We are again faced with a challenge to the regulatory response of an administrative agency to the fuel crisis of late 1973. In this case the petitioner is the United States, acting through the Antitrust Division of the Department of Justice, and the administrative agency under review is the Civil Aeronautics Board.

The United States argues that the Board, faced with a decrease of fuel for commercial aeronautical use, instinctively reacted anticompetitively by permitting three major air carriers to agree among themselves as to the number and scheduling of flights in twenty major markets. The United States wants each carrier to decide on its own which flights to eliminate and when to schedule those flights remaining. The Board rejected unilateral reductions as being inconsistent with the public interest because the Board thought that it would have left lightly traveled markets with little or no service and would have resulted in the 'bunching' of flights at peak demand times.

We hold that, while there were emergency conditions that provided a justification for the Board's summary course in approving capacity reduction agreements in October 1973, its continuance of that approach in its July 1974 order did not have a legally adequate predicate, and therefore must be set aside. The Board's approval of voluntary anticompetitive agreements, which confers an immunity against antitrust liability, must rest on a justification of serious transportation need or important public benefits, with need for a Board showing of an appropriate factual predicate. In this case, the record presents little more than speculation, under emergency conditions, that competitive market response would be wasteful of energy. There was no subsequent procedure, either for a hearing or an experiment in certain markets, no system for testing or opportunity for considered examination of the Board's underlying factual assumptions as to the consequence of the competitive alternative.

The essential facts of this case are rather complex. Faced with a national fuel shortage, on 12 October 1973 the Energy Policy Office, acting pursuant to authority granted by the Economic Stabilization Act Amendments of 1973,1 issued EPO Regulation 1, which established a program of mandatory allocation for middle distillate fuels, which includes jet aviation fuel.2 Under the program air carriers were to receive jet fuel at 1972 levels of consumption, which was expected to be between ninety and ninety-three percent of consumption for the period 1973--74. If fuel stocks were not adequate to acclocate on the basis of 1972 levels (as turned out to be the case), carriers were to be allocated fuel in proportion to those levels.

The same day the Board sua sponte authorized all certificated route and supplemental air carriers to engage in 'discussions to consider adjustment of schedules to the extent necessary to accommodate the fuel allocation program . . ..'3 As a result of these discussions, American Airlines, Inc. (American), United Air Lines, Inc. (United), and Trans World Airlines, Inc. (TWA) reached four agreements to reduce capacity in twenty markets for a six-month period beginning 1 November 1973, the effective date of the fuel allocation program. The agreements were submitted on 23 October 1973 to the Board for approval, along with the request that the approval schedule be accelerated in light of the perceived need to act by 1 November.4 The Board gave interested parties until 29 October to comment on the application. During that period the Departments of Justice and Transportation and Braniff Airways, Inc. (Braniff) filed answers opposing the application, while Delta Air Lines, Inc. (Delta), though opposing capacity agreements in principle, answered that it would not oppose approval under certain conditions.

The Department of Justice's answer contains essentially the same argument it makes now before this court: the carriers had not made any showing that the agreements, anticompetitive on their face, were strictly required in order to secure substantial public benefits. The Department of Transportation made an analogous argument, although it did not oppose continued multilateral discussions. Braniff opposed the agreements out of a fear that reducing consumption in markets subject to the agreements would make more fuel available in non-agreement markets, where competitors such as Braniff would be experiencing fuelshortage problems.

