WRIGHT, Circuit Judge:

This is a petition by Northern Natural Gas Company (Northern) and its subsidiary, Northern Natural Gas Transportation Company (Northern Transportation), to review an order of the Federal Power Commission. The challenged order, issued June 20, 1967, authorized the Great Lakes Gas Transmission Company, a Delaware corporation owned jointly by Trans-Canada Pipe Lines Limited and American Natural Gas Company, to construct and operate a 989-mile, 36-inch, natural gas pipeline extending from the Canadian border in northern Minnesota through northern Wisconsin and the Upper Peninsula of Michigan, across the straits of Mackinac, and through the lower Michigan peninsula to the Canadian border at Sarnia, Ontario. By the fifth year of operation, the pipeline is expected to be delivering 734,000 Mcf of gas per day, with approximately 57,000 Mcf per day being purchased by Michigan Consolidated Gas Company and the remainder going to Trans-Canada in eastern Canada.

Simultaneously with the approval of the Great Lakes proposal, the Commission rejected a competitive application by Northern and Northern Transportation which proposed that the Canadian gas imported into northern Minnesota be utilized by Northern in its northern Minnesota markets and that Northern Transportation fulfill the needs of Michigan Consolidated and Trans-Canada by transporting domestic gas, originating in southwestern United States, from Northern's terminal at Ogden, Iowa, through Michigan to the Canadian border at Sarnia, Ontario. To accomplish this exchange-displacement, Northern Transportation would have constructed a 285-mile, 36-inch pipeline from Emerson, Manitoba, to Sandstone, Minnesota, and a 731-mile, 36-inch line from Ogden, Iowa, to Sarnia, Ontario.

Both proposals contemplated exportation of gas from northern Michigan to Sault Sainte Marie (in each case this would entail construction of a short lateral line) and additional importation of 116,000 Mcf per day of Canadian gas by Midwestern Gas Transmission Company and sale of this gas to Michigan Wisconsin Pipe Line Company. Thus the essential difference between the proposals was that Great Lakes would utilize an entirely new pipeline to transport Canadian gas from western Canada to eastern Canada whereas Northern and Northern Transportation would utilize both existing and new facilities to effect an exchange of Canadian gas for United States production.

The principal question raised by this petition is whether the Great Lakes joint venture substantially lessened actual or potential competition and, if so, whether the Commission adequately took account of this factor. Although the Commission is not bound by the dictates of the antitrust laws, it is clear that antitrust concepts are intimately involved in a determination of what action is in the public interest, and therefore the Commission is obliged to weigh antitrust policy. People of State of California v. F.P.C., 369 U.S. 482, 484-485, 82 S.Ct. 901, 8 L.Ed.2d 54 (1962); United States v. Borden Co., 308 U.S. 188, 198-199, 60 S.Ct. 182, 84 L.Ed. 181 (1939); Lynchburg Gas Co. v. F.P.C., 119 U.S.App. D.C. 23, 27, 30-31, 336 F.2d 942, 946, 949-950 (1964); City of Pittsburgh v. F.P.C., 99 U.S.App.D.C. 113, 126, 237 F.2d 741, 754 (1956); Pennsylvania Water Power Co. v. F.P.C., 89 U.S. App.D.C. 235, 240, 193 F.2d 230, 235 (1951), affirmed, 343 U.S. 414, 72 S.Ct. 843, 96 L.Ed. 1042 (1952). This much is conceded by the Commission and the intervenors. We think that implementation of the Commission's order will have serious anticompetitive effects and that, in issuing it, the Commission gave inadequate consideration to the antitrust policy of the United States. We therefore remand the case to the Commission for further consideration.

I

Petitioners have challenged the Commission's handling of the antitrust issues on three basic grounds: (1) the refusal by Trans-Canada either to deliver Canadian gas to petitioners or to purchase domestic gas from petitioners, the refusal by Michigan Consolidated to purchase gas from petitioners, and Midwestern's refusal to participate in petitioners' project constituted an illegal group boycott which contaminated the comparative proceeding conducted by the Commission and prejudiced petitioners' application; (2) the Commission's finding that the potential benefits from competition between Great Lakes and Northern in the taconite region of northern Minnesota outweighed the possible benefits afforded by the entry of Northern Transportation into the lower Great Lakes region was not supported by substantial evidence; and (3) the joint venture resulted in an illegal division of the consumer market between Trans-Canada and American Natural, substantially lessened competition between United States distributors for the supply of Canadian gas, and illegally eliminated competition between independent applicants (Trans-Canada versus American Natural and Midwestern) in a Commission comparative proceeding.

It is the latter ground which is most troubling. The group boycott, though undesirable, did not overtly affect this proceeding because the Commission weighed petitioners' proposal as if the threatened boycott did not exist. As for the relative benefits of Great Lakes competition in the taconite region and Northern Transportation competition in the lower Great Lakes region, we believe there is sufficient evidence to support the finding in favor of competition in the taconite region. Thus the principal question to which we address ourselves is whether there is sufficient evidence to support the Commission's conclusion that "from the standpoint of enhanced competition, we do not think it very significant whether or not the American Natural system participates in the Great Lakes proposal."

A. The Relevance of Antitrust Law to Regulatory Agencies.

Even though the Commission concedes that it must consider the antitrust implications of its action, in order to determine the required extent of that consideration we think it helpful to examine the overall relationship between antitrust law and regulatory agencies. Despite a continuing debate, it appears that the basic goal of direct governmental regulation through administrative bodies and the goal of indirect governmental regulation in the form of antitrust law is the same — to achieve the most efficient allocation of resources possible. For instance, whether a regulatory body is dictating the selling price or that price is determined by a market free from unreasonable restraints of trade, the desired result is to establish a selling price which covers costs plus a reasonable rate of return on capital, thereby avoiding monopoly profits. Another example of their common purpose is that both types of regulation seek to establish an atmosphere which will stimulate innovations for better service at a lower cost. This analysis suggests that the two forms of economic regulation complement each other.

