delivered the opinion of the Court.

The petitioner, a German manufacturer of automobiles, is one of the largest users of the ports on the West Coast of the United States, delivering through them more than 40,000 vehicles each year, the majority transported there by vessels chartered by the petitioner rather than by common carrier. This case grows out of the petitioner’s claim that charges imposed upon the unloading of its automobiles at Pacific Coast ports are in violation of the Shipping Act, 1916, as amended. 39 Stat. 728, 46 U. S. C. § 801 et seq. The dispute has a long and somewhat complicated history.

The Pacific Maritime Association (the Association) is an employer organization of some 120 principal common carriers by water, stevedoring contractors, and marine terminal operators, representing the Pacific Coast shipping industry. The primary function of the Association is to negotiate and administer collective bargaining contracts with unions representing its members’ employees, of which the International Longshoremen’s and Ware-housemen’s Union (ILWU) is one. In late 1960 the *264Association and ILWU reached a milestone agreement which, it was hoped, would end a long and troubled history of labor discord on the West Coast waterfront.1 The ILWU agreed to the introduction of labor-saving devices and the elimination of certain restrictive work practices. In return, the Association agreed to create over the period from 1961 to 1966 a “Mechanization and Modernization Fund” of $29,000,000 (the Mech Fund) to be used to mitigate the impact upon employees of technological unemployment.2 The agreement specifically reserved to the Association alone the right to determine how to raise the Mech Fund from its members, at the rate of some $5,000,000 a year.

A committee of the Association investigated various possible formulas for collecting the Fund from the steve-doring contractors and terminal operators — i. e., those Association members who were employers of workers represented by the ILWU. A majority of the committee recommended that the Mech Fund assessment be based solely on tonnage handled, and this recommendation was adopted by the Association membership.3 Under this *265formula, general cargo was assessed at 27per “revenue ton.” 4 A revenue ton is based either on weight (2,000 lbs.=one ton) or measurement (40 cu. ft.=one ton). Whether tonnage declarations on a particular item of cargo were to be by weight or by measurement was to depend, with one exception, upon how that cargo had customarily been manifested (and reported to the Association for dues purposes) in 1959. The one exception was automobiles, for which there had been no uniform manifesting custom.5 The Association decided that automobiles were to be declared by measurement for Mech Fund purposes, regardless of how they were or had been manifested.

Unlike shippers by common carrier, the petitioner must arrange and pay for the unloading of its own chartered vessels upon their arrival in port. For this purpose it has since 1954 contracted with Marine Terminals Corporation and Marine Terminals Corporation of Los Angeles (Terminals), which are members of the Association, for the performance of stevedoring and related services in unloading vehicles from the petitioner’s chartered ships in West Coast ports, at a negotiated price. Prior to the Mech Fund assessment agreement, Terminals’ charge to the petitioner for these unloading services was $10.45 per vehicle, of which about a dollar represented Terminals’ profit. When the vehicles were assessed for the Mech Fund by measurement, the assessment came to $2.35 per vehicle — representing, if passed on to the peti*266tioner, an increase in unloading costs of 22.5%.6 If the vehicles had been assessed by weight (0.9 tons) rather than by measurement (8.7 tons),7 the assessment would have been 250 per vehicle — an increase of about 2.4%, comparable to the average Mech Fund assessment of 2.2% for all other general cargo. Assessment by measurement rather than by weight thus resulted in an assessment rate for the petitioner’s automobiles of 10 times that for other West Coast cargo — although automobiles had less to gain than other cargo from the Mech Fund agreement.8 The petitioner and Terminals both protested these seeming inequities to a committee of the Association set up to handle such claims, but without success.9

The petitioner refused to pay any additional charge resulting from the Association’s levy, and Terminals, while continuing to unload Volkswagen automobiles for the petitioner, did not pay its resulting assessment to the Association. The Association sued Terminals in a federal court in California for its failure to pay the Mech Fund assessments; Terminals admitted all the allegations of *267the complaint and impleaded the petitioner as a defendant. The petitioner then obtained a stay of that action to permit it to invoke the primary jurisdiction of the Federal Maritime Commission, in order to determine the following issues:

“1. Whether the assessments claimed from [the petitioner] are being claimed pursuant to an agreement or understanding which is required to be filed with and approved by the Federal Maritime Commission under Section 15 of the Shipping Act, 1916, as amended, 46 U. S. C. 814 (1961), before it is lawful to take any action thereunder, which agreement has not been so filed and approved.

