PHILLIPS, Chief Judge.
The Federal Trade Commission held that the acquisition of Fram Corporation by the Bendix Corporation violated Section 7 of the Clayton Act, 15 U.S.C. § 18.
Reversing the decision of its Hearing Examiner, the Commission directed divestiture of Fram. The case is before this court on the petition to review filed by Bendix.
The complaint was filed June 29, 1967, charging that Bendix had violated not only § 7 of the Clayton Act but also § 5 of the Federal Trade Commission Act, 15 U.S.C. § 45(a) (1). The § 5 charge is not before this court in the present proceeding and will not be discussed further in this opinion.
The complaint alleged violations of § 7 with respect to three lines of com
merce: (1) automotive filters,
(2) aerospace filters,
and (3) liquid separators.
The results of the § 7 violations were averred to be:
“(a) actual and potential competition between Bendix and Fram in the manufacture and sale of automotive filters, aerospace filters and liquid separators will be eliminated ;
“(b) actual and potential competition in the manufacture and sale of automotive filters, aerospace filters, and liquid separators, generally, may be substantially lessened;
“(c) Bendix will eliminate Fram as an independent competitive factor in the manufacture and sale of automotive filters, aerospace filters, and liquid separators;
“(d) concentration in the manufacture and sale of automotive filters will be maintained;
“(e) concentration in the manufacture and sale of aerospace filters and liquid separators will be increased substantially;
“(f) Bendix will obtain a decisive competitive advantage in the manufacture and sale of automotive filters, aerospace filters and liquid separators, to the detriment of actual and potential competition ;
“(g) additional mergers and acquisitions may be fostered with a resultant lessening of competition in the manufacture and sale of automotive filters, aerospace filters and liquid .separators; and
“(h) the entry of new firms and the growth of smaller filter manufacturing companies will be retarded, discouraged or prevented.” (1968-70 Transfer Binder) Trade Reg.Rep. ¶ 17,997 at 20,394-95. (FTC 1967).
As the case proceeded Complaint Counsel dropped several of these theories and eventually reformulated the case. Before the Hearing Examiner he proceeded on three separate theories:
(i) that the acquisition lessened actual competition between Bendix and Fram in the manufacture and sale of automotive filters;
(ii) that Bendix would confer “a decisive competitive advantage” upon Fram in the manufacture and sale of automotive filters; and
(iii) that Bendix was a potential entrant by internal expansion into the business of making and selling passenger car filters in the replacement parts market (known as the “aftermarket”), and that the acquisition of Fram eliminated Bendix’s “potential competition” in this submarket.
The Examiner rejected each of these theories, concluding that Complaint Counsel had failed to sustain his burden of proof. The first theory, lessening of existing competition between Bendix and
Fram, was abandoned by Complaint Counsel on the appeal to the Commission.
On appeal before the Commission most of the Examiner’s findings of fact and conclusions of law were accepted without change, and his treatment of the three theories was not disturbed. The Commission held:
“To the hearing examiner it was dispositive of this case that the evidence showed that Bendix would not enter the passenger ear filter aftermarket by internal expansion but only through merger with another firm already in that market. The examiner assumed that, from the standpoint of Section 7, it made no difference whether Bendix merged with Fram, an established market leader, or with a smaller firm which, by combining with Bendix’s vast resources and managerial skills, would become a new and substantial competitive force in the market confronting Fram and the other established leading firms.
“The examiner’s analysis of the potential competition problem — focusing exclusively on the probability of Bendix’s entry by internal expansion and neglecting the likelihood of entry by merger other than with Fram — was unduly narrow and must be rejected, because it rests upon a misconception of the basic purpose and policy of Section 7. Various forms of merger entry other than through acquisition of a leading company — for example, a ‘toehold’ acquisition of a small company capable of expansion into a substantial competitive force — may be as economically desirable and beneficial to competition as internal expansion into a relevant market, and must be considered in assessing the potential competition of the acquiring firm which has been eliminated as a result of the challenged merger.