This theory of complementary regulation appears to be borne out by the Supreme Court cases holding that regulated industries must, to some degree at least, accommodate the antitrust laws. F.M.C. v. Aktiebolaget Svenska Amerika Linien, 390 U.S. 238, 88 S.Ct. 1005, 19 L.Ed.2d 1071 (1968) (ocean carriers); United Mine Workers of America v. Pennington, 381 U.S. 657, 85 S.Ct. 1585, 14 L.Ed.2d 626 (1965) (labor union); United States v. El Paso Natural Gas Co., 376 U.S. 651, 84 S.Ct. 1044, 12 L.Ed.2d 12 (1964) (natural gas distributors); United States v. Philadelphia National Bank, 374 U.S. 321, 83 S.Ct. 1715, 10 L. Ed.2d 915 (1963) (banking); Silver v. New York Stock Exchange, 373 U.S. 341, 83 S.Ct. 1246, 10 L.Ed.2d 389 (1963) (stock exchange); United States v. Radio Corporation of America, 358 U.S. 334, 79 S.Ct. 457, 3 L.Ed.2d 354 (1959) (television communication); Georgia v. Pennsylvania R. Co., 324 U.S. 439, 65 S.Ct. 716, 89 L.Ed. 1051 (1945) (railroads); United States v. South-Eastern Under-writers Ass'n, 322 U.S. 533, 64 S.Ct. 1162, 88 L.Ed. 1440 (1944) (insurance); McLean Trucking Co. v. United States, 321 U.S. 67, 64 S.Ct. 370 (1944) (trucking); United States v. Borden Co., 308 U.S. 188, 60 S.Ct. 182 (1939) (agricultural cooperatives). Moreover, the Court has held that even where there are specific statutory exemptions for regulated industries from the antitrust laws, such exemptions are to be very narrowly construed. See e.g., People of State of California v. F.P.C., supra, 369 U.S. at 485-486, 82 S.Ct. 901; Maryland Virginia Milk Producers Ass'n v. United States, 362 U.S. 458, 80 S.Ct. 847, 4 L. Ed.2d 880 (1960).

The complementary regulation theory is also supported by congressional directives requiring certain regulatory agencies to enforce portions of the antitrust laws. The Federal Power Commission, while not included on the list of enforcement agencies, has been instructed to "transmit * * * evidence * * * concerning apparent violations of the Federal antitrust laws to the Attorney General." Congress has also explicitly advised certain agencies to consider basic issues of competition while regulating the industries within their jurisdiction. For other agencies the obligation to act in favor of "public convenience and necessity" has been construed as implying a duty to recognize and weigh traditional antitrust concepts. For example, the Federal Maritime Commission has formulated a rule that acts of shipping conferences interfering with the policies of antitrust laws will be aproved only if the conferences can "`bring forth such facts as would demonstrate that the * * [act] was required by a serious transportation need, necessary to secure important public benefits or in furtherance of a valid regulatory purpose of the Shipping Act.'" F.M.C. v. Aktiebolaget Svenska Amerika Linien, supra, 390 U.S. at 243, 88 S.Ct. at 1009. In approving this standard the Supreme Court noted that "[b]y its very nature an illegal restraint of trade is in some ways `contrary to the public interest.'" Id. at 244, 88 S.Ct. at 1009. And while the Supreme Court did not say that the F.M.C. was obliged to display the degree of deference for antitrust laws suggested by its rule, the Court did conclude that "the antitrust test formulated by the Commission is an appropriate refinement of the statutory `public interest' standard." Id. at 246, 88 S.Ct. at 1010.

This is not to suggest, however, that regulatory agencies have jurisdiction to determine violations of the antitrust laws. See People of State of California v. F.P.C., supra, 369 U.S. at 490, 82 S.Ct. 901; United States v. Radio Corporation of America, supra, 358 U.S. at 350 n. 18, 79 S.Ct. 457; National Broadcasting Co. v. United States, 319 U.S. 190, 223-224, 73 S.Ct. 998 (1943); Mansfield Journal Co. v. F.C.C., 86 U.S.App.D.C. 102, 107, 180 F.2d 28, 33 (1950). Nor are the agencies strictly bound by the dictates of these laws, for they can and do approve actions which violate antitrust policies where other economic, social and political considerations are found to be of overriding importance. In short, the antitrust laws are merely another tool which a regulatory agency employs to a greater or lesser degree to give "understandable content to the broad statutory concept of the `public interest.'" F.M.C. v. Aktiebolaget Svenska Amerika Linien, supra, 390 U.S. at 244, 88 S.Ct. at 1009. But because competitive considerations are an important element of the "public interest," we believe that in a case such as this the Commission was obliged to make findings related to the pertinent antitrust policies, draw conclusions from the findings, and weigh these conclusions along with other important public interest considerations. Johnston Broadcasting Co. v. F.C.C., 85 U.S.App.D.C. 40, 46, 175 F.2d 351, 357 (1949). See also Baltimore Ohio R. Co. v. United States, 386 U.S. 372, 402-403, 436-437, 87 S.Ct. 1100, 18 L.Ed.2d 159 (1967) (Mr. Justice Brennan concurring); Scenic Hudson Preservation Conference v. F.P.C., 2 Cir., 354 F.2d 608 (1965), cert. denied, 384 U.S. 941, 86 S.Ct. 1462, 16 L.Ed.2d 540 (1966).

B. Background of the Joint Venture.

In 1961 Trans-Canada concluded that the growing demand for gas in eastern Canada necessitated additional facilities to supplement the gas then being carried by its Canadian pipeline. Trans-Canada subsequently determined that it would be advantageous to build the new pipeline through the United States because: (1) it would be more economical than "looping" (building a parallel line) its existing line and again traversing the solid granite of the pre-Cambrian shield in Ontario, and (2) a United States line offered the potential of expanding Trans-Canada's export sales. This interest in marketing gas to United States distributors while moving additional gas to eastern Canada prompted Trans-Canada to hold meetings in 1962 with several United States gas distributing companies to determine whether any of them might be interested in a joint project.