“2. Whether the assessments claimed from [the petitioner] result in subjecting the automobile cargoes of [the petitioner] to undue or unreasonable prejudice or disadvantage in violation of Section 16 of the Shipping Act, 1916, as amended, 46 U. S. C. 815 (1961).

“3. Whether the assessments claimed from [the petitioner] constitute an unjust and unreasonable practice in violation of Section 17 of the Shipping Act, 1916, as amended, 46 U. S. C. 816 (1961).”

The petitioner then began the present proceedings by filing a complaint with the Commission raising the above issues. The petitioner alleged that the Association was dominated by common carriers10 which had agreed upon the assessment formula in order to shift a disproportionate share of the Mech Fund assessment onto the peti*268tioner,, which did not patronize those common carriers.11 The Commission, after a hearing, upheld the initial decision of its examiner and dismissed the complaint, with two dissents.12 The Court of Appeals for the District of Columbia Circuit affirmed,13 and we granted certiorari to consider important questions under the Shipping Act.14

I.

The petitioner’s primary contention — supported by the United States, a party-respondent — is that implementation of the Association’s formula for levying the Mech Fund assessments was unenforceable, because the agreement among Association members imposing that formula was not filed with the Commission in accord with § 15 of the Act. That section provides that there be filed with the Commission “every agreement” among persons subject to the Act

“fixing or regulating transportation rates or fares; giving or receiving special rates, accommodations, or other special privileges or advantages; controlling, regulating, preventing, or destroying competition; *269pooling or apportioning earnings, losses, or traffic; allotting ports or restricting or otherwise regulating the number and character of sailings between ports; limiting or regulating in any way the volume or character of freight or passenger traffic to be carried; or in any manner providing for an exclusive, preferential, or cooperative working arrangement. . . 15

Until submitted to and approved by the Commission, “it shall be unlawful to carry out in whole or in part, directly or indirectly, any such agreement . . . 16 The

Commission is directed to disapprove any agreement *270be unlawful, and agreements, modifications, and cancellations shall be lawful only when and as long as approved by the Commission; before approval or after disapproval it shall be unlawful to carry out in whole or in part, directly or indirectly, any such agreement, modification, or cancellation; except that tariff rates, fares, and charges, and classifications, rules, and regulations explanatory thereof (including changes in special rates and charges covered by section 813a of this title which do not involve a change in the spread between such rates and charges and the rates and charges applicable to non-contract shippers) agreed upon by approved conferences, and changes and amendments thereto, if otherwise in accordance with law, shall be permitted to take effect without prior approval upon compliance with the publication and filing requirements of section 817 (b) of this title and with the provisions of any regulations the Commission may adopt.” 46 U. S. C. § 814.

*269“that it finds to be unjustly discriminatory or unfair as between carriers, shippers, exporters, importers, *270or ports, or between exporters from the United States and their foreign competitors, or to operate to the detriment of the commerce of the United States, or to be contrary to the public interest, or to be in violation of [the Act]. . . .”17

17 “The Commission shall by order, after notice and hearing, disapprove, cancel or modify any agreement, or any modification or cancellation thereof, whether or not previously approved by it, that it finds to be unjustly discriminatory or unfair as between carriers, shippers, exporters, importers, or ports, or between exporters from the United States and their foreign competitors, or to operate to the detriment of the commerce of the United States, or to be contrary to the public interest, or to be in violation of this chapter, and shall approve all other agreements, modifications, or cancellations. No such agreement shall be approved, nor shall continued approval be permitted for any agreement (1) between carriers not members of the same conference or conferences of carriers serving different trades that would otherwise be naturally competitive, unless in the case of agreements between carriers, each carrier, or in the case of agreement between conferences, each conference, retains the right of independent action, or (2) in respect to any conference agreement, which fails to provide reasonable and equal terms and conditions for admission and readmission to conference membership of other qualified carriers in the trade, or fails to provide that any

