejected the 10% calculation; he by witnesses for four of the leading members of this industry that their companies were profitable, and pJannin documents of respondent stating iL operations are profitable, complaint counsel made use of two in camen exhibit. prepared by respondents ' employees and economic expert. TheBe exhibitb are based on data from Echlin own books and records, originally prepared to establish an economic modd of relative costs of production at thre different hypothetical levels of output. Complaint counsel, over the vigorous objection ofruspondents, asserted the this model enabled complaint counsel to est.ablish the profitability of a. firm that operated at about 10% of tlmarket, or s,des of 1.5 milion kits. Comparing these data with 1978-1982 Census Bureau Ql1arterly Financi Reports (QFR) five-year averages for I) return on asaets for alj manufacturing, 2) average return on atockholdel equity, and even 3) return on assets for wholesaling, complaint counsel argles that all of these QFR "benchmark are very substantially exceeded by the profit data derived from respondents' economic exhibit.. The degree which these calculated "profiis" exceeded the benchmarks (50%) was web above the level agreed by both sid economic experts to indicate Sl.prac.mpetitive profiability. (Complaint counsel's profit calculations yielded" counting" rates ofreturn. Such acc01mting profit.s are potentially unreliable because they do not take into acco certain ofa finn s costs; however, complaints01msel explain that in this industry, accounting rates of return clOI approximate economic rates of return, considered more reliable evidence ofprofitabjJjty by some economists addition, although inferences to be drawn are limited, there is evidence that respondent was able to retail against one new market entrant by offering discounts on selected kit. ranging from 5-'10%- The Comm expressly rules that these sales were not below variable cost. There is the suggestion that respondents' usual p are normally above a competitive level I FlhauJd note here that the DOJ guides ated caselaw do not provideperforse legality for merger.' in decl industries.

ll.D.F. 286. p. 61 20 To assess the level of scale economies, complaint. counsel again relied on ex 543 and 544, an .malysis market at three hypothetical levels of output, prepared by respondent's employees from respondents' bool records. These in camera exhibits explain certain characteristics of three different sized firms-a fin produces 15 milion carhuretor kits imntHdly, representing-just over 10% of the market, a firm producing c kil annually or about 2% of the market, and a firm with about 1% of the market, orahou.t 100 000 kit.'. TJ showed, according to complaint counsel, substantially different unit costs associated with each si7. of outp (footnote Dissenting Opinion 105 F.T. agreed that there were some scale economies in this industry but considered them insignificant. However, if new entrants were faced with economies of scale of 10%, achievement ofthis reduction of unit costs would give a decided cost advantage to the larger incumbent firms, and saddle (13) entrants with a major competitive burden to attain these same advantages without prospect of doing so simply from capture of any market growth. Incumbent firms capable of output at these scales could also deliberately flood the market to deter entry with lower prices. Because potential entrants perceive this to be the case, the need to achieve scale economies deters entry. Third and fourth, complaint counsel also argue that the recent record of new entrants is especially poor in this industry, and that the record shows at least one case of'targeted market retaliation by the market leader against a new firm.

Areeda and Turner endorse a shorthand test for barriers to entry by assessing the historical record of entry over the past few years in the market.2! The 1984 DOJ Guidelines propose a two-year period in which to assess new entry in response to a "small but significant non-transitory increase in price" (about 5% lasting one year)."" The 1982 FTC statement emphasizes the importance of the historical So does the body of traditional caselaw. I believe ecord on entry.23 hat judging the historical record on entry has always been, and emains today, the simplest and most practical way to deal with most arrier to entry analysis situations. (14) While these historical tests emphasize the importance ofthe tradi- )nal study of the simple record of entry, they also emphasize the Iportance of the size and scope of such entry. The 1984 DOJ Guides mid take into account the "magnitude" of entry.24 The 1982 FTC Itement declares: " Evidence of substantial expansion by firms alldy in an industry, especially non-dominant firms, may persuasiveindicate that barriers to larger scale are not high. Conversely, :Jence of frequent entry by fringe firms on a small scale without lificant expansion may also suggest the existence of barriers to ,er scale 25 (Emphasis added).

he record in this case shows that over more than a decade only he "firm " the more decided the cost advantilge. There is additional evidem:(' that this upper range of output proximate the optimum in tbe business expansion plans of another competitor (Holley). allowing it the cost , of automated packaging-. Respondent is 8C-omful of the economies of scale argument. OIJ methodological , on the grounds that the numbers are purely hypothetkal, on the fact that only two of the six markd n1.5 would meet the 10% criteria (although the smaller finns have t.he exist.ing capability to expand on to 10% , according to romplflint counseb industry expert), and on the fact that the r1Umerous smfJller pear to be profitable ,reeda and Turner $upm. 917c, at. 88 (1980) Merger Guides, Section 3.3 (1984) Merger Statement Section In A(J) (1982) Merger Guides Section, 3 (1984) Merger Statement Section III A(J) (198 .

ECHLIN MANUFACTURING CO., ET AL. 499 410 Dissenting Opinion fringe competitors have entered and only to the extent of a total combined market share of about 2%. Moreover, expansion or increased profitability has not occurred over a ten year period for some of these firms, and three years or more for others. The majority conclusion that expansion is "easy and rapid " is not supported by the historical record.

There are serious questions in my mind that these struggling fringe competitors represent any disciplining threat on the prices of the market leaders, and unless they do, their entry is not "significant quite apart from the question of their size. (15) The AU acknowledged only one firm, Sherman, to be a "significant" competitive entry. The record shows that Sherman s 1981 efforts at assembly and sale of a line of 120 kits at low prices continued for two years with little success. Sherman s survival, with about one-third of one percent of the market may be owed mostly to an agreement to supply kits to another firm for resale, after that firm (Carter) gave up on the assembly and sale of kits on its own (for corporate reasons apparently not related to the kit market). Sherman obtained this account in 1983 by underbidding Borg-Warner s carburetor kits subsidiary. About the time that the Carter/Sherman agreement was implemented, respondent targeted Sherman with special and unusual discounts on 19 lines of fast-moving kits. The Commission s observation about the targeted response to Sherman s entry and Sherman s capture of the lifesaving Carter business after two years of struggle, is that "A price war is evidence of competition, not the absence of competition." That is Sherman s presence tempered the market leader s prices overall. Thi, is simply not consistent with the selectivity ofthe response, or the fac: that two years passed before it even began. A prospective entran might take the following view: that a recent entrant, Sherman, wa targeted for selective price cuts by a leading firm that almost certair Iy possessed a cost advantage in calculating the degree of its retaliatt ry discounting (5-30% off list price, according to the record).27 Ro' might such a prospect affect the next firm s (16) decision to enter tt market? To me, this is a relevant question bearing on the likelihor of any further new entry, even at the margin. Of course, it can be sa that Borg-Warner s superior effciency (scale economies) is technic, Iy available to all. But even taking this into account, is the risk wor taking in terms of anticipated post-entry return? The most obvious pool of potential new entrants are the numerc firms that are private label resellers of kits assembled by the f firms that populate the assembly market. Some ofthese were form 26 In contrast, in the recentCalmer case, thi! judge found l1umerous entranL'J , some of which had am; subst"ntial market shares. U.S. v. Calmar, supra at 65, 927-28 (D.XJ. 1985). The fact that the Commissiun found none of these retaliatory discounts to hI! predatorysuggc!\U; that pre- ,. J.;"''''r t. han in a competitive environment, , Dissenting Opinion 105 F. ly assemblers, and all possess some access to distribution systems and the advantage of some name recognition and familiarity with customers. However, the same factors that operate to discourage new entrants, or expansion by fringe entrants, operate to deter resellers particularly since resellers may be even more knowledgeable and alert than others about the dismal record of entry in this market. Most kit resellers handle small percentages of kit resales, and all but one (Ford) have 3% of such resales or less. Therefore, even firms with cus-established distribution wil be forced to capture "changeover" tomers ifthey are to achieve scale economies. But the most significant factor about potential reseller entry is that no reseller has ever successfully entered into kit assembly. The market is clearly unattractive to the new entrants best poised to make the effort, and some factor must account for this fact. (17) In these circumstances one could predict that further entry is likely to be judged very risky indeed, and given the unlikelihood of any significant newmarket growth, doomed. With such a poor record on entry, the presumption of market power that attends high market shares, high concentration and Herfindahl levels should remain standing, somewhat battered to be sure, but unbowed. If expansion is not occurring, the Commission opines, it simply must be attributable to some factor other than incumbent firms' market power-power oormally inferred from the enormous "numbers" in complaint coun- ;el' s prima facie case. To suggest that the failure to expand can be Jased on the invisible evidence of some invisible hand is such a specral conclusion that it is less credible to me than the anticompetitive nferences to be drawn from the traditional market tests so recently onfirmed in the 1982 FTC and 1982 and 1984 Justice Department lerger frameworks. There are no additional arguments to add to this .Iance, since there are no credible arguments that this merger enmces effciency, or that it is likely to promote competition in some her fashion.