As a result of these preliminary discussions, American Natural and Trans-Canada carried on joint studies to determine the feasibility of a joint project. The parties were unable, however, to reach an agreement on the division of ownership and the amount of gas to be purchased by Michigan Wisconsin, the American Natural subsidiary. Consequently, Trans-Canada decided to have Great Lakes, a wholly owned subsidiary incorporated in Delaware, construct and operate a 36-inch pipeline following the route now planned for the joint venture. In accord with Section 7 of the Natural Gas Act, 15 U.S.C. § 717f (1964 ed.), Trans-Canada sought the approval of the Commission.

Shortly thereafter American Natural and Midwestern formulated an alternative to the transportation proposed by Trans-Canada, whereby Midwestern would take gas from Trans-Canada and transport it to Wisconsin, to be utilized there by Michigan Wisconsin in its domestic market, and Michigan Wisconsin would transport domestic gas from its Michigan storage fields to the Canadian border. Before submitting this alternative to the Commission as a competitive application, American Natural and Midwestern proposed to Trans-Canada that it utilize the American Natural — Midwestern alternative. Trans-Canada rejected this proposition because it believed its own proposal was more economical and offered it greater flexibility. American Natural and Midwestern then filed their alternative proposal with the Commission.

Before the Commission had an opportunity to conduct competitive hearings on these mutually exclusive proposals, Trans-Canada and American Natural, subsequently joined by Midwestern, reopened their negotiations for a joint venture. As a result of these negotiations, the three companies withdrew their previous competitive proposals and in their places filed the joint Great Lakes proposal which has been certificated by the Commission. Different reasons were expressed for the withdrawals of the three parties and their concurrence in the Great Lakes joint venture which was to be owned equally by American Natural and Trans-Canada: Mr. E.H. Holstead, a vice president of Trans-Canada and vice president and general manager of Great Lakes, stated that Midwestern withdrew its prior application on the condition that Trans-Canada would sell to it, and Michigan Wisconsin would buy from it, an additional 113,000 Mcf of gas. Mr. Holstead further stated that Trans-Canada would not have withdrawn its original proposal unless Michigan Consolidated agreed to purchase 57,000 Mcf of gas, and, although no direct statements were discovered in the record which revealed the thinking of American Natural, other statements made therein and its oral argument make it clear that it was willing to withdraw its prior application and make additional purchases of Canadian gas under the subsequent joint proposal in order to guarantee itself a portion of the ownership of Great Lakes and a more direct link with the Canadian supply.

C. The Standards for Determining Whether the Joint Venture Was Contrary to Antitrust Policies.

The antitrust law most relevant to these facts is Section 7 of the Clayton Act:

"No corporation engaged in commerce shall acquire * * * the whole or any part of the stock or other share capital * * * of another corporation * * * where in any line of commerce in any section of the country, the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly." 15 U.S.C. § 18 (1964 ed.).

The Supreme Court has emphasized many times that the purpose of Section 7 is "to arrest * * * the tendency to monopoly, before the consumer's alternatives disappeared through merger * *." United States v. Philadelphia National Bank, 374 U.S. 321, 367, 83 S.Ct. 1715, 1744 (1963). (Emphasis by the Court.) In determining the applicability of Section 7 in specific cases, the judiciary has considered various elements. We may quickly dispose of several of these. First, the natural gas industry is obviously a "line of commerce," United States v. El Paso Natural Gas Co., supra, 376 U.S. at 657, 84 S.Ct. at 1744, and the pipelines are in interstate commerce. Second, the relevant geographic market is easily defined by reference to the disputed pipelines. Also, it is no longer questioned that Section 7 applies to the acquisition of newly formed companies, such as this joint venture, as well as to the acquisition of companies already engaged in commerce. United States v. Penn-Olin Co., 378 U.S. 158, 167-168, 84 S.Ct. 1710, 12 L.Ed.2d 775 (1964). Thus the only unanswered question is whether the probable effect of the joint venture is substantially to lessen competition.

The Commission concluded that American Natural's purchase of a 50 per cent interest in Great Lakes would not have a detrimental effect on competition because "there cannot be competition for * * * the transportation of western Canadian gas to eastern Canadian markets." Furthermore, the Commission found "that the Great Lakes line may well serve as a competitive stimulus" in the taconite region, a market previously dominated by Northern. We do not dispute either of these findings. We do believe, however, that the joint venture will have substantial anticompetitive effects on the marketing of gas in the upper Midwest, since the joint venture appears to have effectively prevented competition from arising among natural gas suppliers selling to distributors in Michigan and Wisconsin, and between the supplier of Canadian gas and those suppliers seeking to market domestic gas.

1. Competition Among Suppliers in the Michigan-Wisconsin Market.

In considering the first of these anticompetitive effects, two facts should be recalled: (1) American Natural is the dominant supplier in Michigan and Wisconsin, supplying over 50 per cent of the gas utilized in both states; and (2) as originally proposed the wholly-owned subsidiary of Trans-Canada contemplated substantial sales of gas in the United States in addition to transporting gas to eastern Canada. From this we may conclude that, had the Trans-Canada subsidiary been certified by the Commission, the accessible Michigan-Wisconsin market, now dominated by American Natural, would have been open to competition.

In determining whether the formation of the joint venture precluded this competition and, if it did, whether this competition would have been substantial, two questions must be considered: (1) Was Great Lakes, as a wholly-owned subsidiary of Trans-Canada, a probable entrant to the Michigan-Wisconsin market — that is, was Great Lakes a true potential competitor? (2) Could competition between Great Lakes and American Natural have a substantial effect on the marketing of natural gas in Wisconsin and Michigan?