“Although previous cases have only involved potential entry in one form, i. e., by internal expansion, it is clear that the
form
of entry was not controlling in these decisions. What was determinative in each of these cases was (1) the actual elimination of the additional decision-making, the added capacity, and the other market stimuli which would have resulted had entry taken a pro-competitive form, such as internal expansion; and (2) the anti-competitive consequences of the removal of the discipling effect of a potential competitor from the market’s edge. We believe that these adverse effects on competition may result from the elimination of a potential entrant who might have entered by internal expansion or who might have entered by a toehold acquisition.” 3 Trade Reg. Rep. 19,288 at 21,439, 21,445 (1970).
On the basis of this toehold theory of illegality, the Commission found that there was a substantial likelihood of a lessening of competition in the automotive filter aftermarket as a result of the acquisition and that the acquisition therefore violated § 7.
In this court Bendix does not seriously question whether the toehold theory is a valid theory of illegality under § 7. It contends that to establish a violation of § 7 based on the toehold theory, Complaint Counsel must prove that the effect of the acquisition “may be substantially to lessen competition” in some line of commerce in some section of the country.
Bendix asserts that Complaint Counsel failed to prove the essential points necessary to establish a toehold violation. In addition to contesting the efficacy of the proof, Bendix raises two probing questions: (1) Is the toehold theory of illegality as applied in the instant case inconsistent with the Commission’s approval of the Examiner’s finding that Bendix was not a likely potential entrant by internal expansion? and (2) Did the Commission violate the Administrative Procedure Act and the Fifth Amendment due process clause in deciding the case on a theory that was never considered by the Examiner or raised by Complaint Counsel and without affording Bendix an opportunity to present proof to rebut this theory?
To dispose of the present review we find it necessary to consider only the issue under the Administrative Procedure Act. For the reasons set forth later in this opinion, we conclude from the record before this Court that the Commission violated § 5 of the Administrative Procedure Act, 5 U.S.C. § 554, when it decided the case on a theory of illegality which was never charged, raised, nor tried during the administrative hearing; never presented for consideration by the Hearing Examiner; and not raised as an issue or discussed by Complaint Counsel in the appeal to the Commission from the order of the Hearing Examiner dismissing the complaint. Bendix had no notice that it was charged under the toehold theory of illegality and was accorded no opportunity to present evidence in defense against this theory. The order of the Commission therefore must be vacated and
the
case remanded.
Although we do not pass upon the merits of the § 7 charge, we recite in some detail certain merit-related factors as disclosed by the record which are relevant to the determinative issue under the Administrative Procedure Act.
Relevant Market
Filters sold in the relevant market are used for replacement of spent filters.
The relevant market is clearly distinguished in sales techniques, distribution channels, product outlets, and customers from the related market of filters for original equipment manufacturers (OEM’s). The OEM market involves sales of the same types of filters to companies such as Chrysler and American Motors for use as parts
on
their new passenger automobiles. General Motors and Ford have divisions which supply all their needs in the OEM market.
The only proofs offered on the relevant geographic market were directed at the entire United States. For purposes of this review we do not consider any geographic submarket less than the entire United States.
The Industry
The predominate force in the relevant market is General Motors. Through its automobile parts division General Motors sells under the trade name “AC” approximately one-third of the filters distributed in the relevant market. Puro-lator is the second largest company in the relevant market, having approximately 21 per cent of the automotive filter aftermarket business. Fram ranks third in the sale of filters in the aftermarket with approximately 17 per cent of the business.
Other significant market members include Wix Corporation, ranking fourth with 9.5 per cent of the business, Walker Manufacturing Company ranking fifth with 3.9 per cent of the business and Hastings Manufacturing Company, ranking sixth with 3.2 per cent of the business.