In summary, the likelihood of any firm entering this market does ,t depend simply on the existence or absence of Stigler s (18) entry rriers. The presence of supra competitive prices may normally be an :lucement to entry, but, depending on the record on entry, it may ;0 suggest the historical lesson that entry is risky, and therefore ;erred.29 A firm s decision to take the investment risk depends on claim that social C08t. of a rncq:;cr would be small is Ii mu.ch weaker defel1I\ than a claim, as in an economic!: !defense, that a merger would yield social gains. The social interest in attempting to isolate and immunize the Drmer is plainly !e8 than the ocial interest in protecting the Jatter. It may be sensible to absolve the Jaw-loss ases, but whether it is or not depends heavily on the facility with which they can be idl:ntifed. IV Areeda od Turner supru, V917b at 87.

tonebreaker Corporate Profits and the Risk of Entry, " 58 Rev. ofEeoD. and Stat.. 33, 39 (1976). (High profi assciated with high risks of entry, deter entry, and enable established firms to earp supranormal profits. d of small firms at the edg-e of the market an important factor in assssing riak). , , ECHLIN MANUF ACTIJRING CO., ET AL. 501 410 Dissenting Opinion its anticipated post-entry rate of return. Initially, a small market that is stable or declining and promises no new growth for an entrant to capture may not be attractive for entry.30 It might, however, be the sort of market conducive to collusion. Unrecoverable sunk costs are not lightly to be regarded when the ten-year record of entry shows five firms holding two percent, and two ofthe three largest firms recently merged into a single firm with almost half the market. The few incumbent firms may have the scale economy advantage oflower unit costs, which may permit selective retaliatory pricing that is not strictly speaking, predatory, but is, generally speaking, entry deterring. And, if there is excess capacity, as there is alleged to be in the two or three incumbent firms that operate at 5-7% of the market expansion of product "runs" on individual lines of kits could deter entry by easily increasing supply and flooding the market with cheaper (19) kits. ! Finally, assessing once again the historical record of entry, the potential entrant/expander may well be aware that it faces no Stiglerian barriers, and no necessarily enormous capital investment costs in getting a toehold in the market, yet it may anticipate a post entry rate of return that does not justify the effort, given other investment alternatives.

Unlike the majority I regard this as a close case, and, on balance I come out the other way. The major weaknesses arguing against this outcome are the absence of stronger evidence as to supracompetitive profits of incumbent firms, and on economies of scale that may operate to create cost disadvantages. But it is precisely because of the typical-perhaps inherent-(20)ditIculty and potential ambiguity of such evidence in merger cases that the history of merger law shows a struggle to find simplifying assumptions that can proxy for economic proofpositive. Examples ofthese assumptions to which I am willng to give credence, based on my reading of the law, the 1982 and 1984 DOJ Guides, and the 1982 FTC Statement on Horizontal Mergers, are that high market shares and Herfindahls indicate the prospect for 30 DOJ Merger Guides, Section 3.3 n. 21 (1984); Webbink, Entry, l'rce- Cost Margins and Barrers to Entry in 280 4-DiJ;t Industries, 1967- 1972, Federal Trade Commssion Bureau of Econorncs Working Paper :-o. 19 (1979).

''I Spence Entry, Capacity, Investment and Oligopolistic Prcing," 8 Bell J. Econ. 534 (1978); Dixit, "A Model of Duopoly Suggesting a Theory of Entry Barriers " 10 Bell ,J. Econ. 20, 21 (1979) ("The threat of; a large enough post-entry output will make entry seem unprofitable, and then it need never be implemented"); Wenders Excess Capacity as a Barrier to Entry," 20 J. of Ind. Econ. 14 (1971). 32 The entrant should ignore preentry price and profit levels, but attempt to infer the pOBtentry equilbrium price and profit levels. lfthe entrant' s e;lpected profits are negative, he is deterred; the no-ntry profits accrue to the already established finn rather than the equally effcient entrant- Even a more effcient entrant may be deterred by an established finn who has sunk suffcient costs to make rus own exit uneconomical, and hence entry mutually destructive- Salop, "Strategic Entry Deterrence " 69 Am. Econ. Rev. 335 (1979). , Dissenting Opinion 105 F. collusion and that a long record offai!ed or marginal entry raises and inference of high entry barriers.

The sun has probably set on the rule of presumptive ilegality in horizontal merger cases, such as outlined in U.S. v. Philadelphia National Bank 374 U.S. 321, 364 6 (1963).33 Yet have we not gone all the way in the other direction if we say that the only relevant question is whether particularly defined barriers to entry are high or low, irrespective ofthe traditional indicia of antitrust concerns about enhanced potential for collusion? I seriously question, based on the facts of this case, whether any true advance in our knowledge of how this market really works is promoted by a rigid economic theory of absolute" barriers to entry, or a notion of delayed" entry in terms of decades. Certainly such an economic theory is outcome determinative, just as was the old rule of presumptive (21) ilegality. The analysis contained in this opinion ofthe Commission denies us the latitude to consider whether market structure, performance or conduct in a given case leads rational potential entrants to walk away, on the basis oftheir assessment of risks they face and the sorry record ofthe firms the Commission here would call "new entrants." One wonders why the FTC and the Department of Justice have spiled so much ink over how to analyze mergers, if it is all as easy as this. The clear implication of the writing and then rewriting of the DOJ Guides was that merger analysis was complicated stuff, and that fair enforcers should take into account relatively sophisticated analysis of all aspects ofthe market. I suggest that the majority has turned the old Section 7 numbers" game on its head in favor of a "quick look" approach for Stigler s barriers to entry, the new primus inter pares of merger law. One result of such an approach would be to sanction any and all mergers in this market, right up to the point where a pure monopoly results. If there are no barriers to entry, after all, what would be wrong with that?34 The entry barriers "quick look" test leads to a rule of per se legality for many mergers.

It is, of course, always possible that the distinguished and expert majority is dead right with their election of the economic theory to drive the result in this case. But my own brief(22) assessment ofthe literature on this issue shows it long on words and short on empirical findings. There is no surfeit of discussions of the issues, but no agreement on them, either.

What is emerging in Commission merger decisions is by and large JJ However, Judge Winter if!Waste Manugement observed that the Supreme Court ha never expressly stated that ease of entry i one of the circum8tot1ct' that fits the Court's approving as.'cssmcnt of market conditions to supplement the statistical market share and concentration data of the prima facie case, Huch as occurred inu.s. u. General Dynamics Corp. 415 U,S. 486 (1974). 743 F 2d at 982 (1984). 3' Or, as the Commission opinion pul In the absence of barriers to entry, incumbent firms cannot exercise market power, regardles.' of the concentration in the nomimd ' market,' and even if that ' market' hair been .

....n ...u, "UH' 410 Dismissal Order the rule that, according to the urrew"- economic learriing, a merge-r is almost always legal. The Commission has charted a new course away from the great body ofthe traditional caselaw, and indeed abandoned the assumptions that have attended merger enforcement policy of both old and recent vintage, substituting a well-nigh theological-and surely theoretical-economic deus ex machina. FINAL ORDER This matter has been heard by the Commission upon the appeals of complaint counsel and respondent The Echlin Manufacturing Company from the initial decision and upon briefs and oral argument in support of and in opposition to the appeals. For the reasons stated in the accompanying Opinion, the Commission has determined to affrm the initial decision. Accordingly, It is ordered That the complaint is dismissed. Commissioner Bailey dissented.

ADVISOHY OPINION 505 Proposed development of a relative value guide for physician services to be distributed to member internists would likely to have anticompetitive effects that probably would not be outweighed by any countervailng efficiency justifications. (American Society of Internal Medicine, P84 3532) April 19, 1985 Dear Mr. Kopit:

This letter responds to your request for an advisory opinion concerning the legality of the American Society of Internal Medicine ("ASIM") proposal to develop and disseminate relative value guides RVGs ). The Commission has determined, on the basis of the information provided by ASIM and additional information gathered by Commission staff, that there is substantial danger the proposed conduct would lead to a combination or conspiracy that unreasonably restrains competition among physicians in violation of Section 5 of the Federal Trade Commission Act. The Commission, therefore, cannot give advance approval to ASIM's RVG proposal. This advisory opinion begins with a brief summary of ASIM' proposal. It then discusses two central questions-first, whether there is substantial danger of an agreement in restraint of trade resulting from the proposed conduct, and second, whether, were such an agreement to result, it would restrain trade unreasonably. The letter then indicates alternative actions, unlikely to raise antitrust problems that ASIM can pursue to redress the alleged reimbursement disparities about which it is concerned.