The Supreme Court has suggested that, in deciding whether one is a true potential competitor, consideration should be given to "the nature or extent of [the] market, the nearness of the absorbed company to it, that company's eagerness to enter the market, its resourcefulness, and so on." United States v. El Paso Natural Gas Co., supra, 376 U.S. at 660, 84 S.Ct. at 1049. See also F.T.C. v. Procter Gamble Co., 386 U.S. 568, 580, 87 S.Ct. 1224, 18 L. Ed.2d 303 (1967). The nature of the natural gas market was accurately described by Mr. Justice Douglas in El Paso:

"This is not a field where merchants are in a continuous daily struggle to hold old customers and to win new ones over from their rivals. In this regulated industry a natural gas company (unless it has excess capacity) must compete for, enter into, and then obtain Commission approval of sale contracts in advance of constructing the pipeline facilities. In the natural gas industry pipelines are very expensive; and to be justified they need long-term contracts for sale of the gas that will travel them. * * * Once the Commission grants authorization to construct facilities or to transport gas in interstate commerce, once the distributing contracts are made, a particular market is withdrawn from competition. The competition then is for the new increments of demand that may emerge with an expanding population and with an expending industrial or household use of gas." 376 U.S. at 659-660, 84 S.Ct. at 1049. (Emphasis by Mr. Justice Douglas.)

In this case the additional demand in the eastern Canadian market made a new pipeline feasible. Because this market is reached most easily by traversing a part of the large Lower Michigan market, now dominated by American Natural, and also the as yet undeveloped Upper Peninsula, there was additional incentive for Trans-Canada to seek the right to construct a pipeline to eastern Canada. In fact, every application made for the right to carry gas to eastern Canada contemplated the construction of a pipeline with a capacity in excess of the amount of gas needed to meet the Canadian demands since each hoped to market the excess gas in Michigan. Thus the nature and extent of the market would have been attractive to the Trans-Canada subsidiary.

The second consideration — the nearness of the absorbed company to the market — also suggests that the Trans-Canada subsidiary was a true potential competitor. Trans-Canada has a market and pipelines on both sides of the Michigan-Wisconsin market and, therefore, a link between the two was reasonable and practicable from a business viewpoint. This is obvious in light of Trans-Canada's conclusion that a pipeline traversing the United States would be a more economical means of transporting gas to eastern Canada than looping its existing line in Canada.

Finally, it appears that Trans-Canada possessed the desire and resources to enter the market. Its intent is clearly demonstrated by its original application. And the mere size of its present operation would suggest that financing would not hinder its entry. Thus all factors indicate that a wholly-owned Trans-Canada subsidiary could have entered the Michigan-Wisconsin market.

If a wholly-owned subsidiary of Trans-Canada would have become an actual competitor of suppliers in Michigan and Wisconsin, we believe that the effect would have been substantial and that the northern Wisconsin and Michigan markets could have expected significant benefits. This is so, in large part, because competition, even in a regulated industry, secures benefits which might otherwise be unattainable. Admittedly the Commission possesses a rate-making power and this power is designed to protect the consumers of natural gas. But it is clear that this power is largely a negative one. Thus the Commission may set a selling rate for a supplier only after it has been demonstrated that the present charge is unjust, unreasonable, unduly discriminatory or preferential, a heavy burden even for specialists as intimately familiar with the natural gas industry as is the Commission. On the other hand, if competition exists, albeit in a limited area, there would be incentives for innovation by the regulated companies themselves and for their coming forward with proposals for better services, lower prices, or both. And once innovations or proposals are forthcoming by a supplier, the Commission could more easily act to universalize these benefits than it could have acted to extract them initially.

Consider the types of natural gas markets which exist. There are some markets which are natural monopolies — that is, where the most efficient allocation of resources results in a single supplier. In such markets, competition is sacrificed to avoid wasteful duplication of services and investment, and hence regulation by the Commission is the only protection a consumer has. In virtually all other natural gas markets, there is a tight oligopoly or partial monopoly. In these markets the fortunes of the few sellers are highly dependent, and therefore there is an incentive for the sellers to arrive at a price which will offer the highest return to all of them. Such a uniform price, whether arrived at by formal agreements or merely through price leadership, is in essence a monopoly price and yields monopoly profits. Where the few sellers strive for and attain a uniform monopoly price the only significant difference between a tight oligopoly market and a monopoly market is that in the former the monopoly profits are shared among several sellers. Thus direct rate regulation is as necessary for the protection of the consumer in these markets as it is in monopoly markets.

The above analysis suggests what might happen in a tight oligopoly market. But in practice it appears that firms selling in such a market do not always seek a uniform monopoly price and, if sought, do not always attain it. Among the reasons why uniform monopoly prices are not sought or attained is that one seller may believe he can maximize his profits by expanding his total sales rather than by taking a maximum profit on each sale. Also, even the mere addition of one seller to an oligopoly market makes the market more complex and less predictable. Therefore, there may be competitive actions and reactions in an oligopoly market.

One instance of such activity in a tight oligopoly market within the natural gas industry was noted by the Supreme Court in United States v. El Paso Natural Gas Co., supra, 376 U.S. at 654-655, 84 S.Ct. 1044. There Pacific Northwest Pipeline Corporation, a potential supplier to the California natural gas market, sought, in an attempt to gain entrance to that market, to attract a major customer from El Paso. Pacific Northwest offered lower prices and an uninterruptible supply to this customer whose El Paso supply was then subject to interruption during peak demands. Although El Paso was able to hold this customer and thereby prevent Pacific Northwest from entering the California market, it was able to do so only by giving the customer a firm supply and by dropping its selling price 25 per cent. It is significant that these benefits were initially the result, not of Commission regulation to which El Paso had always been subjected, but rather of the competition of a single potential entrant to the market. Thereafter, since Section 4(b) of the Natural Gas Act, 15 U.S.C. § 717 c(b) (1964 ed.), prohibits suppliers of natural gas from maintaining preferential and unreasonable rates and Section 5(a), 15 U.S.C. § 717 d(a) (1964 ed.), empowers the Commission to set reasonable rates after it has established that the prior rates were unjust, it is probable that at least a portion of the 25 per cent drop in selling price enjoyed by this single customer who was the subject of the competition was subsequently extended to other customers of El Paso as well. Thus it appears that the competition and direct regulation would complement each other to the benefit of consumers generally.