In addition to the companies previously mentioned, several companies sell filters in the aftermarket on a reduced scale. There are approximately 20 companies with less than one per cent of the business. These companies do not have a volume of business comparable to that of General Motors, Purolator and Fram. That fact does not mean, however, that these smaller companies are not significant market factors. Most of them are making profits, and many have well
trained marketing staffs that vigorously promote their company’s products in the market.
Sales in the aftermarket are made through four principal channels of distribution. First, there is the traditional aftermarket channel in which filter manufacturers sell to warehouse distributors, who in turn sell to jobbers. The latter in turn sell to automotive establishments such as gasoline stations and repair garages which service the cars of their customers. Second, filter manufacturers sell replacement filters to passenger-car manufacturers who in turn sell these filters to their franchised automobile dealers and also to independent warehouse distributors. Third, oil and tire companies constitute another channel of distribution; they purchase filters from the manufacturers and sell them primarily to franchised service stations but also to independent distributors. And fourth, there is still another main group of large-volume purchasers, i. e., the mass-merchandisers (such as Sears Roebuck & Co., automotive chain stores and discount houses); they sell directly to consumers through their own retail outlets.
Mass-marketing and intense promotion as well as very substantial advertising are the hallmarks of successful distribution in the filter aftermarket. The distributional and promotional techniques including purchase bonus stamps, contests and prizes are aimed at the trade rather than the ultimate consumer, since the latter in most instances is not brand-conscious. These techniques are designed to differentiate the promoted and advertised brand from competing brands and so gain and retain widespread brand acceptance by the trade. These efforts are implemented by large sales staffs and great numbers of promotional men, so-called “missionary men,” who are organized on a territorial, regional and national basis. The three leading filter makers, viz., the A.C. Spark Plug Division of General Motors, Purolator and Fram, each employed over 100 missionary men.
Large sales staffs are not limited to the top companies in the industry. Several of the companies having a smaller share of the market employ as many as 100 of the so-called “missionary men.”
These “missionary men” are essential in the aftermarket where entry is not precluded by any substantial technological barriers. There are no basis patents essential to any of the products or processes used in manufacturing operations. Since the set-up cost and cost of production are not prohibitively high, the primary factor for success in the business is promotion. The firms that promote their products and achieve acceptance in the distribution channels are the firms that succeed.
The Parties
A. Bendix. Bendix is one of the largest industrial corporations in the United States. In 1966 its consolidated net sales were $1.05 billion. The company was founded shortly before World War I as a maker of starter drives for automobiles. It became diversified over the years with its manufacturing operations extending to components and assemblies for aerospace, automotive, automation, scientific, oceanic and other uses. In recent years more than one-half of Bendix’s sales have been concentrated in sophisticated equipment for aircraft, missiles and spacecraft. Thus the company was heavily dependent on government sales and particularly government defense contracts. In 1967 the highly sophisticated technical products accounted for 53 per cent of all of Bendix’s sales.
In 1966 Bendix sold $229 million worth of automotive products. That figure includes sales abroad by foreign subsidiaries and affiliates, export sales and sales to the military, nearly all of which were shipped overseas. Automotive parts sold by Bendix in the United States totaled $165 million.
Bendix’s automotive business consists principally of the manufacture of technologically complex parts such as starter drives, ignition components, brakes,
power steering units, carburetors and car radios. About 75 per cent of all its automotive buiness is in the OEM market. The sale of all Bendix automotive products in the aftermarket was about $40 million in 1966. About one-half, or $20 million, was under the Bendix name. Thus, Bendix brand sales in the aftermarket amounted to less than one per cent of the total automotive parts sold in the aftermarket and less than two per cent of the company’s total sales. Although these percentages are small, the dollar volumes cannot be called “insubstantial.” Bendix is certainly no novice in the automotive parts aftermarket. Its principal aftermarket sales division, the Automotive Service Division (ASD), employs about 80 salesmen for all products sold by the ASD. However, it does not sell automotive filters in the aftermarket. At the time of the acquisition of Fram it did not make or sell any filters designed for use on passenger cars, and the ASD had never emphasized promotion and merchandising techniques, nor was it particularly adapted to selling and promoting low technology, consumer-type products, such as automotive filters. The 80 employees of the ASD call primarily on garages and repair shops and they do not customarily call on service stations, the main outlet for automobile filters in the aftermarket.