On 31 October 1973 the Board approved the four capacity limitation agreements. It is this order, Order 73--10--110, which is the subject of No. 73--2276.5 Although the agreements were set to terminate on 28 April 1974, United, American, and TWA on 6 February requested an extension of its agreements through 14 June. At that time they also requested the extension of a four-market capacity limitation agreement entered into on 23 May 1973 and approved by the Board on 27 July 1973.6 This four-market agreement was prompted by economic factors unrelated to the fuel crisis and was approved only through 15 March 1974. The Air Line Pilots Association (ALPA) petitioned for review of that order arguing (1) that the CAB should not have granted interim approval to the agreement without holding an evidentiary hearing to determine whether the anticompetitive effects of the agreement outweighed other public interest considerations and (2) that the CAB should have required labor protective arrangements as a condition for interim approval. Both these arguments were recently rejected by this court.7

The Board did not act to approve or disapprove this request for an extension until 24 July, three months after the twenty-market agreement by its own terms expired, and four months after the interim approval on the four-market agreement had expired. In the interim on 5 April the carriers asked approval of six agreements, which, but for the fact that they did not cover one of the twenty-four markets previously subject to the two sets of agreements, were virtually identical to the expired agreements. These agreements were to stay in effect until 14 December 1974.

The Departments of Justice and Transportation again opposed approval on the grounds that the agreements were anticompetitive. In addition, they argued that the Arab oil boycott had been lifted, and as a result the fuel shortage was easing. On 24 July 1974 the Board by a vote of 3--2 approved the 5 April agreements under certain conditions. This order, Order 74--7--105, is the subject of No. 74--1900. It approved all the capacity reduction agreements on the basis of the need to conserve jet aviation fuel. As a result, the extension of the four-market agreement lost its differing character and entered the mainstream of this controversy.8

More recently, the Board has authorized the carriers to submit requests for an extension9 and on 22 November an appropriate agreement was submitted covering the period through 14 June 1975. Again the Board granted interim approval of the agreement10 with the net effect that the agreements authorized by the orders under review are still in effect. Although no one has suggested mootness, we have held in the past that as long as reductions in service authorized by an order under review are still in effect, there remains a live controversy which we are permitted to adjudicate.11 In addition, review of these expired orders is proper because the Board's orders have clearly shown themselves to be 'capable of repetition' and their period of effectiveness is so short they would otherwise evade our review.12

Section 412(a) of the Federal Aviation Act requires that air carriers submit all agreements affecting air transportation entered into with any other air carrier relating, inter alia, to

. . . preserving and improving safety, economy, and efficiency of operation, or for controlling, regulating, preventing, or otherwise eliminating destructive, oppressive, or wasteful competition, or for regulating stops, schedules, and character of service, or for other cooperative working arrangements.13

The agreements previously mentioned were filed as required by section 412(a) and the orders approving these agreements were entered pursuant to section 412(b), which provides:

The Board shall by order disapprove any such contract or agreement, whether or not previously approved by it, that it finds to be adverse to the public interest, or in violation of this chapter, and shall by order approve any such contract or agreement, or any modification or cancellation thereof, that it does not find to be adverse to the public interest, or in violation of this chapter . . ..14

Such approval immunizes the agreements from the operation of the antitrust laws.15

The initial question posed by the parties is the appropriate legal standard to be applied by the Board in determining whether approval of an agreement submitted under section 412(a) having anti-competitive aspects is consonant with the public interest. The Board, relying upon its decision in the Local Cartage Agreement Case,16 argues that the appropriate standard is the one there set out:

Where an agreement has among its significant aspects elements which are plainly repugnant to established antitrust principles, approval should not be granted unless there is a clear showing that the agreement is required by a serious transportations need, or in order to secure important public benefits.17

The United States argues that the Board cannot find an agreement to be in the public interest unless the Board determines that the end sought cannot be achieved at all in a less anticompetitive way. Although these two standards are not necessarily inconsistent (and in fact the United States seems to derive its standard from the Board's standard), the Board's past applications of the Local Cartage standard and it application of the standard in these cases indicate that the Board does not understand Local Cartage to set down as rigorous a standard as that desired by the Department of Justice.