*271An agreement filed with and approved by the Commission is immunized from challenge under the antitrust laws.18

The Commission held that, although the Mech Fund assessment formula was a “cooperative working agreement” clearly within the plain language of § 15, it nonetheless was not the kind of agreement required to be filed with the Commission under that section:

“Although the literal language of section 15 is broad enough to encompass any ‘cooperative working arrangement’ entered into by persons subject to the Act, the legislative history is clear that the statute was intended by Congress to apply only to those agreements involving practices which affect that competition which in the absence of the agreement would exist between the parties when dealing with the shipping or traveling public or their representatives.

“It is not contested that the membership of [the Association] entered into an agreement as to the manner of assessing its own membership for the collection of the ‘Mech’ fund. Such an agreement, however, does not fall within the confines of section 15 as, standing by itself, it has no impact upon *272outsiders. What must be demonstrated before a section 15 agreement may be said to exist is that there was an additional agreement by the [Association] membership to pass on all or a portion of its assessments to the carriers and shippers served by the terminal operators.” 9 F. M. C., at 82-83.

The Court of Appeals affirmed. That court felt itself confined by our decision in Consolo v. FMC, 383 U. S. 607, to determining simply whether the Commission’s ruling was supported by “substantial evidence.” With “due deference to the expertise of the Commission,” it concluded “(albeit with some hesitation) that there is substantial evidence in the record considered as a whole to support the Commission’s decision.” 125 U. S. App. D. C., at 290, 371 F. 2d, at 755.

The issue in this case, however, relates not to the sufficiency of evidence but to the construction of a statute. The construction put on a statute by the agency charged with administering it is entitled to deference by the courts, and ordinarily that construction will be affirmed if it has a “reasonable basis in law.” NLRB v. Hearst Publications, 322 U. S. 111, 131; Unemployment Commission v. Aragon, 329 U. S. 143, 153-154. But the courts are the final authorities on issues of statutory construction, FTC v. Colgate-Palmolive Co., 380 U. S. 374, 385, and “are not obliged to stand aside and rubber-stamp their affirmance of administrative decisions that they deem inconsistent with a statutory mandate or that frustrate the congressional policy underlying a statute.” NLRB v. Brown, 380 U. S. 278, 291. “The deference owed to an expert tribunal cannot be allowed to slip into a judicial inertia . . . .” American Ship Building Co. v. NLRB, 380 U. S. 300, 318. Cf. FMB v. Isbrandtsen Co., 356 U. S. 481, 499-500 (where this Court overturned the Commission’s construction of § 14 of the Shipping Act).

*273In limiting § 15 to agreements which “affect competition” and in finding that the assessment agreement did not so “affect competition,” the Commission in this case used that phrase in a highly artificial sense — by requiring “an additional agreement by the [Association] membership to pass on all or a portion of its assessments . . . There is no question that the assessment agreement necessarily affected the cost structures of, and the charges levied by, individual Association members. Most, though not all, of the stevedoring contractors and terminal operators did pass the assessment on. The economic realities were such that many of them had no choice — a fact of which they apprised the Association at the time the assessment arrangement was being devised.19 In the case of Terminals, the assessment it had to pay on Volkswagen automobiles was more than twice its profit margin.

The Commission thus took an extremely narrow view of a statute that uses expansive language. In support of that view, the Commission argued in this Court that a narrow construction of § 15 should be adopted in order to minimize the number of agreements that may receive antitrust exemption. However, antitrust exemption results, not when an agreement is submitted for filing, but only when the agreement is actually approved; and in deciding whether to approve an agreement, the Commission is required under § 15 to consider antitrust *274implications.20 FMC v. Aktiebolaget Svenska Amerika Linien, ante, p. 238; see also Isbrandtsen Co. v. United States, 93 U. S. App. D. C. 293, 211 F. 2d 51.21