In regard to automotive filters, Bendix, through its Filter Division, focused upon the production and sale of heavy duty oil and fuel filters for the OEM market for use in large trucks, tractors and off-highway equipment. The Bendix Filter Division was one of the company’s smaller divisions and in 1966 its sales of automotive filters were about $1 million, and of this total about $283,-000 represented sales of filters manufactured for Bendix by a competitor, i. e., Wix Corporation. Most of the filter sales were made in the OEM market, with the lucrative aftermarket accounting for a small amount. The only automotive filter manufactured by Bendix specifically for passenger cars was an air filter element made in the late 1950’s primarily for Ford Motor Co. After losing money on the venture, Bendix in 1960 began subcontracting these filters until 1963, when its agreement with Ford was terminated.
B. Fram. Organized in 1934, Fram is a leading producer of various kinds of filters, including automotive filters, aerospace filters and filter water separators. In 1966, its net sales totaled $66.7 million, its assets amounted to $39.1 million, and net earnings were $4.4 million. Sales, earnings and assets in 1966 were the highest in its history. Fram’s net profit on sales before taxes increased from 10.6 per cent in 1963 to 13.0 per cent in 1966, and return on stockholders’ equity rose from 14.7 per cent to 19.8 per cent during the same period. It was thus a flourishing and growing company.
Prior to its acquisition by Bendix, Fram had six operating divisions, of which the Automotive Division was by far the largest. The latter was engaged in the manufacture and sale of automotive filters, primarily for passenger cars, but also for trucks, and other equipment utilizing internal combustion engines. Net profits on sales by the Automotive Division rose from 14.7 per cent in 1965 to 17.1 per cent in 1966. In that year, automotive filters accounted for 55 per cent of Fram’s total sales.
Although Fram’s Automotive Division sold filters in all distribution channels, about 90 per cent of its filter sales were made in the aftermarket for passenger cars. In that market Fram ranked third, accounting in 1966 for 17.2 per cent of total sales.
The Administrative Proceedings
The dispositive issue on this appeal is whether the Commission denied Bendix and Fram the opportunity to defend the charges with which they were ultimately faced; and, if so, did that denial violate § 5 of the Administrative Procedure Act (APA), 5 U.S.C. § 554.
Prior to the time the complaint in this case was filed, a number of decisions had
been rendered under § 7 of the Clayton Act condemning the elimination of potential competition.
See, e. g.,
United States v. Penn-Olin Chemical Co., 378 U.S. 158, 84 S.Ct. 1710, 12 L.Ed.2d 775; United States v. El Paso Natural Gas Co., 376 U.S. 651, 84 S.Ct. 1044, 12 L.Ed.2d 12. Foremost among these decisions is the Clorox case, FTC v. Procter & Gamble Co., 386 U.S. 568, 87 S.Ct. 1224, 18 L.Ed.2d 303, wherein the Court said:
“[T]he merger would seriously diminish potential competition by eliminating Procter as a potential entrant into the industry. Prior to the merger, the Commission found, Procter was the most likely prospective entrant, and absent the merger would have remained on the periphery, restraining Clorox from exercising its market power. If Procter had actually entered, Clorox’s dominant position would have been eroded and the concentration of the industry reduced.” 386 U.S. at 575, 87 S.Ct. at 1229.
The evils attending elimination of potential competition also have received exhaustive treatment in the periodicals.