ASIM' s Proposal ASIM, a national professional society consisting of approximately 000 doctors of internal medicine, proposes to develop an RVG and distribute it to its member physicians and to private and governmentaJ third-party payors on an advisory basis. ASIM plans to request that these parties consider using the RVG as a guide in developing reimbursement programs consistent with the approach contained in the RVG. The RVG would cover services that are provided by physicians who specialize in internal medicine ("internists ). ASIM proposes in the future to work with other physician organizations including surgical societies, to develop RVGs for other medical and surgical services.

The proposed RVG would Jist medical services by descriptive codes. ASIM intends to assign numeric values to each coded service, relative to one another, determined on the basis of costs, time, complexity, and the Jevel oftraining required to perform each service. The RVG would not in itself be a fee schedule, but could be converted to a fee schedule by physicians or third-party payors simply by multiplying the relative values by a dollar conversion factor. ASIM has indicated that it would 506 FEDERAL TRADE G'OMMISSION DECISIONS not provide conversion factors with its RVG; that the RVG and the other proposed aspects of ASIM's conduct would be voluntary and advisory" in nature; and that there would be no explicit or implicit threats or coercion against physicians or third-party payors to induce them to use the RVG.

ASIM has stated that it wants to develop the RVG to redress an alleged disparity in reimbursement for "cognitive" and "procedural" services provided by physicians. According to ASIM, a high level of insurance reimbursement now encourages physicians to use and sometimes overuse costly t'procedural" services such as surgery, electrocardiograms, x-rays, and other technology-intensive services. At the same time, ASIM submits, relatively low levels of reimbursement discourage physicians from using more time.consuming " cognitive services such as the diagnosis of patient health care problems, preventtative education, and life style evaluation. ASIM's members are internists, most of whom are chiefly engaged in primary care and the delivery of cognitive services. ASIM proposes to increase the relative value of cognitive services and decrease the relative value of procedural services to encourage use of more cognitive services and discour- , ifage overuse of procedural services. ASIM states that its RVG widely adopted, would reduce health care costs by creating incentives to substitute low-cost care for high-cost care. It further states that an increase in the relative amount at which cognitive services are reimhursed, as compared to procedural services, would encourage physicians to spend more time in personalized aspects of care and would provide new incentives for physicians to choose primary care specialties utilizing relatively large amounts of cognitive services. ASIM plans to use the "Delphi technique" to reach consensus on the relative values to be assigned to each of the services commonly provided by internists. In separate maiJ surveys, representatives of internal medicine subspecialty! organizations and two ASIM state affliates would be asked anonymously to assign relative values to medical services on the basis of time, complexity, costs, and training. Median and average figures computed by ASIM based on the first round of responses would then be submitted to the same physicians to use in making a second round of responses. The process would continue until a consensus or as much uniformity as possible was reached. ASIM's Resource Cost Committee would then review the product of each of these survey determinations and determine reJative values using the DeJphi consensus-building technique. The resulting RVG would then be submitted to ASIM's Board of Trustees for approval or disapproval without modification. ASIM also plans to send a "white paper" to physicians and third- I Internal medicine subspecillJties include cardiology, gastroenteroiol'Y, allerl!V. endocrinolol!. hernatnlrwv ADVISORY OPINION party payors that would explain the cognitive/procedural reimbursement disparity and use of its RVG to reduce the disparity. It -would also ilustrate how to use the RVG to "change the reimbursement structure from the current procedural service basis to a cost of resources basis." The stated purpose of the "white paper" would be to persuade and not to coerce.

Legal Analysis The antitrust issue raised by ASIM's proposal is whether it presents a substantial danger of an agreement that unreasonably restrains trade.2 The threshold question in resolving this issue is wljether there is danger of an agreement in restraint of trade. If there is a substantial danger of such an agreement occurring, the second question is whether there is a substantial danger that it would unreasonably restrain trade. The antitrust laws prohibit, of course, only those agreements that restrain trade unreasonably. Danger of Agreement in Restraint of Trade ASIM' s adoption and dissemination of an RVG, as it proposes, could involve or facilitate two types of agreements in restraint of trade: (1) an agreement among ASIM, its members, and possibly other physicians to adhere to the RVG in determining charges for their services; and (2) an agreement between ASIM, acting on behalf of its members and one or more third-party payors, possibly resulting from coercion that the third-party payor(s) wil adhere to the RVG in reimhursing physicians for covered services. The Commission concludes there is a substantial danger that the first type of agreement may occur; there does not appear to he a substantial danger of the second type of agreement.

With respect to the first type of agreement, if a professional association expressly or implicitly suggests or advises marketplace conduct on the part of its members or other competitors and the intent or likely consequence of the communication is that association members or others wil concertedly or interdependently modify their behavior in the marketplace to restrain trade, both the professional association and the individuals so acting could properly be found to be parties to an agreement in restraint of trade. 4 In contrast, when an association provides information or advice to its members or others that could be The Commssion di!\usss the antitrust risks of ASIM's proposal in tem3 of"s1.bstantial danger" because til advis.ry opinion seeks approval for propose future conduct, the precise nature andspecific effects of which cannonowUnited States u.beStandard Oil rleternnedCo., 221 U.S. 1 59-0 (1911). . See generally Interstate Circuit, Inc. v. United States,306 U.s. 208, 226-27 (1939) ("It was enough (for unlawful conspiracyJ that, krwwing that concertd action was contemplated 2nd invited, the(yJ gave their adhe ence to the scheme and participate in it. Each. . . was advised that the others were asked to participate; ea, knew that cooperation was esntial to successful operation of the plan. They knew that the plan, jfcarried O! would result in a restraint ofc.commerce, which.h. . was unreasonable..., and knowing it, al!participated in t plan.

, used by its recipients unilaterally in the marketplace and it is neither intended nor likely that the communication wil result in concerted interdependent action to restrain trade, the association would probabJy not he found party to an agreement in restraint of trade. 5 In this matter, the substance and market context of ASIM' s communications and physician actions in response to them would be critical in determining the existence of an agreement in restraint of trade. Although any action by ASIM, an association of individual practitioners many of whom compete with one another, to develop an RVG would reflect an agreement to take that action, ASIM' s development of an RVG, standing alone, would not constitute an agreement restraint of trade. No one would be party to an understanding by which he or she is committed to any particular course of conduct in the marketplace. However, antitrust analysis of ASIM's proposal must be focused on the entire course of conduct planned by ASIM to determine whether the proposal is intended to or could be expected to involve or facilitate an agreement in restraint of trade. Several factors indicate there is substantial danger that ASIM' proposed conduct would be intended to or would result in concerted interdependent action by physicians to adhere to the RVG in pricing their services. Despite the disclaimers it would make in its distribution of the RVG, ASIM appears to be proposing implicitly to invite physicians to adhere to the RVG in determining their charges. ASIM plans to send members the RVG on a "purely advisory" basis, leaving individual members "free to make independent fee decisions, with a white paper" that would "illustrate 'how to' use the (RVGJ to change the current reimbursement structure." The RVG would be prescriptive in nature, describing a set of pricing relationships that ASIM would be supporting as what should be. The RVG would be designed to change future market transactions with respect to physician charges and output. Such pricing information programs are more likely to result in agreements in restraint oftrade than are exchanges of descriptive data, which merely describe or reflect historical or ;urrent market transactions.6 Indeed, there is a danger that use ofthe lVG could lead to an agreement among physicians on a single converion factor to apply to each service on the RVG. Thus, the RVG could easily become the means for physicians in at least some communities o coordinate a collusive pricing scheme.

Further, for a number of reasons there appears to be a substantial See generally MOl1anto v. Spray.Rite Service Corp. 104 S.Ct. 1464, 1471 (1984) (To find an agreement (tJhere 1St be evidence that tends to exclude the possibility that the manufacturer and the nonterminated distributors re acting independently.. . (TJhe antitrust plaintiff should present direct or circumstantial evidence that sonably tends to prove that the manufacturer and others 'had a conscious commitment to a common scheme signed to achieve an unlawful objective.''' (citation omitted));Nat'lFirstBank u.Cities Service Co., 391 U.S. 253 -88 (1962) (the inference of a conspiracy does not JogicalJy follow in the ab!Ince of either direct conspiratorial leDee at motive to enter a tacit agreement). See Goldfarb v. Virginia.Slate Bar 421 U.S. 773, 781 (1975); Maple Flooring Mfrs ' Ass n u. United Slates, 268 563 585-86 (1925) ADVISORY OPINION 509 danger that concerted adherence to the BVG by physicians in response to ASIM's invitation would be widespread. The invitatioilto use the RVG would emanate from a leading national medical specialty association and would presumably have the support of the other medical organizations that would have helped to build the "consensus" the RVG reflects. Concerted adherence to an ASIM RVG would appear more likely than if an independent outside organization were to formulate an RVG. Moreover, ASIM' s invitation to adhere to the RVG would be attractive to the many primary care physicians who would benefit financially from increased reimbursement forcognitive services. Also, the RVG would be circulated in a form easily used by individual physicians in setting their prices. It would require only that the physician identify the appropriate code for each medical service rendered and apply a conversion factor he or she selects to each Jisted relative value to determine his or her charge for every service.