There is no dispute that the Board in considering whether an agreement is in the public interest must consider the agreement's effects on competition. The Federal Aviation Act sets out competition as one of six factors to be considered by the Board in making a public interest determination.18 This court has explicitly stated on prior occasions that the Board must consider the anticompetitive implications of agreements submitted for approval under section 412.19 In addition, the Supreme Court has recognized in analogous circumstances, even absent a specific statutory statement, that an evaluation of the 'public interest' includes a consideration of anticompetitive effects.20

In examining the statute and the previously mentioned cases, we have been unable to find a basis for the rigorous 'impossibility' standard advocated by the United States. In reviewing an order of the Federal Power Commission (FPC) approving the construction and operation of a natural gas pipeline extension, we stated that the FPC was '. . . obliged to make findings related to the pertinent antitrust policies, (and) draw conclusions from the findings . . .'; however, in the end the FPC was only required to '. . . weigh these conclusions along with other important public interest considerations.'21 More directly, this court has considered international air fare agreements approved under section 412. In those cases this court came in contact with the Local Cartage standard in a context in which it was not necessary to determine its substantive validity. However, the court did indicate its view that competition was but one factor to be considered in reaching the public interest

balance. For example, in National Air Carrier Assoc. v. CAB (NACA I),22 the court said:

In short, the essential question, from an antitrust standpoint, is whether the existence of a market structure conducive to maximum feasible competition will be imperiled by approval of the agreement. 'Furtherance of the economic or policy implications of the agency's particular statutory charter may indeed compel overriding of antitrust principles,'¹⁰

10. Cities of Statesville v. AEC, No. 21,706 (D.C.Cir. Dec. 5, 1969) (en banc) (slip op. at 45; Judge Leventhal, concurring).

but first the proper antitrust questions must be asked and answered. Thus it is not really responsive to an antitrust claim, except perhaps as a counterbalance to the projected anticompetitive effects, to say that an agreement will provide the public with lower fares, if there is a risk that these lower fares are merely a step toward an ultimate increase in concentration. Price cutting, after all, is a time-honored tool of the aspiring monopolist.23

In National Air Carriers Assoc. v. CAB (NACA II),24 the Local Cartage standard was again considered without firm resolution:

Petitioners make much of the fact that the Board included in its statement of differences with the examiner its view that Local Cartage is not strictly applicable to each and every IATA-made fare. We are not convinced that the resolution of this difference is critical to our decision, since it is by no means clear to what degree the Local Cartage standard diverges from the statutory 'public interest' requirement of Section 412; and, even by Local Cartage, it would appear that the Board here found that the statutory standard was met by evidence of both 'transportation need' and 'important public benefits.'

What the Board meant, it seems to us, is that Section 412 applies to all manner of carrier agreements, not just those involving international air fares. When one of such other agreements is filed, it may or may not be violative of the antitrust laws.⁹

9. IATA agreements involving group boycotts, as distinct from concerted action on fares, have failed to secure the Board's approval. See IATA Credit Agreements, 30 C.A.B. 1553 (1960); IATA, Extension of Credit, 31 C.A.B. 1041 (1960); IATA Resolution re Visual In-flight Entertainment, C.A.B. Order E--22240 (1965). In each such case the Board had to decide first whether the particular agreement was contrary to antitrust principles, and then to determine whether, even if it was, there were overriding justifications.25

If the Board finds that it is, it must not be approved without a showing of serious transportation need or important public benefits sufficient to overcome the antitrust violation. In the case of IATA fare agreements, however, the Board has already decided that the concerted making of fares by IATA, although clearly in conflict with antitrust principles, is not 'adverse to the public interest' within the meaning of Section 412. Thus, the filing of an IATA fare agreement does not in each case present the Board with the threshold question of whether the agreement is a violation of the antitrust laws. So long as its earlier decision stands, the Board need not go over that ground again but may proceed directly to the question of its conformity to the statutory standard of adversity to the public interest.

Of course, if the Board finds that the particular IATA fare agreement before it is unfairly designed and directed in purpose and effect to the elimination of legitimate competitors, it is obviously out of keeping with the statutory standard. Here, the Board has found that the particular agreements before it are not so constituted. If that finding is supportable, there is no necessity for the Board to reopen, and to examine anew, the question of whether the IATA rate conference machinery is in the public interest.