See, e. g.,
Turner, Conglomerate Mergers and Section 7 of the Clayton Act, 78 Harv.L.Rev. 1313 (1965). Of the many consequences of conglomerate mergers, Professor Turner listed only one involving potential competition that he considered to be of possible antitrust significance :
“The acquisition may lessen ‘potential’ competition either (a) because the acquiring firm would have entered the industry in any event by internal expansion of its own facilities, so that acquisition reduces the number of competitors by one; or (b) because the acquiring firm, although it might not have entered, was a sufficient threat to existing firms to exert a competitive influence on their price and other decisions.”
Id.
at 1323.
No mention was made by Professor Turner of the effect of the availability of a toehold firm on the acquisition actually consummated. So far as our research has disclosed, the earlier considerations of the effects of § 7 on the elimination of potential competition were concerned with acquisitions where the acquiring firm was a potential entrant by internal expansion into the market of the acquired firm.
The toehold theory was not unheard of, however, at the time of the filing of the complaint. A clear formulation of the toehold concept is found in the Commission opinion in Beatrice Foods Co., (1965-67 Transfer Binder) Trade Reg. Rep. ¶ 17,244 at 22,332 (FTC 1965).
The pertinent portion of the opinion reads as follows:
“The competitive effects are more difficult to predict when a very small factor in the market is acquired by a substantial potential competitor. The merger may increase competition in the market by injecting a substantial firm, one capable of challenging the dominant firms in the market, in place of a firm too small to be a significant competitive factor.
* * * * if *
“The competitive effects are likely to be most serious where the merger is between one of the dominant firms in a concentrated market and a substantial potential competitor. In such a case there is no improvement in the competitive structure of the market— for one dominant firm has simply been replaced by another — and substantial potential competition is eliminated. The dominant firms in the market no longer have to concern themselves with the consequences of entry by the potential competitor; he is already in. Nor need they cope with any additional competition as a result of his entry; he has not increased the number of substantial competitors in the market but simply taken the place of one of those competitors. The potential competitor enters the market in circumstances where there is no change in the competitive structure of the market, except that he is eliminated as a prospective entrant. Such a merger is even more injurious to competition if the acquired firm is a potential com
petitor in one or more of the acquiring firm’s markets. For then potential competition is hurt twice: the merger eliminates the acquiring company as a potential entrant in the acquired firm’s markets, and the acquired firm is eliminated as a potential entrant in the acquiring firm’s markets.”
Complaint Counsel’s brief before the Commission did not mention the toehold theory in discussing the alleged violation. The only section that is pertinent to potential competition was headed “The Hearing Examiner in Concluding that Bendix Was Not a ‘Potential Entrant into the Automotive Filter Replacement Market
by Internal Development’.”
(Emphasis added.)
This is not a case where the initial complaint was couched in broad generalities but subsequently was tried on the specific theory that ultimately justified a finding that § 7 had been violated. In Kroger v. FTC, 438 F.2d 1372 (6th Cir.), petition for cert, filed, 39 U.S.L.W. 3528 (No. 70-202) (May 24, 1971), Mr. Justice Tom C. Clark, speaking for this court on the argument that the case was tried on a different theory before the Commissioner than that pursued before the Hearing Examiner, said:
“Kroger’s own counsel repeatedly objected before the Hearing Examiner to the relevance of certain evidence as to Kroger but not to Beatrice on the ground that ‘independent acts of Party A had nothing to do with the responsibility of Party B.’ That is exactly the theory on which the Commission acted in finding Kroger responsible.”
Id.
at 1378.
In the present case counsel for Bendix specifically asked both the Hearing Examiner and Complaint Counsel for the theory upon which the case would be tried. In response to this questioning Complaint Counsel limited the ease to the theory that the merger eliminated potential competition by virtue of the fact that Bendix was a potential entrant into the filter aftermarket by internal expansion. Throughout the hearing before the Examiner, the case was tried on the issue whether Bendix would have entered the passenger car filter aftermarket by internal expansion if it had not acquired Fram. Repeated statements by counsel, witnesses, and the Examiner showed that every one believed this to be the only issue with respect to potential competition. Complaint Counsel never suggested that Bendix could have and would have entered the automotive filter replacement business by acquiring a toehold firm if Fram had been unavailable. At the conclusion of the hearing, Complaint Counsel submitted elaborate proposed findings and conclusions, reiterating their position that Bendix was a potential entrant through internal expansion. There was no mention of potential entry by a toehold acquisition.