Widespread adherence to the RVG by physicians in local communities would likely be interdependent because it would probably not be in the economic self-interest of individual physicians to charge on the basis of the RVG unless they believe most competing physicians would be doing Jikewise. Physicians choosing to price in conformance with the RVG would likely do so to effect increases in the absolute If only a few physicians level of their charges for cognitive services.7 were to increase their charges for cognitive services, insurers might refuse to pay the increased amounts on the ground that for each such physician, it reflected a fee exceeding the "usual and customary charge of internists. In light of increasing competition at the primary care level, individual physicians considering adherence to the RVG would know that patients who were not fully reimbursed by insurance and who incurred higher out-of-pocket costs for cognitive services could over time go elsewhere for their medical care (e. to other private practice physicians or to health maintenance organizations). If most primary care physicians in an area adhered to the RVG, in contrast, insurers ' Ucustomary" screen levels would over time increase, likely resulting in higher reimbursement allowances. Patients would then have less incentive to seek, and could less easily find, a lower-cost provider. Thus, concerted or interdependent conduct by a very substantial number of physicians could succeed, and would probably be necessary to succeed, in raising the relative price level of cognitive services. If concerted conduct were not necessary, physicians concerned about the disparity identified by ASIM could address it unilaterally, and presumably already would have in their own 7 Adherence to the RVG would likely require physicians to depart from their current fee schedules- It would be unkely that physicians would voluntarily elect to confonn to the RVG and choose conversion factors that would keep their prices for cognitive services at roughly their current levels and would result in redltction oftheir charges for procedural services practices, by increasing charges for cognitive services. ASIM's likely knowledge that individual physicians probably could only effectively use the RVG interdependently, or concertedly, would help support a finding that ASIM contemplated a concerted response by physicians to its promulgation of the RVG.

Finally, to the extent third-party payor coverage "desensitizes insured patients to price increases, concerted conformance to the RVG in an effort to raise charges for cognitive services would be more likely contemplated, attempted, and successful in the medical marketplace than restraints of trade in other market contexts in which consumers are more "price sensitive. " If the RVG were adhered to by a substantial number of physicians, then "discounting" might not be as advantageous for physician competitors as it typically is for other competitors seeking to undercut higher charges resulting from collusion. As noted above, general adherence to the RVG would Jikely create a new range of "customary" charges, so that third-party payors would, as time passes, likely recognize and pay higher charges for cognitive services. In this event, even if some physicians did not adopt the RVG and charged lower prices for cognitive services, they might not be able to undercut effectively those adhering to the RVG because fees below those recognized as customary by insurers might not attract many insured patients away from other physicians. Patients with paid-in-full insurance coverage or only small co-payment obligations, who are treated hy physicians whose fees are within the "customary" range, could have little, if any, monetary incentive to switch physicians.

The Commission recognizes, notwithstanding the foregoing discussion, that substantial arguments can be made against the likelihood that concerted adherence to the RVG would result from ASIM' proposed conduct. For example, physicians' different cost structures and diversity across the country in practice patterns and pricing relationships among various medical subspecialties may make it unlikely that physicians would reach a common understanding to utilize any single RVG. Also, the cost-containment practices of third-party payors would pose a major obstacle. Nonetheless, although the Commission cannot, in this advisory opinion context, predict that widespread concerted conformance to the RVG would necessarily result from its dissemination by ASIM, the available information on this specific RVG proposal indicates that this type of agreement in restraint of trade is a substantial danger.

ASIM also proposes to disseminate its RVG to insurers and other third-party payors and encourage them to adopt the RVG as a basis for their reimbursement structures. This conduct raises the question of whether ASIM' s proposal may lead to the second type of possible y, ADVISORY OPINION 511 tween ASIM, on behalf of its member, and third-party payors that such third-party payors wil adhere to the RVG in their reimbursement systems. Such an agreement between ASIM and a third-party payor could lessen competition among ASIM's members over the terms of their dealings with the third-party payor. The Commission does not find a substantial danger that this type of an agreement would result from the proposed conduct.

No agreement in restraint of trade involving ASIM occurs if a third-party payor decides to adopt an RVG as the basis for its reimbursement system, even if its decision results from discussions with ASIM, so long as the third-party payor s decision is a unilateral one is not the result of coercion by ASIM or of an agreement coerced or voluntary, between ASIM and the third-party payor. In this regard, ASIM's request letter states that its discussions with third-party payors would he advisory and not coercive in nature. ASIM also states that it would not be acting as a common agent or in a representative capacity for its members in its dealing with thirdparty payors; rather it would simply seek to persuade third-party payors of the effcacy of reimbursement systems based on the RVG. Based on these representations and the absence of factors indicating serious risk in this regard, the Commission does not believe there is a substantial danger that ASIM wil negotiate an agreement with, or coerce, third-party payors to use the RVG, so as to constitute an agreement in restraint of trade)O Reasonableness of Potential Agreement in Restraint of Trade Because the Commission has concluded there is a substantial danger that ASIM's proposed conduct would involve an agreement in restraint of trade among ASIM and physicians to concertedly adhere to the RVG, the remaining issue is whether such an agreement would unreasonably restrain trade and therefore be ilegal. The Commission concludes that such an agreement would be inherently suspect, in Jight of its purposes and likely anticompetitive effects. The Commis. sion further concludes that the likely anticompetitive effects of sucl an agreement probahly would not be outweighed by any countervaiJ s Because of the conclusion reached in this advisory opinion it is not nl)cessary to reach the question ofwheth an agreement to adopt and promulgate an RVG by a medical society cdl ld result in an unreasonable ft:straint trade, even absent a finding of concerted adherence to it by phygicians or an agreement between the medica) HOC), aod any third.party payors See Virginia Academy of Clinical Psychologists u. Blue Shield of V4. 624 F.2d 476, 483 (4th Cir. J980), e denied 450 U.s. 916 (1981);Michigan State Medir:l Soc 101 F. C. 191 286 (198.1) MSMS" ); Monsonto Ct. at 1471 10 ASIM' s dcvelopment of an RVG for dissemination to third-party payors, although not raising an appa ubstantial danger of agreement in restraint of trade between ASIM and third. party payon!, CQuld noneth, result in an agreement between ASIM and its members to adhere to the RVG. If ASIM were to develop an Ilnd support it in discus-Gians with third-party payors, its members' knowledge of this action and the 8pp.cif the RVG col1ld result in concerted adherence to it for many of the same reasons stated above. Although evidentiary situation would be different., this conduct could raise antitrust risks like those resulting from, dissemination of the RVG to ASIM's members with encouragcrnet1t to use it. , y, ), ing effciency justifications that may flow from ASIM's proposed conduct. As a result, as is discussed below, an agreement among physicians to adhere to the ASIM RVG would be likely to restrain trade unreasonably. The Commission, therefore, cannot approve ASIM' proposed actions.