The Commission itself has not heretofore limited § 15 to horizontal agreements among competitors, but has applied it to other types of agreements coming within its literal terms. See, e. g., Agreements Nos. 8225 and 8225-1, Between Greater Baton Rouge Port Commission and Cargill, Inc., 5 F. M. B. 648 (1959), affirmed, 287 F. 2d 86, and Agreement No. T—4: Terminal Lease Agreement at Long Beach, California, 8 F. M. C. 521 (1965), applying § 15 to lease agreements.22 In the latter case, decided only four months before its decision in the case before us, the Commission said:

“Section 15 describes in unambiguous language those agreements that must be filed; it does not speak of agreements per se violative of the Sherman Act. *275Since the wording of section 15 is clear, we need not refer to the legislative history; there is simply no ambiguity to resolve.” 8 F. M. C., at 531.

To limit § 15 to agreements that “affect competition,” as the Commission used that phrase in the present case, simply does not square with the structure of the statute.23

The legislative history offers no support for a different view. The genesis of the Shipping Act was the “Alexander Report” in 1914.24 FMB v. Isbrandtsen Co., 356 U. S. 481, 490. While it is true that the attention of that congressional committee was focused primarily upon the practices that had cartelized much of the maritime industry, it is clear that the concerns of its inquiry were far more broadly ranging. The report summed up the testimony before the committee:

“Nearly all the steamship line representatives . . . expressed themselves as not opposed to government supervision . .. and approval of all agreements or arrangements which steamship lines may have entered into with other steamship lines, with shippers, or with other carriers and transportation agencies. On the other hand, the shippers who appeared as witnesses . . . were in the great majority of instances favorable to a comprehensive system of government supervision . . . [and] the approval of contracts, agreements, and arrangements, and the general supervision of all conditions of water transportation which vitally affect the interests of shippers.” Alexander Report, at 418. (Emphasis added.)

*276The committee recommended, among other things:

“That all carriers engaged in the foreign trade of the United States, parties to any agreements, understandings, or conference arrangements hereinafter referred to, be required to file for approval ... a copy of all written agreements (or a complete memorandum if the understanding or agreement is oral) entered into (1) with any other steamship companies, firms, or lines engaged directly or indirectly in the American trade, or (2) with American shippers, railroads or other transportation agencies.” Alexander Report, at 419-420.

Nothing in the legislative history suggests that Congress, in enacting § 15 of the Act, meant to do less than follow this recommendation of the Alexander Report and subject to the scrutiny of a specialized government agency the myriad of restrictive agreements in the maritime industry.25

This is not to say that the Commission is without power to determine, after appropriate administrative proceedings, that some types or classes of agreements coming within the literal provisions of § 15 are of such a de minimis or routine character as not to require formal fifing. Since the Commission’s decision in the present case, Congress has explicitly given it such authority:

“The Federal Maritime Commission, upon application or on its own motion, may by order or rule exempt for the future any class of agreements between persons subject to this chapter or any specified activity of such persons from any requirement of this chapter, or Intercoastal Shipping Act, 1933, *277where it finds that such exemption will not substantially impair effective regulation by the Federal Maritime Commission, be unjustly discriminatory, or be detrimental to commerce.

“The Commission may attach conditions to any such exemptions and may, by order, revoke any such exemption.” 26 46 U. S. C. § 833a (1964 ed., Supp. II).

But the agreement with which we deal here — levying $29,000,000 over five years, binding all principal carriers, stevedoring contractors, and terminal operators on the Pacific Coast, and necessarily resulting in substantially increased stevedoring and terminal charges — was neither de minimis nor routine. We hold that this agreement was required to be filed under § 15 of the Act.

*278It is to be emphasized that the only agreement involved in this case is the one among members of the Association allocating the impact of the Mech Fund levy. We are not concerned here with the agreement creating the Association or with the collective bargaining agreement between the Association and the ILWTJ. No claim has been made in this case that either of those agreements was subject to the filing requirements of § 15. Those agreements, reflecting the national labor policy of free collective bargaining by representatives of the parties’ own unfettered choice, fall in an area of concern to the National Labor Relations Board, and nothing we have said in this opinion is to be understood as questioning their continuing validity. But in negotiating with the ILWU, the Association insisted that its members were to have the exclusive right to determine how the Mech Fund was to be assessed, and a clause to that effect was included in the collective bargaining agreement. That assessment arrangement, affecting only relationships among Association members and their customers, is all that is before us in this case. Moreover, so far as the record shows, only the assessment on automobiles is now challenged, and there is no reason to suppose that the Commission will not consider expeditious approval of so much of the agreement as is not in dispute.