This example demonstrates the important role competition can play as a complementary force in regulated industries. In the instant case, as the Commission recognized. Trans-Canada's excess gas could have been marketed in the taconite area of northern Minnesota where Northern now enjoys a monopoly position. On the theory set out above, it is likely that, if there were competition in that region, the benefits of this competition would spread throughout the Northern system. Even more probable is that the wholly-owned Trans-Canada subsidiary would seek to sell its excess gas in that portion of northern Michigan which is not now served by any natural gas pipeline. If American Natural's subsidiaries chose not to compete with the Trans-Canada subsidiary for this market, there would not be any additional benefits to consumers. But it is more likely that the American Natural subsidiaries would compete for this market and, once again, any benefits derived from that competition could spread throughout the American Natural system.

In its analysis of the joint venture in the instant case, the Commission ignored these potential benefits of increased competition. Indeed, by permitting American Natural to buy a half interest in the Great Lakes pipeline, the Commission enabled American Natural to protect its Michigan and northern Wisconsin markets from any competition by an independent competitor, even though there were no apparent economies of scale to be gained. It should be noted, moreover, that an acquisition tending to produce similar results was quite recently condemned by the Supreme Court as violative of the antitrust laws. Thus the Court held in United States v. El Paso Natural Gas Co., supra, a case brought by the Justice Department, that El Paso, the supplier of over 50 per cent of the California natural gas market, could not lawfully acquire Pacific Northwest, a pipeline that did not then supply the California market but had the potential for doing so in the future, since such a horizontal merger tended to lessen substantially competition in violation of Section 7 of the Clayton Act. The fact that in the instant case this protection of the market was accomplished through a joint venture in which American Natural owned only a half interest, whereas in El Paso there was a complete merger, is not significant for, as the Supreme Court observed in United States v. Penn-Olin Co., supra, 378 U.S. at 168, 84 S. Ct. at 1715, "[r]ealistically, the parents would not compete with their progeny." According to the corporate agreement creating the joint venture, American Natural is permitted to elect half of the members of the board of directors. Consequently, it is clear that American Natural will have a veto power over at least those matters which the corporate agreement requires the board to approve, such as changes in rate schedules, contracts for the purchase, sale or transportation of gas, construction contracts, and applications submitted to governmental authorities. We believe that this degree of control by American Natural will not only lessen competition between Great Lakes and American Natural, but will actually foreclose it. See United States v. Penn-Olin Co., supra, 378 U.S. at 168, 84 S.Ct. 1710.

It should also be noted that the fact situation in the instant case presents the most clear example of a joint venture lessening competition, since the half-owner of the joint venture, American Natural, was already the dominant force in the market which the joint venture planned to enter. In Penn-Olin the Supreme Court noted that joint ventures should not always be analyzed in the same manner as mergers because a merger eliminates an existing market participant whereas the joint venture might create a new competitive force therein. 378 U.S. at 168, 84 S.Ct. 1710. This distinction was crucial in Penn-Olin because the District Court had found that neither of the owners of the joint venture would have entered the market and that the market in which entry was contemplated was then divided between two other companies. In spite of these considerations, however, the Supreme Court reversed the District Court's finding that the joint venture would not substantially lessen competition and remanded the case for a determination of whether "one would have built `while the other continued to ponder.'" 378 U.S. at 173, 84 S.Ct. at 1718. But before reaching this conclusion, the Court hypothesized a situation wherein a joint venture would enter a market already dominated by one of the participants of the joint venture, which is the situation presented by this case. The Court dismissed this hypothesis as an obvious violation of Section 7:

"* * * The rule of United States v. El Paso Natural Gas Co., 376 U.S. 651 [ 84 S.Ct. 1044, 12 L.Ed.2d 12] (1964), where a corporation sought to protect its market by acquiring a potential competitor, would, of course, apply to a joint venture where the same intent was present in the organization of the new corporation." 378 U.S. at 170, 84 S.Ct. at 1716.

Finally, we note that, although Congress has limited entry into natural gas markets by providing that new pipelines may not be built without prior certification of public convenience and necessity, 15 U.S.C. § 717 f(c) (1964 ed.), it has not evinced an intention to restrict competition among qualified operators or a preference for monopoly service. To the contrary, the language of the Natural Gas Act carefully guards against such inferences. Thus the provision relating to the granting of certification is stated positively:

"* * * [A] certificate shall be issued to any qualified applicant * * if it is found that the applicant is able and willing properly to do the acts and to perform the service proposed and to conform to the [relevant law and regulations] * * * and that the proposed service * * * is or will be required by the present or future public convenience and necessity * * *." Section 7(e), Natural Gas Act, 15 U.S.C. § 717 f(e) (1964 ed.).

And Section 7(g) of the Natural Gas Act provides:

"Nothing contained in this section shall be construed as a limitation upon the power of the Commission to grant certificates of public convenience and necessity for service of an area already being served by another natural-gas company." 15 U.S.C. § 717f(g) (1964 ed.).

In accord with this statutory language, this court approved the Commission's certification of an American Natural subsidiary seeking to invade the territory previously monopolized by Panhandle Eastern Pipe Line Company, and in so doing made the following observation:

"* * * [N]othing in the Natural Gas Act suggests that Congress thought monopoly better than competition or one source of supply better than two, or intended for any reason to give an existing supplier of natural gas for distribution in a particular community the privilege of furnishing an increased supply. * * *" Panhandle Eastern Pipe Line Co. v. F.P.C., 83 U.S.App.D.C. 297, 300, 169 F.2d 881, 884, cert. denied, 335 U.S. 854, 69 S.Ct. 81, 94 L.Ed. 402 (1948).

And in Lynchburg Gas Co. v. F.P.C., supra, 119 U.S.App.D.C. at 30-31, 336 F.2d at 949-950, Judge Washington, joined by Judge Burger, responded to an assertion that a certain rate schedule was in violation of the antitrust laws as follows:

"* * * Investors in the natural gas industry, although granted an opportunity for a `fair return', are by no means guaranteed freedom from risk or competition. Such assurance would, in a case such as this, deprive competitors of the right to compete, inhibit efficient allocation of resources and deny ultimate consumers the lowest prices to which they are entitled."