Naked horizontal agreements to restrict output or tamper with price are per se ilegal. No elaborate inquiry into market power or actual effects is required for condemnation of such agreements under the antitrust laws, and insistence on the "need" in the marketplace for such arrangements cannot provide a defense'!! An agreement on precise fees is not required for a finding ofiJegality,!2 "Any combination which tampers with price structures is engaged in unlawful activity. . . . (T)o the extent that they raised, lowered, or stabilzed prices they would be directly interfering with the free play of market forces. 13 As the Supreme Court has recognized, "An agreement to payor charge rigid, uniform prices would be an ilegal agreement under the Sherman Act. But so would agreements to raise or lower prices whatever machinery for price-fixing was used. 14 Thus, an agreement to use a particular formula, like an RVG, could support a finding of ilegal price-fixing,'5 because "tampering with the means of setting prices is tantamount to tampering with reimbursement levels. 16 Similarly, the Supreme Court has condemned as per se unlawful a horizontal agreement to fix only one element of price, such as credit terms,!7 If an agreement encompassing promulgation and adherence to the I\SIM RVG is found, the agreement would have purposes and likely ,ffects that under the foregoing precedent would condemn the agree- Gent as per se unlawful absent plausible effciency justifications. The agreement would tamper with market pricing structures, and pose a "ious danger of higher prices, at Jeast with respect to some medical 'rvices, and other anticompetitive effects. United Stale v. Socol1y- Vacuum Oil Cu. 310 U.S. 150, 218 (1940);NCAA v. Board of Regents of University )k/ahomu, 104 S.Ct. 2948, 2960, 2965 (1984). Per se cond.emnation has been deomed appmpriat€ even when 1) suggested prices were used hy a trade dation s members only as a "starting point" for individualnegotialiol1s and price competition continued wut/! IJealersAss n of N Cal. v- United States,279 F.2d 128, 132 (9th Cir. 1960), and 2) there were no sanctions 1St members not adhering to the suggested prices and suggested prices in fllet were not strictly adhered to members United States u. Nationwide Trailer Rental $Y8. , Inc. 156 F.supp. 800, (D, Kan. afrd per curium S, 10 (1957), In regard to "advisory" price schedules possibly supporting a finding of agreement in restraint de compare dictum in Goldfarb 421 U.S. at 781 (1975) ("(a) purely advisory fee schedule issued to provide mes without a showing of an actual restmint on trade, would present us with a different question with i States v. Nation.al A8. n of Rent Estate Ed-s. 339 U.B- 485, 488-9 (1950) (in regard to a "non-mandatory hedule lsJubtle idf1uences may be just as effective as the threat or use of forma! sanctions to hold people izona v. Maricopa COtinty Med. Soc 457 U.S. 332, 346 (1982) (quotingSocony- Vacuum). See ols" MSMS at291 on.y. Vacuum 310 UB. at 222 (€mphasis added). Morrison v. NissanMotor Co., 601 F.2d 139 (4th Cir. 1979) (regarding an autumobi1€ r€pair flat rate I-S, 101 F, C. at 291- '10M, Inc. v. TargetSQlcI$, Inc., 446 U.S, 643 (1980), ADVISORY OPINION 516 First, the agreement would tamper with the market' s pricing structures by locking competing physicians into use of a particular pricing formula, if not uniform prices. It would fix the relationships among each physician s prices for different services, so that ratio would not depend upon the production costs or quality of each physician s service, nor on the degree of demand he or she faces for various services. Second, use by physicians of the proposed RVG would apparently be designed to achieve, and would likely result in, payment and reimbursement for cognitive services at higher absolute levels than prevail currently. It can be inferred from ASIM's own statements that ASIM' s purpose in developing the RVG includes raising . prices for cognitive services on an absolute basis as well as on a relative basis. Examples include an ASIM resolution in 1983 stating: "Resolved, that in ASIM's campaign to reduce the discrepancy in reimbursement between cognitive and procedural services, the Board of Trustees continues to actively promote enhanced reimbursement for cognitive services. . . . 18 In proposing in 1982 that the Department of Health and Human Services adopt an RVG demonstration project, ASIM stated, "A new schedule of allowance providing for increased reimbursement for internist's cognitive services would he created. 19 Moreover, ASIM very likely knows that its members would have every incentive to use the RVG to increase the absolute level of their prices for cognitive services. As noted above, physicians who voluntarily convert their current fee schedules to the new ASIM RVG would be unlikely to adopt a conversion factor that would result in Jower prices for procedural services and no increase in their cognitive services charges. The effect of widespread use of the ASIM RVG by physicians to bil higher fees for cognitive services, even without the RVG' explicit adoption hy third-party payors, would likely be incorporated into third-party payors' physician fee profie data for computing usu- , customary, and reasonable charges and raise third-party payors reimbursement levels for cognitive services. Third, a "fragmentation" phenomenon of new biling categori81 being created has apparently arisen with use of some other RVGs an. could in the instant case result in increases in overall biling charge, For example, a study of the California Medical Association RVG CO! cluded that more detailed and fractionalized RVG descriptive cod, resulted in overall increases in payments to physicians.20 Here, the is some danger, for example, of ASIM's RVG providing for ll charges by procedure-oriented physicians for cognitive aspects ofs, vices, when patients were previously charged only for a procedu '8 ASIM, Reference Committee B Report 1983 House of Delegates 4 (emphasis added) 19 ASIM, Proposal (or Demonstration Project 3 (Ocl 1982) (emphasis added). 20 Sobaski Health Ins. Statistics USDHEW Effects of the1969 California Relative Value Stltdies on Physician Services Under SMI Pub. No (SSA) 75-11702 (June 20 , 1975), at 5; see aim; Urban Institute Alter Method of Developing Relative Value Scale of Physicians ' Services: Year End Report 7-8 (1983) (" Institule Study , D14 FEDERAL TRADE COMMISSION DECISIONS Such a tendency in the ASIM RVG could arise in the evolution of the consensus" needed among both cognitive and procedure-oriented physicians for the RVG's contents.

Fourth, widespread adherence to the RVG could also tend to stabilize prices artificially. Such a phenonemon could be in contrast to the stability of price one might expect in a competitive market in which homogeneous, fungihle goods are sold. The price relationships of different services, and possibly ahsolute prices as well 22 would be stabilized to a degree not already effected by third-party payment and without regard to differences in the quality of each physician s services or his or her effciency.

Fifth, the RVG may also faciJtate direct price fixing. In the absence of an RVG the diffculties in forming a consensus among physicians on a fee schedule would involve deciding which services to include in a price-fixing ageement and agreeing on what value each service should have in relation to another. Agreement on a full-blown fee schedule would be faciltated by adherence to the RVG and would involve additional agreement only on a conversion factor. Although ASIM has disclaimed any intent to encourage such conduct, agreements among physicians to use a particular conversion factor with an RVG can arise. Agreement on conversion factors does not appear to be a part of ASIM's plan nor an inevitable result of it, but it is a possible result of ASIM's conduct, particularly in subspecialties at local levels.

Finally, in addition to affecting price, concerted adherence to "-SIM's RVG would also appear to fix or restrict output of certain ervices, also a type of agreement that can be per se ilegal. ASIM' lated intent is to change the mix of cognitive and procedural services elivered by internists, if not all physicians, through change in reimlrsement levels. ASIM's proposal apparently contemplates a reducm in the output of procedural services, and could, depending on the pact on demand of any significant price increases for cognitive vices, reduce output of cognitive services. 3ecause of its apparent purpose to raise price levels for some seres and the substantial danger of anticompetitive effects on price e Am. Med News, Nov. 23/30 1984, at30, col 3 (au.gical80ciety otIcial quoted as stating there is "cognition OR (operating roo011 too ); see alsa Am. Med. News Oct. 14, 1983, at 14, co!. 1 (gurgica! society offcial quoted ing "surgeons use cognition before, durng and aftr surgery; physician quoted a8 stating that "resistance :eons to concept of reducing cognitive and procedural aervices disparity began to "diaappear" in state 1 society when internsts explained to surgeons that hecaul! oftbe reimbursement bias surgeoll do . cognitive consultations that they have to consider throwaways" or "time lost" empirical study suggests that HVG use is assiated with less fee dispen;ion, although not necessrily fees. B. EiBeuberg,Informa.tiun Exchange Among Competi/urs: The Issue o( Relative Vulue Scales. for ' Services, 23 J.L. & F.-on. 441 , 457-58 (1980). The study s data were not suffcient to indicate whether ction in dispersion reflected more effcient market perfonnance, or a p()SIibly unwarranted tret1d toward i7.ed prices by physicians offent1g differing qualily service. The study did fit1d a positive a/lodation RVG U8C Hod higher price!!, but the assciation was not statistically gjgnificant. A, 104 S.Cl. at 2948;National Macaroni Mfrs. v. FTC 65 F. C. 583 (1964),a.frd, 345 F.2d 421 (7th Cir. , ), ADVISORY OPINION 515 and output, the agreement to adhere to the RVG that could result from ASIM's proposal would be inherently suspect or prima~ facie anticompetitive. Such an agreement would not be condemned outright under the per se rule as a naked restraint of trade if a plausible procompetitive effciency rationale existed for it, but the burden would be on ASIM to establish justifications Jegitimizing the agreement. ' If ASIM established procompetitive effciency justifications they would then be weighed against the anticompetitive effects of the conduct to determine net competitive effects. If such justifications were not shown to be valid, the prima facie anticompetitive or inherently suspect conduct in question would be condemned without further proof of anticompetitive effects. This method of analysis can be deemed a truncated, quick-look, or limited rule of reason analysis. Showing adequate justification for concerted promulgation and adherence to ASIM's RVG would he particularly critical given the power of ASIM and those physicians, both ASIM members and others who might concertedly use the RVG to effect significant changes in the marketplace. ASIM membership includes 19 000 physicians, a significant portion ofthe nation s 63 000 internists. The Commission understands that ASIM also has the support of at least 12 other physician organizations in its effort to make cognitive services reimbursement more "equitable." Concerted action by physicians who are members of these organizations to adhere to the RVG would Jikely have a substantial effect on the marketplace. In addition, primary care physicians who are not members of these organizations might also he attracted to the ASIM RVG because use of it would be in their financial interest. FinaUy, future proposed ASIM RVG activity encompassing aU physician services could command widespread, acrossthe-board adherence by physicians in aU specialties. 2- NCAA 104 S.Ct. at 2967; cf, MSMS 101 F. C. at 291; seeRl'nerally Brunet Streamlining Antitrll.'/ Litigation By "Paciol Examinu.tiu/I Of Restrains: The Burger Caur/And The Per Se Rule Of Reusan Distinction 60 Wash L. Rev. l (1984).