II.

The petitioner also attacked the Association’s assessment of its automobiles under § 16 and § 17 of the Shipping Act. Section 16 makes it unlawful “to subject any particular person, locality, or description of trafile to any undue or unreasonable prejudice or disadvantage,” 27 and *279§ 17 forbids any “unjust or unreasonable” regulation or practice “relating to or connected with the receiving, handling, storing, or delivering of property.”28 The Commission ruled that neither of these sections had been violated, and the Court of Appeals affirmed.

If the agreement is now filed under § 15, the Commission will be called upon again to consider the effect of §§16 and 17, since an agreement that violates a specific provision of the Act must be disapproved.29 Accordingly, it is not inappropriate, without now passing upon the ultimate merits of the §§16 and 17 issues, to give brief consideration to the Commission’s handling of those issues upon the present record.

The Commission ruled that the petitioner had failed to demonstrate any “undue or unreasonable prejudice or disadvantage” under § 16 solely because it had not shown any unequal treatment as between its automobiles and other automobiles or cargo competitive with automobiles. In so ruling, the Commission applied the “competitive relationship” doctrine which it has developed in cases concerning rates for carriage of goods by sea.30 *280But the Commission, in cases not involving freight rates and the particularized economics that result from a vessel’s finite cargo capacity,31 has often found § 16 violations even in the absence of a “competitive relationship.” See, e. g., Practices, etc., of San Francisco Bay Area Terminals, 2 U. S. M. C. 588 (1941) and 709 (1944), and Storage Practices at Longview, Washington, 6 F. M. B. 178 (1960), involving storage charges; and New York Foreign Freight Forwarders and Brokers Assn. v. FMC, 337 F. 2d 289, involving freight forwarders’ fees. In a proceeding subsequent to its decision in the present case, the Commission explicitly dispensed with the competitive relationship requirement with respect to port “free time.” Investigation of Free Time Practices — Port of San Diego, 9 F. M. C. 525 (1966); cf. California v. United States, 320 U. S. 577. See also Investigation on Household Goods, North Atlantic Mediterranean Freight Conference, F. M. C. Docket No. 66-49 (June 30,1967). When the agreement in the present case is filed, the Commission may consider anew whether the mere absence of a competitive relationship should foreclose further § 16 inquiry.32

With respect to § 17, the Commission found that the assessment upon the petitioner’s automobiles was not *281“unreasonable,” because the petitioner had received “substantial benefits” in return for the assessment, and there was no showing of a deliberate intent to impose an unfair burden upon the petitioner. This, we think, reflects far too narrow a view of § 17. It may be that a relatively small charge imposed uniformly for the benefit of an entire group can be reasonable under § 17, even though not all members of the group receive equal benefits. See Evans Cooperage Co. v. Board of Commissioners of the Port of New Orleans, 6 F. M. B. 415.33 But here a relatively large charge was unequally imposed. The benefits received by the petitioner may have been substantial, but other cargo received greater benefits at one-tenth the cost.34 Moreover, the question of reasonableness under § 17 does not depend upon unlawful or discriminatory intent. As the Commission itself has said:

“[Sections 16 and 17] proscribe and make unlawful certain conduct, without regard to intent. The offense is committed by the mere doing of the act, and the question of intent is not involved.” Hellenic Lines Ltd. — Violation of Sections 16 (First) and 17, 7 F. M. C. 673, 675-676 (1964).

*282Cf. United States v. Illinois Central R. Co., 263 U. S. 515, 523-526; ICC v. Chicago G. W. R. Co., 209 U. S. 108.