In sum, Congress, the Supreme Court, and this court have concurred in the belief that competition has a role to play in the natural gas industry. Both courts have recognized specific instances where the goals of direct regulation and the antitrust laws have coalesced. This would seem to be increasingly true as the natural gas markets grow, often demanding new facilities because existing pipelines have reached their ultimate capacity. And when new facilities must be built, the competitive advantages afforded by a new entrant might often be more meaningful than any economies of scale which could be attained by permitting the present monopolist, or dominant market force, to construct the new facilities and fulfill the increased demand. Even limited competition would seem to encourage suppliers of natural gas to become more aggressive in proposing new rates and services, and thereby increase the effectiveness of regulation by the Commission. The joint venture approved in this case had the effect of foreclosing this development and, therefore, tended to lessen competition substantially among suppliers in the Wisconsin and Michigan markets. This effect should have been appraised by the Commission along with other relevant considerations.

2. Competition Between Canadian Source Gas and United States Source Gas.

In addition to the effect on suppliers marketing gas to consumers and distribution companies in Michigan and Wisconsin, the joint venture appears to have eliminated competition between Trans-Canada, the only source of Canadian gas east of the Rocky Mountains, and United States sources of supply. That is, the joint venture agreement gave Trans-Canada an assured outlet for additional sales of 170,000 Mcf of gas per day without having to compete with United States sources. And these sales were agreed to, not because the Canadian gas was cheaper or because these sales were necessary to accomplish the transfer of western Canadian gas to eastern Canada, but rather because American Natural, the parent of the buying subsidiaries, seemed willing to sacrifice the interests of its consumers in order to protect its markets and enhance economically the joint venture in which it owned a half interest.

From the beginning, it was obvious that the major reason Trans-Canada sought a United States partner in its venture to transport gas from western to eastern Canada was to help Trans-Canada market gas in the United States. Thus E.H. Holstead, the general manager of Trans-Canada, testified that one of the principal reasons Trans-Canada preferred a United States pipeline over a second Canadian line was to afford Trans-Canada greater access to United States markets. He further said that Trans-Canada was "particularly" interested in combining with United States companies which were "distributor-oriented" and that Trans-Canada "would not have entered into [the joint venture] agreement [with American Natural] without [the] sales [to Michigan Consolidated]."

American Natural agreed to purchase this gas despite the fact that its system had an excess capacity of 300 to 350 MMcf of gas per day which could be utilized with greater flexibility to transport United States gas to its markets at prices equal to or less than the cost of the Canadian gas delivered at Marshfield, Wisconsin. When Karl E. Schmidt, vice president and chief engineer for American Natural, was asked why American Natural agreed to the purchase when it had this excess capacity, the only answer which was forthcoming was:

"Well * * * as you know * * we are a partner in the Great Lakes Transmission Company and the gas which we purchase and make a market for, as far as Great Lakes is concerned, enhances Great Lakes' economics and that certainly was a consideration given in our incentive in being a purchaser at this time."

Furthermore, American Natural flatly stated that if the joint venture were not certificated it would have no desire to take this Canadian gas from Midwestern. That these sales are not necessary to facilitate the transfer of western Canadian gas to eastern Canada is demonstrated by the fact that, after the exchange-displacement method of transferring ends in 1969, the agreement requires Michigan Wisconsin to continue purchasing, via Midwestern, 113,000 Mcf of Canadian gas per day.

From the above it is clear that Trans-Canada will be able to market an additional 170,000 Mcf of gas per day in the United States without having to meet or beat the competition provided by United States source gas, and that American Natural's agreement to purchase this gas through its subsidiaries was part of the bargaining price it had to pay in order to obtan a half interest in the joint venture. This had an effect not only on those who market United States gas at its source, but ultimately on the consumers in the American Natural system. The staff of the Commission detected this and commented:

"* * * Thus the consumers of Michigan Wisconsin and Michigan Consolidated, to whom service could be rendered more readily and more economically through the combined American Natural System, are the sacrificial pawns by which American Natural received (1) half ownership in Great Lakes and (2) Midwestern's quiet acquiescence in the withdrawal of their joint competitive proposal."

To conclude this discussion of the antitrust issues, we believe that the joint venture substantially lessened competition among suppliers in the Michigan-Wisconsin consumer market and between Trans-Canada and suppliers of gas from United States sources. Unless the Commission finds that other important considerations militate in favor of the joint venture and that these considerations are more beneficial to the public than additional competition, the antitrust policies should be respected and the joint venture set aside.

D. Other Undesirable Effects of the Joint Venture.

The joint venture has two other undesirable effects which should have been weighed by the Commission. The first of these relates to its effects on administrative proceedings in general; the second to the opportunity it provides for future anticompetitive activities.

Petitioners have aptly noted that comparative proceedings before regulatory agencies are "sensitive mechanism[s] for weighing the relative merits of * * * rival * * * projects" and one of the "main competitive arenas" of the natural gas industry since it is there that the sellers challenge one another for the favor of the Commission. This process could easily be distorted if the Commission permitteed potential applicants to get together to decide how a market would be divided before submitting their proposals to the Commission, for then private parties rather than the Commission would be determining what means of meeting a market demand is most closely in accord with the public interest. We cannot permit such an abrogation of administrative responsibility.

The danger of allowing parties to agree among themselves prior to submitting their proposals to the Commission becomes all the more apparent when it is remembered that the Commission's power is largely a negative one; it must rely heavily on private initiative to propose projects to meet consumer needs. Indeed, the judiciary has many times emphasized the importance of maintaining free and vital proposals. Thus the Fourth Circuit, whose views were later endorsed by this court, stated:

"One of the most important duties of a public utility, inherent in its franchise to serve the public, is the duty to take the initiative in proposing reasonable rates and rendering adequate services, taking into account changing conditions; and the utility is not relieved from this duty because its activities are subject to governmental regulation, for a regulatory commission is not clothed with the responsibility or qualified to manage the utility's business. * * *" (Footnote omitted.)