25 United StatestI. American Soey of Anesthesiologists, Inc. 473 F.supp. 147 (S. Y. 1979) ABA" cited by ASIM, warrants discussion, In that case, the court rejected the Department of Justice s contention that AS/' committed a plr . violation of the Shennan Act through its dissemination of an RVG fot anesthesia services am fOUDd no violation under the rule of reason. TheABA case was tried solely on pera 51' theory 80 there was no fu exposition of possible anti ompetitive effects. The court, in fad, found no agreement to adhere to thlJ RVe wit the purpose or effectofrai ing or tabilizing price ASA at 159, and instead found that ABA had not "encouragei anyone to use its RVG. No evidence was cited showing any intent on the part of ASA to achieve an j!1crea fee levels. ASA at 159-0. ThlJ court also empha ized that delivery of anesthesia services i somewhat unique medical practice-little or no contad with patients prior to surgery and virtually 100 percent insurance covera of fees-ld. The present matter differs significantly in these respects. 2fiSee, e. , General asewnys v. National Truck Lensin,; Ass 744 F.2d 588, 59:)-96 (7th Cir. 1984) (prelimim injunction); Brief for the United Swtes AmicusliS Curiae in Support of Affnnance NCAA at 9-12 In referel to the quick-IDok rue of reason (sJeasoned antitrust lawyers recognize that the threshold facial examinatio, Dot that novel and is entirely consistent with older landmark cases.United S ates I). Addys/on. Pipe and Steel evideuces the historical foundation underlying the quick look method. " Brunet supra note 24, at 22. Even full-blown rule of reason analysis were the appropriate mode of analysis for an aweement encompassing conee adherence to ASIM's RVG, the apparently anticompetitive purposes and potential effects discussed above, an( likelihood that ASIM members and other physicians coj1ectively using the RVG could exercise market pow. discussed below, would very likely make out aprima facie case, once estabiished in an evidentiary rp.cord. would then, as in a truncated rule of TeaSOD analysis, have to proffer evidence showing proeompetitive efTe' greater or at least equivalent weight uJ.O FEDERAL TRADE COMMISSION DECISIONS ASIM' s stated justification is esseniialJy that imperfections in the insurance payment system for reimbursing physicians have created wrong incentives a high level of reimbursement for costly technological and procedural services, which encourages overuse and more expensive medical care, and a low level of reimbursement for cognitive services, which discourages their use. ASIM further claims that redressing the reimbursement disparity between procedural and cognitive services through its proposed RVG would encourage greater use of more personalized cognitive services and provide new incentives for more physicians to choose primary care specialties. ASIM claims that its proposed RVG, besides influencing physicians to hetter meet the public s overall health needs, would be designed to reduce health care costs.

ASIM' s purported objective-a lower-cost medical services marketplace, with concomitant health benefits to patients-is laudabJe. However, that objective would not provide a cognizable justification or defense under the antitrust laws for an agreement to supplant determination of prices by market forces on the ground that prevailing prices were not at a level the parties to the agreement believed was optimal for them or society.27 In Professional Eng the Supreme Court confirmed that activities of professional societies are subject to the traditional antitrust test of reasonableness- whether the challenged agreement is one that promotes competition or one that suppresses competition -and may not be defended on the ground that the special characteristics of professional services markets make comoetitiveJy determined prices undesirahle.

The diffculties in recognizing the availability of such a defense for II agreement to adhere to the ASIM RVG are ilustrated by the issues hat would have to be resolved to determine its validity. A principal lctual issue would be determining the accuracy of ASIM' s claim that ricing levels and output in the medical services marketplace are not . appropriate levels. If such nonoptimal performance is proven, one Juld then have to determine whether the results of the conduct in estion would be improvement or worsening of the market. A court ght well have to assess the likely result of physicians forming and ;ing upon a subjective consensus judgment, based on cost and other tors, of what pricing relationships would prevail in the market e the market working properJy. This inquiry would be akin to the ulatory determination ofa public utility commission and would ond any inquiry undertaken in prior antitrust cases. Indeed, to mpt to resolve empiricalJy whether competition and consumers Id ultimately be served or harmed by concerted agreement on a ng formula would require an inquiry that courts have long esred in antitrust cases- , to Hset sail on a sea of doubt" seeking Nation..l Sucy of Prof. Eng rs v. United Stutes 435 UB. 679, 688, 692 (1978) !!. g., ADVISORY OPINION to decide "how much restraint of competition is ill the public interest and how much is not " with the court trying to assess the reasonableness of the prices charged. Also, even if it were demonstrated that the market changes ASIM proposes would in fact produce prices and output at a more optimal level in the immediate short term, they could over time produce unreasonable prices and output, with it being virtually impossible to police the ongoing effects of such concerted use of ASIM's pricing formula.

Even if a defense by ASIM premised on the appropriateness of agreed-upon changes of industry pricing structures and output levels were legally cognizable as an effciency-enhancing device;1t is doubtful that ASIM could successfully establish, on the facts, that market performance would improve through its proposed conduct to a doser approximation of optimal market pricing. For example, ASIM' proposal may drive up those prices that are now dose to or at a competitive Jevel, while Jeaving largely undisturbed prices for procedural services that may he reimbursed excessively. It is possible that market forces may be permitting above-optimal prices for procedural services, while keeping the prices of cognitive services at approximately optimal levels.3D If so, ASIM' s efforts could raise cognitive service prices above competitive levels. Thus, if it enhanced reimbursement for cognitive services, concerted adherence to ASIM' proposed RVG could distort the market to a point even further from optimal competitive performance than now exists. There is certainly no assurance that the price of procedural services wil be reduced by the ASIM RVG in the long run. ASIM' s efforts to reduce the reimbursement disparity between procedural and cognitive services have reportedly met with concern from representatives of some internal medicine subspeciaJty groups whose members engage more heavily in procedural services. These groups reportedly do not object to increasing reimbursement for cognitive services, but question a decrease in reimbursement for procedural services." It i. very possible that, once consensus on the RVG is reached, use in th, market ofthe relative values accorded different services would resuJ in increases in reimbursement for cognitive services with Jittle or n decrease in reimbursement for procedural services. Moreover, 1 reach consensus among physician representatives with divergent i 2" United States u. Addyston PijJ Sled Co. 85 F. 271, 283-84, 291 (6th Cir. 1898),a.(rd as modified 175 211 (1899).See also MSMS 101 F.T.G. at 293. 29 See Uniled Stotes Tnmton Potteries Co. 273 U.S. 392, 397-98 (1927) 3B Market forcl's that may be restraining the price of cognitive services to a greater degree than procca Aervices could include better COOi\umer knowledge of what a "fair" price is for mgnitiv€ services;mote acti' effective consumer involvement in determining whether and when to Reek primary care services; a gr proportion of out-or.pocket costs for patients ruceiving cognitive services because of the term of indu coverage; the growth of ambulatory care centers that have extended hounI; and growing competition frOII physician health care providers.See, e. AMA Council on Med. Service Effects of Competiti(Jn in Medicin 1. AM.A. 1864 (Apr. 1983); ASIM Reimbursement for Physician. ' Cognitive and Procedural Services: A Puper 1 (Jan. 1981).