Similarly, the Supreme Court has approved a formulation essentially identical to the Local Cartage standard in the context of the Federal Maritime Commission's approval of shipping conference agreements.26

Turning to the Federal Aviation Act itself, it should be noted that competition is merely one of six or more factors to be considered by the Board in determining the public interest.27 Conceivably, in any given case, a less anticompetitive means of achieving the same result would be possible, but it might be uneconomical, less safe, or discourage development of an air transportation system properly adapted to America's future needs (to name three other considerations set out in section 102). In that event, the statute would appear to contemplate that the factor of competition could be outweighed by one or more of the other factors. Even if the value of competition is given a preferred status relative to other aspects of the public interest, such a preferred status does not lead one inexorably to the Government's 'impossibility' standard. In fact, the Local Cartage standard itself appears to give competition a preferred status in the public interest balance.28

Underlying this whole discussion is the fact that the entire regulatory scheme set up by the Federal Aviation Act is severely anticompetitive. We have noted on a prior occasion that '(o)f course, most Section 412 agreements will have anti-trust implications.'29 The fact that Congress specifically immunized such agreements, if approved by the Board, from the operations of the antitrust laws is further indication that anticompetitive effects were contemplated.30 The Department of Justice may well be dissatisfied with the balance drawn by Congress between competition and other aspects of the public interest. However, the remedy is not to seek judicial reinterpretation of statutory language that permits the Board, in an appropriate case, to override antitrust considerations, but to seek from Congress amendment of the Board's statutory charter.

III. The Conflicting Models of Competitive Activity

On a first reading of the Board's 31 October order, one might miss the Board's discussion of anticompetitive effects. However, the Board's brief indicates that the following language constitutes its consideration of the issue:

In the Board's view, a method is needed to insure that those cutbacks are made in a manner that will provide the public with the best air service possible given the limited levels of service necessitated by the fuel shortage. We believe that fuel-shortage-caused service reductions made by each of the carriers on a unilateral basis would in many cases not meet that challenge. We have long urged that carriers exercise unilateral capacity restraints. Indeed, we have taken steps making it important for their economic well being for them to do so. Nonetheless, we have from time to time approved capacity reduction agreements where we believed the circumstances so warranted. Moreover, where fuel rationing, as opposed to economic factors, is the basis for service reductions, we are particularly concerned that such reductions might be made in ways that do not conform to the public interest.

Our concern stems from several considerations. First, it appears all too likely that airline economics, in a fuel shortage situation, dictate that carrier managements, if acting unilaterally, will choose to retain relatively high capacity levels in highly competitive markets, while cutting back capacity in thin or monopoly markets beyond what would be warranted by passenger-demand considerations. Secondly, schedule patterns that are entirely appropriate from the viewpoint of individual carrier managements can be plainly inefficient in terms of the needs of the public. Third, if competitive carriers are forced at the same time to make unilateral cutbacks, they may choose to withdraw simultaneously from thin markets in favor of each other--thus leaving these markets totally without service. That needless, uneconomic disruption can be avoided if the carriers can adjust their operations with advance coordination. Fourth, because of the fuel shortage and the consequent limit on the total amount of service that can be performed, decisions affecting the level of service in any one market will necessarily affect service elsewhere in the country. Absent capacity agreements, decisions about how much of the limited amount of the air transportation service available should go to one city, and how much should be taken away from another, would be made by the various carrier managements, acting unilaterally. We believe that in the circumstances of a fuel shortage, the nation's passengers, shippers, communities and Postal Service ought to have an impartial arbitrator in a position to weigh their competing needs regarding relative levels of service. Capacity agreements will permit the Board to perform that role by its power to condition such agreements.31

Although the order states four considerations, they really boil down to only two reasons for favoring multilateral over unilateral reductions. These are, according to the Board's brief:1. That carriers acting unilaterally would tend to retain relatively high capacity in highly competitive markets and reduce service in lightly travelled or monopoly markets below acceptable levels in terms of passenger needs. Moreover, unilateral cutbacks in relatively lightly travelled competitive markets could leave those markets with no service.