Because independent proposals are so important to the administrative process, the Commission as well as this court must be chary about permitting corporate agreements which limit the nature of the proposals submitted to the Commission. There are few opportunities for consumers of natural gas to choose among the several suppliers offering a variety of services and prices. It is therefore extremely important that a competitive edge be maintained in Commission proceedings. This will increase the chance that the public will be given better service at a lower price. If the Commission determines, after reviewing individual proposals, that a joint project would be more advantageous, it can at that time refuse to certify the individual plans and itself suggest a joint application.

A second undesirable aspect of the joint venture is that it increases the risk of joint action between the parents in future endeavors. This increased risk was recognized in Timken Roller Bearing Co. v. United States, 341 U.S. 593, 600, 71 S.Ct. 971, 976, 95 L.Ed. 1199 (1951), where Mr. Justice Black said, with respect to an international joint venture, that if it were "not severed, the inter-company relationships will provide in the future, as they have in the past, the temptation and means to engage in the prohibited conduct." Commentators considering this aspect of joint ventures have concurred in this analysis, explaining the phenomenon as follows:

"* * * The joint venture puts the parents, particularly if they are competitors, in dangerous proximity to discuss and act jointly on aspects of their business apart from the joint venture and creates an aura of cooperative team spirit which is apt to dampen competitive fires between the firms involved. * * *" (Footnote omitted.)

There is certainly the opportunity for such joint action from the parents in the instant case since here the officers of the joint venture are not only named by the parent companies but also serve as officers of those companies. There are many forms which this joint action may take. One form has already made its appearance: the threatened joint boycott of Northern and Northern Transportation if their proposals had been certificated. A second possibility is reciprocal dealing — that is, American Natural may continue to buy from Trans-Canada only if Trans-Canada promises not to sell to any competitors of American Natural, or only if Trans-Canada promises to fill its short-term demands for United States gas, such as that now being supplied by the Tennessee Gas Transmission Company, by purchasing from American Natural subsidiaries. Another possible form of joint conduct is that Trans-Canada and American Natural could plan cooperative expansion programs which would divide future markets. Thus it seems that numerous undesirable joint anticompetitive actions may spring from this joint venture relationship.

E. Prior Agency Orders of the Federal Power Commission and the Securities and Exchange Commission.

Intervenors, not the Commission, have suggested that the prior Commission order dated August 6, 1965, permitting Trans-Canada and Michigan Wisconsin and Midwestern to withdraw their competitive proposals, and the order of the Securities and Exchange Commission authorizing American Natural to acquire 50 per cent of the stock of the joint venture, should be respected, and that accordingly we are precluded from requiring the Power Commission to weigh the joint venture against antitrust policies. We disagree.

As for the prior order of the Power Commission, it should be recognized that it was merely an "order granting withdrawal of [the competitive] applications and terminating [the] proceeding" for weighing the original Trans-Canada proposal and the Michigan Wisconsin and Midwestern proposal against the public interest. In issuing the order, the Commission clearly did not intend to put a stamp of approval on the joint venture, for, apart from the fact that the parties anticipated submitting a joint proposal, the Commission had no knowledge of the joint venture. And it must be assumed that its order was made in the belief that any future proposal would have to endure a full certification proceeding.

Moreover, the duty imposed upon the Commission by Section 7 of the Natural Gas Act is not merely to determine which of the submitted applications is most in the public interest, but also to give proper consideration to logical alternatives which might serve the public interest better than any of the projects outlined in the applications. Udall v. F.P.C., 387 U.S. 428, 87 S.Ct. 1712, 18 L. Ed.2d 869 (1967); City of Pittsburgh v. F.P.C., supra, 99 U.S.App.D.C. at 123 n. 28, 237 F.2d at 751 n. 28; Scenic Hudson Preservation Conference v. F.P.C., supra, 354 F.2d at 617-620. See generally Reich, The Law of the Planned Society, 75 YALE L.J. 1227, 1248-1251 (1966). Therefore, the Commission was obliged in its examination of the joint venture to give some consideration to the first set of applications, even though they had been withdrawn. Indeed, the Commission, recognizing this duty, pointed out in its brief that consideration was implicitly given to the original Trans-Canada proposal since, except for ownership, the joint venture and the original Trans-Canada proposal are essentially the same. Since the prior Commission order granting withdrawal did not foreclose the Commission from considering the original proposals, we cannot accept the intervenors' contention that the prior order bars our remanding this case to the Commission for further consideration of antitrust policies.

The Securities and Exchange Commission order permitting American Natural to acquire the joint venture stock is more troublesome. The S.E.C. order was sought and obtained because American Natural is a registered public utility holding company and, as such, requires S.E.C. approval before extending its holdings. Although some of the factors which are relevant to the Power Commission's determination of the public interest may have been considered by the S.E.C. and, therefore, the order is perhaps deserving of some weight, there is nothing in either the Public Utility Holding Company Act or the Natural Gas Act which requires one agency to defer to the judgment of the other. In this case even limited comity is not practical because the S.E.C., before concluding that the proposed acquisition was "appropriate in the public interest and in the interest of investors and consumers," merely recounted the content of American Natural's application. American Natural Gas Co., No. 3-349, Holding Company Act Release No. 15422, March 7, 1966. The S.E.C. held no hearings and made no findings to support its conclusion. Therefore, there is little to compel either the Power Commission or this court to give substantial weight to the order of the S.E.C.

II

A second question raised on appeal is whether there is substantial evidence in the record to support the conclusions of the Commission with respect to cost analysis, expansibility of the proposed pipelines, and supply analysis. We find substantial evidence to support all conclusions other than two aspects of the cost analysis.

The Commission's initial opinion stated that in final analysis the costs of the joint venture were less than the costs of petitioners' project:

"* * * [W]e are convinced * * that the alleged rate savings to Trans-Canada and Michigan Consolidated under [the Northern] proposal would be but shortrun."