3' See Am. Med, News, Ocl 14, 1983, at 14, coj. 1 ,HO FEDERAL TRADE COMMISSION DECISIONS terests, compromises might result in identification of new cognitive services, not previously biled for, that procedural service oriented physicians can bil to insurers and patients. In addition, ASIM's implicit prediction that physicians would switch to providing more cognitive instead of procedural services and thereby contain health care costs is speculative. Even if some switching did occur, would it be enough to onset any increase in the price of cognitive services so as to Jower overall health care costs? Would the resulting output be more heneficial to consumers than the current one? These questions demonstrate the risk inherent in permitting price and output mix to be determined or redirected by private agreement among competitors who have a stake in the outcome. Finally, the means ASIM plans to use to develop its RVG highlights some of the dangers. ASIM proposes to derive appropriate relative values through a "consensus" building process-pollng physicians by means of the Delphi technique. A recent study on the pros and cons of various relative value guide alternatives prepared for the Department of Health and Human Services, noted that if the Delphi technique is used, the representational nature of the polled group is critically important.34 Physicians "would have a financial stake in the outcome of the RVS determinations and thereby have a substantial 35 The studyconflct of interest if em panelled to determine an RVS. explains:

Po the extent that various goals of an RVS would be encouraged as part of a group lecision process, e. , (the goal ofj more adequately reward(ings cognitive services, the roeess becomes less that of finding a solution and more that of achieving the most olitically acceptable choice. The findings of research on formal, group decision-making 'T problem solving tasks (showing the potential effcacy of such efforts) are unlikely be valid for group choice tasks in which participants have a stake in the outcome d no objectively correct solution exists. Id. 'r this and the other foregoing reasons it is not at all clear that lcerted use of the ASIM RVG would achieve the cost reductions and leficial public health policy results ASIM has projected. )ther possible justifications are also unlikely to provide an ade- '.te ground of defense. Widespread adherence to a single RVG could .ide a common benchmark for physician pricing. Arguably, this d facilitate enhanced price competition and comparison shopping ng physicians by consumers and health plans on the basis of the rent conversion factors used by physicians. Some procompetitive fits of this sort could conceivably result from standard adherence ;ingle RVG, but it is entirely speculative how substantial those wpm p. 12.

104 S.Ct. al2948.

o.son Group Decision-Making Me/both, in Urban Institute Study. sl.pm /lote 20, at 123. 121.

ADVISORY OPINION 0)1.;: benefits would he and they would likely hot outweigh the anticompetitive impact of concerted use of the RVG. Although insurers could possibly have benefitted significantly some years ago from such pricing by physicians, most insurers now have or can obtain computer profies on physician fees that provide data on pricing differentials among physicians.

RVGs in some contexts can serve the legitimate, unilateral business needs of third-party payors, promoting competition and effciency. An RVG adopted for use by an insurer, self-insured employer, health maintenance organization, or the government for its ayn use as a third-party payor could well be valuable. Here, when it horizontal agreement among physicians to adhere to the ASIM RVG is a realistic possibility, it is also possible that some procompetitive effciency benefits could be achieved from its unilateral use by individual thirdparty payors. It is, however, unclear how substantial such benefits would be. More important, if third-party payors have had a critical need for an RVG, it is not clear why private entrepreneurs, research centers, or the payors themselves would not have already satisfied that need, with whatever physician consultation was necessary, short of medical society promulgation ofthe proposed RVG with its attendant risks.

Finally, informational benefits could flow from the availability of the ASIM RVG for unilateral use by physicians in the marketplace. Such henefits, however, would not be present when physicians conspire to adhere to the RVG, and do not merely use it as an informational tool. Such effciencies would not, therefore, appear to constitute a valid justification for the unreasonable pricing agreement that is a risk of ASIM's proposal.

The Commission concludes, on balance, that any procompetitiV! effciency benefits flowing from ASIM's proposed conduct would no be likely to outweigh the anticompetitive dangers of the agreemen to adhere to ASIM's RVG that is a serious risk of its proposal t develop an RVG raising the relative prices of cognitive services. Alternative Actions to Address the Cognitive/Procedural Disparit There are actions ASIM can take to further its goal of reducing t alleged reimbursement disparity between cognitive and procedu' services that would not appear to raise antitrust problems and tJ may be helpful to public and private third-party payors. To aid thi party payors in developing sound reimbursement programs and teria, ASIM has available a range of actions that do not requirf incurring antitrust risk through development of a comprehen Jo This information is based upon staff intervjew with representatives of large and small insurers adc partypayoradmin.strators ;n Commission stafr intervews of representatives of insurers genera.JJy indicated a lack of enthuaias" medical sodety developed RVG.

), u..v FEDERAL TRADE COMMISSION DECISIONS RVG and its dissemination to both third-party payors and all its member physicians. For example, ASIM can seek to persuade thirdparty payors to change their reimbursement methods or amounts without running afoul of the antitrust Jaws so long as there is no coercive conduct engaged in or threatened, nor any price agreement entered into between ASIM and any third-party payor lessening competition among ASIM's members. ASIM can lobby Congress or the Department of Health and Human Services for changes it desires in physician reimbursement. Expressions of opinion on the policy question of reducing the reimbursement disparity between cognitive and procedural services as would be contained in an ASIM "white paper do not constitute a restraint of trade and also fall within the ambit of protected free speech. Finally, ASIM can conduct research and analyses that could be used with other information by the Department of Health and Human Services or other third-party payors in constructing an RVG. ASIM could, for example, study, analyze and report on the time, complexity, or costs of specific services performed by internists without developing a formal RVG mechanism and disseminating it to ASIM's member physicians. Conclusion The Commission has determined that the danger of an anticompetiive agreement in restraint oftrade is suffciently great that it cannot ive approval to ASIM's proposed course of conduct. ASIM can lough, legitimately engage in alternative actions to redress the reimlrsement inequities that it perceives. This advisory opinion does not fleet a determination by the Commission that ASIM's proposed nduct would necessarily violate the antitrust laws if undertaken. " does it denigrate ASIM's concerns about the public health and it implications of current third-party payor reimbursement patns. Rather, the Commission has determined only that advance "oval cannot be given for the specific actions ASIM has proposed. s advisory opinion, like all those the Commission issues, is limited he proposed conduct about which advice has been requested. direction of the Commission.

Letter of Request March 1, 1984 Mr. Muris:

behalf of our client the American Society ofInternal Medicine we hereby request an advisory opinion pursuant to 16 1.1 to 1.4, concerning the development and dissemination of ADVISORY OPINION 521 a relative value guide designed to reduce the disparity in reimbursement between cognitive and procedural physician services. I. Bac kground The American Society ofInternal Medicine is the national association of physicians specializing in internal medicine. ASIM believes that current reimbursement policy provides the wrong incentives for utiizing physician services and that these incentives need to be reevaluated. For example, under present insurance payment systems physicians are encouraged, by the high level of reimbursement, to use costly technological and procedural services-EKGs, X-rays and surgical and diagnostic procedures. At the same time, physicians are discouraged, by the low Jevel of reimbursement, from utilizing more time consuming cognitive services--iagnosing patient health problems, providing preventive education, and evaluating lifestyle effects on health.

ASIM believes that the physician reimbursement system should he revalued to increase the relative value of cognitive services and decrease the relative value ofproceduraJ services, in order to encourage the use of more personal services and discourage the overuse ofprocedural services. It is further believed that an increase in the relative amount at which cognitive services are reimbursed, as compared to procedural services, would encourage physicians to spend more time and effort in personalized aspects of care. The increased value of cognitive services should also provide new incentives for more physicians to choose primary care specialities utilizing relatively large amounts of cognitive services.

II. Activities in Question ASIM wishes to promote the concept of changing the physiciar reimbursement incentives, away from costly technical services an' toward less costly personal care services-the cognitive services. IJ this effort, ASIM wil develop and disseminate its evaluation as t how cognitive services can be valued when compared to procedun services to supply an alternative to the current reimbursement sy tem. ASIM wil also propose relative value guides for listed medic, and surgical services based on factors designed to reduce the discre ancy between cognitive and procedural services. Initially, ASIM proposes to develop relative value guides only f those cognitive and procedural services provided by physicians w specialize in internal medicine. ASIM intends in the future to we with other physician organizations-including the surgical society to develop relative value guides for all listed medical and surgi services, also based on factors designed to reflect relative differen in the cognitive skils and costs required to provide each service These relative value guides ("RVG proposal") wil contain no conversion factors or pricing information of any kind and wil in no way constitute a fee schedule but, instead, will merely list medical services against which relative values wil be assigned based on a listed set of relevant factors. As a result, the RVG proposal wil merely describe the relative cost and diffculty of certain cognitive services in relation to other services. In this regard, many third party payors agree with the need to reduce the disparity in payment incentive but point to the unavailabilty of any RVG that accomplishes the desired result. The development of such a relative value guide for internists' services wil necessitate that ASIM consult with various other physician organizations representing the subspecialties of internal medicine (cardiology, gastroenterology, infectious disease, allergy, et aI). Each organization wil be asked to share with ASIM its assessment of the relative time, complexity, training, and overhead costs involved in each service commonly provided by internists in its subspecialty, and the recommended relative value for each based on these factors. ASIM wil then take these recommendations, develop a draft RVG and attempt to build a consensus among the various internal medicine organizations in support of the new RVG as a valid basis for reducing the disparity between cognitive and procedural services. The RVG proposal wil be developed in a "white paper" which wil explain the problems caused by the cognitive-procedural service reimhursement disparity and wil provide written examples, by way of relative value guides, as a method to reduce such disparity. The white paper" wil ilustrate "how to" use the relative value guides ;0 change the reimbursement structure from the current procedural ,ervice basis to a cost of resources basis. The purpose of the "white mper" wil be to pursuade and it wil itself emphasize its voluntary nd advisory nature. It wil contain no explicit or implicit threats of ny kind against physicians, payors or purchasers. Nor wil ASIM 'sue any such threats in connection with its dissemination or otherise.