2. That unilateral scheduling often results in 'bunching' of schedules at the precise point of peak demand and, flight reductions being necessary, that this would leave over-all passenger needs insufficiently served.32

Essentially the same arguments are made in the 24 July 1974 order.33 Referring to its 31 October order, the Board said:

In Order 73--10--110 we expressed our view that because of the nature of airline economics, unilateral action by airline managements probably would not result in the most appropriate apportionment of service among the nation's markets in the circumstances of a fuel shortage, and that in such circumstances capacity agreements enable the Board to perform the role of impartial arbitrator on matters pertaining to, inter alia, the competing needs of communities for airline service. We adhere to that view, especially with respect to the agreements here before us.34

The United States, on the other hand, argues that the rationale advanced by the Board in support of its order is 'speculation' based 'on a record that consisted of no more than the carriers' application and the critical comments of several other parties . . ..'35 The United States goes on to meet the Board's reasons head-on. The argument that the air carriers will retain high capacity in highly competitive markets while cutting back in 'thin or monopoly markets' is said to be 'irrational as a matter of logic, economics, and common business experience.'36 The Department of Justice's position is bottomed on the assumption that monopoly markets are more profitable than competitive markets and that the CAB can independently require that a minimal level of service be maintained if economic or other pressures push capacity to too low a level.

As to the second of the Board's arguments, the Department of Justice views it as merely a response to demand. Air carriers free to compete will only 'bunch' flights at peak times, if on balance, 'the maximum number of air travelers will be convenienced . . .'37 by such scheduling. As a result, 'bunching' constitutes a response 'to overall passenger needs . . .'38 and is not inconsistent therewith.

IV. Lack of a Full Inquiry as to the Conflicting Models of Competitive Activity

It is quite clear that central to this case is a dispute between the Department of Justice and the Board as to (1) the probable behavior of air carriers if they were required to make unilateral capacity reductions and (2) the effect of such behavior on that most elusive of standards, the public interest. What is disturbing to us is the fact that the Board has not in the past seventeen months undertaken a full-scale, non-interim proceeding in which the substantial contentions of the Department of Justice along with the testimony of experts on such matters as air carrier economics and business practice could have been presented for serious evaluation.

In analogous circumstances we have said that 'it is essential in the face of an antitrust claim that the Board's approval (pursuant to section 412(b)) rest upon a sufficient justification for tolerating the restraint.'39 In Pillai v. CAB we also declared that

(t)o the extent that the Board's 'public interest' determination involved factual predicates, the substantive statute requires that they be supported by substantial evidence. (Footnote omitted.) 'Substantial evidence' is the requirement, no matter what the issue, and no matter how esoteric are the facts to be considered in support of the CAB decision on the issue.40

Our NACA I opinion, while discussing the sufficiency of the Board's findings, also suggested that factual findings were more necessary where an agreement submitted under section 412 was being attacked on antitrust grounds. This was because such grounds are unusually 'complex, requiring painstaking assessment of many relevant considerations.'41 As a result, such 'issues are particularly suited to resolution by evidentiary hearing . . . disposition of antitrust questions by other means creates a greater likelihood of administrative error, and invites a more skeptical judicial scrutiny.'42

The Board's answer to all this was that it was

. . . faced with a full-blown emergency with perfectly obvious signs of serious disruption of airline service and widespread public inconvenience. It recognized that exigencies of time had resulted in its having less than optimum information upon which to base a decision . . .. But because the crisis was 'upon us,' time was of the essence and it was the Board's view that its duty was to take such action as it could to insure that the import of the crisis on the public was no more painful than absolutely necessary.43

In the past this court has been willing to accept emergency agency action even though some of the procedures usually required for full consideration had been dispensed with because of the need for expedition. In this particular case the Board has also argued that the fuel crisis of late 1973 presented it with a type of emergency it had not previously experienced and that therefore we should be somewhat less searching in our judicial scrutiny. On the other hand, in Pillai and in NACA I this court held that 'the specter of an open rate situation does not absolve the Board of its responsibility to employ procedures which will allow it . . . (to make an informed assessment of the public interest under section 412).'44 By analogy it might be argued that the specter of unilateral capacity reductions ought not to excuse the Board from gathering an adequate factual basis before it may approve admittedly anticompetitive agreements.