But in the opinion following the rehearing, the Commission stated:

"* * * We are fully aware of the cost savings intrinsic in an exchange transaction and the benefits that may result to consumers. However, here we find, as set out in our original opinion, that Northern Natural will incur significant costs which in a rate case would appear allocable to Northern Transportation's proposed service to Trans-Canada and Michigan Consolidated. Such an allocation does not deny that savings might arise from the exchange transaction; but it would alter the economics of the Northern proposal by increasing the costs allocable to Trans-Canada and Michigan Consolidated to the point where those costs would exceed the cost of the Great Lakes proposal. * * *"

The Commission's conclusion is unfounded. In the initial opinion the Commission found that Northern Transportation's costs were understated because: (1) it failed to add the additional $14,000,000 investment Northern would have to make in Minnesota in order to accommodate the Canadian gas; and (2) it failed to attribute to Northern Transportation any of the costs of transporting gas from its United States source to Ogden, Iowa, which the Commission alleged was necessary under its rolled-in-costs system of determining rates.

While it may be true that Northern would have to invest $14,000,000 to handle the Canadian gas, Northern also claimed that by utilizing the Canadian gas in Minnesota it would save an additional expenditure of $29,000,000 which was otherwise necessary to meet the Minnesota market demand with gas from United States sources. Since this claimed saving was not disputed by the Commission, it seems that Northern would in fact experience a $15,000,000 saving in investment costs.

As for allocating to Northern Transportation part of the transportation cost incurred by Northern in transporting the gas to Ogden, Iowa, this would not detract from the overall savings of petitioners' project. True, it would effect a rise in the costs of the Northern Transportation project which would be reflected in higher rates for its customers. But such a shift in costs would produce a corresponding decrease in costs allocable to Northern, thereby permitting lower rates for its customers. Since the public interest which the Commission is obliged to consider includes all natural gas resources and consumers, this overall saving should have been treated as a beneficial aspect of petitioners' exchange-displacement project.

The second aspect of the cost analysis which is faulty relates to the Commission's justification of the cost estimates of the joint venture. The staff's minimum adjustments brought the total cost estimates for the joint venture project to $214,000,000 from the $211,000,000 estimated by the joint venture. $5,000,000 was added by the staff to reflect the cost of installing 36-inch pipe. Great Lakes never made a detailed study of the costs of laying 36-inch pipe; rather its study related to 30-inch pipe and the costs so computed were later increased by a factor of 1.135, thereby adding 13.5 per cent to the costs arrived at for 30-inch pipe. The Commission staff, relying on a 1965 report of the Commission, said that "[a]ctual industry experience indicates a differential, between 30- and 36-inch laying costs, ranging from 32 to 39 percent." By using the minimum 32 per cent differential, the staff estimated the installation costs for Great Lakes to be $5,000,000 more than was estimated by the joint venture.

The Commission rejected this adjustment by its staff, saving that "the Great Lakes estimate * * * was not only substantiated by the testimony of Mr. David Williams, Jr., of the Williams Brothers Construction Co., but that company offered to complete the major construction effort within the Great Lakes estimate." Careful review of Mr. Williams' testimony in the joint appendix does not reveal any testimony attempting to justify a 13.5 per cent adjustment for raising cost figures to obtain an approximation of the costs of installing 36-inch pipe. Nor were we able to discover any "offer" by Williams to complete the Great Lakes project within his estimates. The closest he came to making an "offer" was the following conditional statement:

"Q. Does William Brothers intend to submit a bid to Great Lakes should they get certificated? A. Well, it depends upon the prior commitments that we may have. * * * [W]e would be prepared to submit a proposal for the work in 1967 — that is, the remaining eight hundred-some-odd miles — within our estimate, with possibly the contingency that applies strictly to the construction alone. * * * I doubt if we would submit on individual sections."

This statement does not foreclose Williams from declining to bid for the work altogether or, if a bid is made, from adjusting his costs; it is not an offer to complete the job within the previous cost estimates. We do not mean to suggest that the Commission could not have accepted a lower adjustment factor than was indicated by its 1965 study, but only that it was imprudent to do so for the reasons it has advanced.

Thus we believe that the Commission did not have sufficient evidence to accept the Great Lakes cost estimate in its entirety, and that the estimated cost saving of petitioners' proposal has not been adequately explored.

III

The final question raised by petitioners is whether the corporate structure of Great Lakes, which gives each owner a veto power over the principal actions of the corporation, renders Great Lakes incapable of discharging its responsibilities as a regulated natural gas company. We find this contention without merit. As the Commission has noted, in most instances there would be a sufficient community of interest between the parties so that an agreement could be reached on policy issues. And in those few instances in which agreement would not be reached, the arbitration provisions of the corporate agreement seem sufficient to resolve any deadlocks. See Ringling v. Ringling Bros. — Barnum Bailey Com. Shows, 29 Del. Ch. 610, 53 A.2d 441, 447 (1947).

IV

To conclude, we find that the Commission failed to apply proper standards to determine relevant antitrust policy and consequently ignored significant anti-competitive effects of the joint venture. We therefore remand this case for further consideration in accord with this opinion. On reconsideration the Commission should weigh the foreseeable gains from limited competition along with other economic, social and political factors encompassed within the "public interest" concept. Furthermore, unless it is determined that other considerations, such as gas quality, greater expansibility and convenience afforded by a new line running directly from western Canada through the upper Great Lakes region to eastern Canada, outweigh any cost advantages of an exchange-displacement project, then the Commission must also reconsider those elements of its cost analysis which we have found to be unsupported by sufficient evidence.

In remanding this case, we are fully aware of the pressures under which the Commission was working. But, as Mr. Justice Holmes once noted, cases such as this often result in "bad law" because they "exercise a kind of hydraulic pressure which makes what previously was clear seem doubtful, and before which even well settled principles of law will bend." Northern Securities Co. v. United States, 193 U.S. 197, 400-401, 24 S.Ct. 436, 48 L.Ed. 679 (1904) (dissent). The judiciary and administrative agencies must strive to resist such pressures.

Remanded.