ASIM intends to send the RVG proposal to its members, third party iyors, and preferred provider organizations C'PPOs ) and request at they consider using the RVG proposal as a guide to develop a imbursement program consistent with the approach contained in e proposal. ASIM also intends to hold follow up discussions and ,etings with interested payors. ASIM wil further advise each ofits ysician members to evaluate the proposal and wil recommend that vsicians supportive of the concept request a reevaluation of the ltive values of physician services in existing or proposed PPO angements, and to participate in any such reevaJuation. ASIM wil , however, engage in coercive conduct of any kind in this connec- , nor wil it serve as common agent for any of its members which , J\U".iI_ may be negotiating payments, pricing and reimbursement issues with PPOs, third party payors or purchasers, or other persons or organiza- Finally, to facilitate the critical discussion of the issues involvedtions. and to promote the concept of RVG within the medical profession, ASIM wil exchange information regarding the relative value ofvarious services with the American Academy of Family Physicians, the American Academy of Pediatrics or other primary care organizations. However, no price information of any kind wil be considered, discussed, or transmitted in any such communication, nor wil any explicit or implicit threat or coercive conduct of any kind be undertaken.

III. Statutory Provision The foregoing activities raise questions under Section 1 ofthe Sherman Act' and Section 5 of the Federal Trade Commission Act. IV. Present Law It is clear that agreements to fix prices or to set fee schedules, or other arrangements which produce anticompetitive effects, violate the antitrust laws. Arizona v. Maricopa County Medical Society, 102 Ct. 2466 (1982) Maricopa ' Tbe Supreme Court has recently determined that an agreement to establish a fee schedule constitutes per se price fixing. See Maricopa. Although the Federal Trade Commission ("FTC") has not litigated that issue, it has obtained consent orders by which certain professional association agreed to withdraw and to refrain from developing in the future any RVG which also contained conversion factors. In In re MinnesotaCalifornia Medical Association 93 F. C. 519 (1979); State Medical Association, 90 F. C. 337 (1977). Nevertheless, the mere development and circulation of an RVG, without conversion factors, is price neutral and does not necessarily have an anticompetitive purpose or effect. Moreover, activities proposed herein would not involve the implementation of any particular RVG. Indeed individual physicians, third party programs and purchasers of health care all would be explicitly advised to make independent decisions regarding the implementation of any such proposal. Finally, it is significant that the ASIM proposed RVG is designed to reduce total health care costs by creating incentives to J lSU. C.1 215 U.S.C. 45 :, Although the complete facts are not ktlown, the I"TC has alao made agreements and obtained con8etlt orderl wherein certin professiotla! society have agreed, among other things, to refrain from developing relative vau. studies where no mention of conven3ion factor waa made.In re Amerimn College ofRQdioLogy, 89 F. C. 144 (1977: In reAmericQ'I College ofOrthopoedicSurgeons, 88 F. C. 968 (1976);In reAmerimn College ofObstetricians an. Gynecologists 88 F. C. 955 (1976).

). ). ). substitute low cost care for high cost care. Many of the existing physician reimbursement systems create precisely the opposite incentive. Judicial precedent has determined that without conversion factors the circulation of relative value guides among association members, individual physicians, component societies, agencies of federal and local government and third party payors, does not violate the antitrust prohibitions against price fixing when the RVG is intended only as suggested methodology for arriving at appropriate fees. Jinited States v. American Society of Anethesiologist 473 F. Supp. 147 (1979). ("ASA" ). In ASA the court found that the publication and circulation of RVG did not amount to an agreement to use the RVG in pricing services and that the circulation of RVG did not curtail competition or interfere with the setting of prices in the marketplace. Id. Indeed, the FTC has recently addressed the issue of whether the mere circulation of even price related information amounted to price fixing and concluded that merely providing information, opinions or advice does not violate Section 5 ofthe Federal Trade Commission Act FTCA" In re Michigan State Medical Society, Dkt No. 9129 , 3 Trade Reg. Rep. (CCH) n 21 991 (Feb. 17 , 1983). ("Michigan Although the Commission in Michigan determined that the medical society s agreement to fix prices and to boycott third party payors violated Section 5 of the FTCA, the Commission exempted exchanges of information which did not amount to an imposition of a price related agreement. Accordingly, price related communications that do not involve an explicit or implicit agreement or an attempt to reach an agreement were held not to violate Section 5. The Supreme Court has also recognized that even the mere sharing of price related information does not in itself violate the antitrust laws. United States v. United States Gypsum Co. 428 U.s. 422 (1979). ("Gypsum In this regard the Court has stated that: ttJhe exchange of price data and other information among-competitors does not invariably have anticompetitive effects; indeed such practices in certain circumstances increase economic effciency and render markets more, rather than Jess, competitive. 438 U.s. 422, 440 n. 16 (1978). Thus, the Supreme Court refused to mandate the conclusion that a price fixing violation automatically follows from a finding that the exchange of price information has an impact on prices.

With regard to discussions with third party payors, an association may send information and may express the views of its membership. Michigan at 22,469 (where the Commission determined that the order 0 cease and desist did not reach noncoercive expressions of the medi- :al society s views or the views of its members when made to third ,arties).4 Moreover, advocating physician participation in decision , Although binding joint negotiation with third party payors may amount to price fix"'" , O:"'".; ADVISORY OPINION 525 making, does not create anticompetitive effects; as long as such advocacy is noncoercive. Michigan at 22,470; Maricopa 102 S.Ct. at 2477-78.

V. Conclusion ASIM believes that its activities with regard to the development and circulation of the RVG proposal wil not have a probable or foreseeable anticompetitive effect on physician s fees and should not be precluded by the antitrust laws. ASA; Michigan at 22,469. Indeed we believe that our proposal, if adopted, wil be procompetitive and reduce medical care costs. As previously indicated, ASIM' s efforts wil be purely advisory in nature and are intended only as a suggested methodology to reduce the disparity in relative values between cognitive and procedural physician services. In addition, no conversion factors wil be developed, proposed or disseminated so that the proposal is price neutral.

The circulation of the RVG proposal among individual physicians PPOs and third party payors wil not interfere with the PPO' s or third party payor s abilty to establish whatever pricing strategy they so desire. Moreover, ASIM wil not hind nor attempt to bind or coerce its members in any way. Each member wil continue to be free to make independent fee decisions.

We wil be happy to supply additional information and to discuss this proposal with you in person. We appreciate your prompt attention to this matter.

Respectfully submitted, Isl Wiliam G. Kopit Epstein Becker Borsody & Green preclude third party payor discussioJJS where each member reserves and exercises independence 11' pricinl siol1l1. Virginia Excc/sior Mils Inc. v.FIC("Virginia ) 256 F.2d 538, 541 (where the court noted that the a common agent is not ilegal when each member reserves and exercises independence jn pricing and materia! matters). Here, of course, ASIM has stated explicitly that it would not act as a bargaining agent "mhers or any other person or organization attempting to implement the RVG-

................ . . . . . .. .. .... ................................................................................................................................................................................................................................................................................................................ ............................ .. ..... ..... ..... ..... ..... ....

TABLE OF COMMODITIES * DECISIONS AND ORDERS Page Ai cleaning appliances. . . . . . . 317 Automotive parts. 410 Aviation weather detection systems. . . . . . . . 381 Computer products. . . . . . . . . .. 230 Consumer credit. . . . . . . 310 Descent systems. . . . . . . . . 280 Electrc adjustable beds. . . . . . 366 Intrastate property carrers. 406 Land sales. .. .. .. . . . .. . .. . .. . . .. . .. .. .. .. . . .. . .. .. Medical fee schedules. . . . . . . . 248, 277 Motor vehicles. . . . . . . . . . . . . . . . . 194 Nutritional supplements. . . . . . . . 291 Petroleum coke. . . . . . . 345 Petroleum products. . . . . . . . . 228 Residential housing builders and sellers. . . . . . . . . . . . . . . 250, 347 Scrap metal. . . . . . . . . . . . . 192, 246 Shoes. . . . . . . . . . . . . 191 . . . . . .. 1 , 304 . . . . . . . .. . . . . . . . . . . . . . . . reguation. Taxicab Ultrasonic pest control devices. . . . . . . . . 197 Commodities involved in dismissing or vacating orders are italidzed.