On balance we decide that the Board's initial emergency order can be upheld in light of the Board's reasonable belief that emergency circumstances required an expeditious response. In addition, because of the Board's general unfamiliarity with fuel shortages, we are willing to accept a less detailed statement of reasons and findings than we would otherwise accept even under emergency conditions. However, by July 1974 there had been time available for a detailed inquiry into the arguments presented by the Department of Justice to, but ignored by, the Board. Far from acting with 'deliberate speed, majestic instancy,' the Board has not acted at all; we therefore reverse the Board's later order.

The Board may argue, as it did in Pillai, that no factual inquiry is required of the Board at all in this case since it is '. . . a case in which the decision subject to review was the sort of policy-making judgment-call which is not to be tested by 'substantial evidence' criteria in the first place.'45 Whatever may be the merits of such an argument in other contexts (it was rejected in Pillai) it is clearly inappropriate in this case. As we said in NACA I, questions as to probable competitive behavior and its probable effects are susceptible to determination in the light of business and economist testimony and exhibits, and cross-examinations concerning analyses and underlying assumptions. Such matters are not to be determined exclusively by reasoning in the abstract, based upon logical speculation.

So much was virtually conceded by the Board when it established a full-scale evidentiary inquiry into the earlier economic capacity reduction agreements. If that matter was susceptible to evidentiary inquiry, so certainly must this matter.

The latter inquiry might be distinguished on the ground that it dealt with 'final' approval of capacity reduction agreements, while the orders under review deal with 'interim' approvals. This court has indicated that where Board actions 'obviously contemplate future administrative modification and supplementation, its authority is unusually expansive.'46 However, when we have upheld emergency or interim action, we have also required that such action must be promptly and fully reexamined.47 No such prompt and full reexamination has taken place here in the over seventeen months since these agreements have been given interim approval. Although we uphold the initial October order on the ground that it was only an interim or emergency action, at some point in time the Board is required to reconsider the entire matter on a non-emergency basis. Certainly by the following July such a point had been reached.

The only remaining contention is one made by intervenor Northwest Airlines that the agreements in this case constituted an acquisition of control by each of the signatories of the other signatories and as a result required approval under section 408(a).48 Although the statute indicates that the acquisition of control can be 'in any manner whatsoever,' 'direct or indirect,' we are satisfied that the agreements here at issue were too limited in scope to come within even a liberal definition of 'acquisition of control.'

We recognize that the Board was faced with a difficult situation in the airline industry in October 1973, and that it moved with commendable speed to act in what it deemed the public interest. In that situation the one certain fact which the Board confronted was that airline capacity (number of flights) would inevitably be reduced because of the drastic cutback in fuel supply, whether by Board or individual carrier action. Everything else appears to be utterly speculative, based on conjecture and surmise. The airlines offered no evidence, not even statements of their plans to reduce capacity, if left to their own choices; the Board asked for none.

With competitive and noncompetitive solutions available, and no evidence to support the choice of either, the Board reacted quickly (the United States argues, instinctively); it chose the noncompetitive solution. Instead of waiting to see what unilateral action the supposedly competing airlines would take, and monitoring the results of that, the Board chose to approve noncompetitive agreements reducing service to twenty markets.

The Board's original action was not a principled choice of alternatives based on evidence, even evidence of individual airline plans to curtail service; its sole justification was and must be emergency, which may then have sufficed. That justification had expired long before the Board issued its July 1974 order. That order is set aside.

So ordered.