The question under § 17 is not whether the petitioner has received some substantial benefit as the result of the Mech Fund assessment, but whether the correlation of that benefit to the charges imposed is reasonable. The “substantial benefits” measure of unreasonableness used by the Commission in this case is far too blunt an instrument. Nothing in the language or history of the statute supports so tortured a construction of - the phrase “just and reasonable.” The Commission has cited no similar construction of the phrase by any other regulatory agency or court. Indeed, in past decisions the Commission itself has not applied any such test. See California Stevedore & Ballast Co. v. Stockton Elevators, Inc., 8 F. M. B. 97 (1964), and Practices, etc., of San Francisco Bay Area Terminals, 2 U. S. M. C. 588 (1941), affirmed, 320 U. S. 577, where the Commission found violations of § 17 even though the benefits received were clearly substantial. The proper inquiry under § 17 is, in a word, whether the charge levied is reasonably related to the service rendered.

The judgment of the Court of Appeals is reversed and the case is remanded for further proceedings consistent with this opinion. 1

r. . , , It is so ordered.

concurring.

Although I agree with the conclusions reached by the Court in this case, I deem it desirable to amplify the reasons, as I see them, for what is decided today. More especially, I think that further justification is needed for the Court’s decision (1) that the “assessment agreement” falls within the Commission’s jurisdiction under *283§15 notwithstanding its intimate connection with the underlying collective bargaining agreement; and (2) that the Commission should give further consideration to the §§16 and 17 issues notwithstanding that it has already determined them.

I.

The Pacific Maritime Association is a multi-employer collective bargaining group. Its “assessment agreement” directly in question here is closely related to a collective bargaining agreement covering a subject about which employers are required to bargain, “terms and conditions of employment.”1 This underlying labor agreement was, according to apparently unanimous industry and expert opinion, a highly desirable step forward in the shipping industry.

Multi-employer collective bargaining units have long been recognized as among the unit classifications that the National Labor Relations Board may deem “appropriate.” In Labor Board v. Truck Drivers Union, 353 U. S. 87, we held that Congress intended

“that the Board should continue its established administrative practice of certifying multi-employer units, and intended to leave to the Board’s specialized judgment the inevitable questions concerning multi-employer bargaining bound to arise in the future.” Id., at 96.

We specifically referred to longshoring as an industry with a long history of multi-employer bargaining, and we noted

“cogent evidence that in many industries the multi-employer bargaining basis was a vital factor in the effectuation of the national policy of promoting labor peace through strengthened collective bargaining.” Id., at 95.

*284The Board has authorized a multi-employer bargaining unit for West Coast shipping, and the labor agreement that forms the background to this case is additional “cogent evidence.”

At the same time, the very existence of multi-employer units, and the obvious need for the employers involved to agree on collective policy, must invariably have competitive effects. The signatories to a collective bargaining agreement are frequently, by the very act of signing, agreeing with their own competitors on matters such as labor costs, certain nonlabor costs, services to be provided to the public, and (indirectly) price increases.

Multi-employer collective bargaining must therefore be reconciled with the sometimes competing policies of federal laws promoting and regulating competition, viz., the antitrust laws and, in the case of maritime labor relations, the Shipping Act. This is a problem on which Congress has provided relatively little direct guidance,2 but it is one of a kind that the Court has repeatedly grappled with since Allen Bradley Co. v. Union, 325 U. S. 797. It is a problem of line-drawing.

The Court, noting that the assessment agreement levied $29,000,000, thus “necessarily resulting in substantially increased stevedoring and terminal charges,” ante, at 277, holds that the assessment agreement must be filed under § 15 of the Act. It says that the underlying labor agreement is not before us and the “continuing validity” of that agreement is not brought into question by today’s decision. Ante, at 278.

*285On the other hand, my Brother Douglas argues that on the Court’s premise the assessment agreement could not be distinguished from any collective bargaining agreement that “raised labor costs beyond the point at which PMA members could be expected to absorb those costs without raising prices or charges.” Post, at 313. He further contends that if part of a collective bargaining agreement is subject to Commission approval, this will stifle labor negotiation.3 Consequently, he suggests that a proper accommodation between “labor” and “competition” interests can be reached by exempting both labor agreements and labor-related agreements from the filing requirement of § 15 but leaving them subject to the specific prohibitions of the antitrust laws and §§16 and 17 of the Shipping Act.