If there would have been any difference in defending the action based on the internal expansion theory as compared to the toehold theory, Bendix was entitled to an opportunity to defend against the theory upon which the Commission acted. It is readily apparent that different defenses and proofs would be used in defending against these two theories. As an example of the difference in proofs, Bendix states in its brief that:
“[An officer of Bendix] was specifically asked whether Bendix would have entered the passenger car filter aftermarket by
internal expansion
if the Fram acquisition had been unavailable. He said no. (A. 1421) He and every other Bendix official who testified would have been asked if Bendix would have made a ‘toehold’ acquisition of a small firm and expanded it, had it been known the Commission would decide the case on that issue.
“Numerous other examples can be given of how the case would have been tried differently,
e. g.,
by further proof that a ‘toehold’ acquisition was not feasible for Bendix; by a more comprehensive analysis of alleged ‘toehold’ candidates to demonstrate in detail why they did not meet the acquisition criteria established by the Bendix management; by the presentation
of further proof that entry into the passenger car filter aftermarket through a 'toehold' acquisition, followed by its expansion, would have entailed the same investments, risks, and activities as entry by internal growth; and by evidence of the economic problems of attempting to expand a ‘toehold’ into a significant competitive factor in the context of this particular industry.”
Section 5 of the Administrative Procedure Act, 5 U.S.C. § 554, provides in relevant part: “(b) Persons entitled to notice of an agency hearing shall be timely informed of * * * (3) the matters of fact and law asserted.”
This court repeatedly has held that an administrative agency must give a clear statement of the theory on which a case will be tried.
See, e. g.,
NLRB v. Tennsco Corp., 339 F.2d 396 (6th Cir.); NLRB v. Johnson, 322 F.2d 216 (6th Cir.), cert, denied, 376 U.S. 951, 84 S.Ct. 968, 11 L.Ed.2d 971. See also S. S. Kresge Co. v. NLRB, 416 F.2d 1225, 1234-1235 (6th Cir.).
In the leading case on this issue involving the FTC, the United States Court of Appeals for the District of Columbia Circuit set aside an FTC order because the Commission sustained a charge on a ground neither alleged in the complaint nor tried before the Hearing Examiner. The Court held:
“[I]t is well settled that an agency may not change theories in midstream without giving respondents reasonable notice of the change. * * * By substituting an issue * * * for the one framed by the pleading, * * the Commission has deprived petitioners of both notice and hearing on the substituted issue. The evil at which the statute [Section 5 of the APA] strikes is not remedied by observing that the outcome would perhaps or even likely have been the same. It is the
opportunity
to present argument under the new theory of violation, which must be supplied.” Rodale Press, Inc. v. FTC, 132 U.S.App.D.C. 317, 407 F.2d 1252, 1256-1257. (Emphasis in original.)
Bendix’s case was prepared and presented in response to certain enumerated theories. The witnesses were questioned and cross-examined in terms of these issues. The documentary proof was keyed to these theories. Bendix was not accorded the opportunity to present proof and argument under the toehold theory of violation.
We hold that the Commission violated § 5 of the Administrative Procedure Act and that the case must be remanded to permit the parties to offer additional evidence. On remand, both Complaint Counsel and Bendix shall be permitted to offer evidence pertinent to the toehold theory of illegality and any related defenses. Both Commission Counsel and Bendix shall be accorded the right to require the presence of any previously sworn witness for additional testimony either on direct or cross examination. Upon rehearing the Commission may consider any evidence previously taken and now on file and such additional evidence as may be admissible and relevant to the issues involved.
Vacated and remanded.