This suggested accommodation seems to me demonstrably wrong. In the first place, as the Court notes, the filing requirement of § 15 was drafted broadly, and the filing-and-approval process includes review of questions arising under §§16 and 17, and specifically creates an exemption from antitrust attack. Hence, if the question were simply whether substantive challenge to a maritime agreement (dealing with labor or with any other matter) is to take place in advance of implementation of the agreement or, instead, during its operation, I should have thought it clear that Congress chose the former alternative. Furthermore, I would find it very difficult to see why provision for advance approval and exemption of labor-related agreements would not be *286preferable, from the standpoint of facilitating collective bargaining, to the “wait-and-see” approach.

The real difficulty in this case is not to distinguish between agreements that must be filed and agreements whose impact on competition will be evaluated after implementation, but to define the Commission’s jurisdiction in such a way that (whether challenges arise before or after implementation) the Commission will not improperly be brought into labor matters where it does not belong. The Court’s only suggestion is that the labor agreements involved in this case “fall in an area of concern to the National Labor Relations Board.” Ante, at 278.

More circumspect analysis than this is needed, I believe. In the first place, since the later validity and antitrust immunity of all agreements subject to § 15 depend upon filing, it is desirable that signatories to agreements be given more precise instructions than that they need not file if they are in an area of Labor Board “concern.” Furthermore, I see no warrant for assuming, in advance, that a maritime agreement must always fall neatly into either the Labor Board or Maritime Commission domain; a single contract might well raise issues of concern to both.

The Commission took the position that § 15 of the Act, requiring filing, was meant to apply “only to those agreements involving practices which affect that competition which in the absence of the agreement would exist between the parties when dealing with the shipping or traveling public or their representatives.”4 I agree with the Court’s conclusion that proper application of that principle to this case would require the opposite result from the one the Commission reached. The difficulty, however, is that the principle is exces*287sively broad: any significant multi-employer agreement on economic matters “affects competition” with respect to prices and services to the public, even if it is a collective bargaining agreement or an employer agreement collateral thereto.

Since maritime employers are permitted to bargain as a group, and since they are required to bargain about certain subjects, the resulting agreements must have some exemption from the filing requirements of § 15 and from successful challenge under the antitrust laws or under the substantive principles in §§16 and 17 of the Shipping Act. The exact extent of the “labor exemption” or “labor immunity” from statutes regulating competition has troubled this Court before;5 however, since no collective bargaining agreement in the maritime industry is now before us, it would be inappropriate to suggest the affirmative extent of the immunity. The important point in this case is an opposite and two-edged one: the assessment agreement before us is not immune or exempt, for it raises “shipping” problems logically distinct from the industry’s labor problems; at the same time, Commission review itself must be circumscribed by the existence of labor problems that it is not equipped to resolve.

The assessment agreement was, of course, consequent upon the labor agreement committing PMA to raise the fund. The union side was concerned with a guarantee that the fund would be raised somehow, and the *288labor agreement guaranteed only that much. But whenever any multi-employer bargaining unit agrees to provide benefits for employees there arises a problem of how to allocate the costs among the various employers and (in consequence) among their customers.

Often, the “allocation” decision follows directly from the terms of the labor agreement. In the case of a multi-employer agreement to raise wages, for example, each employer simply bears the cost of benefiting his own employees. In the present case, had it been possible to make the levy on each employer directly proportional to, and roughly simultaneous with, the savings to that employer from modernization, two things would have followed: the “allocation” decision could be said to stem directly from the terms of the labor agreement, and the modernization program would “pay for itself” as it went along, leaving shipping customers unaffected.

The PMA, however, did not (and presumably could not) apportion costs in this manner. To the extent that, under the plan chosen, individual employers were unable to absorb the levy and debit it against future savings from modernization, the decision how much each employer was to pay necessarily affected that employer’s customers as a class. To the extent that the plan went on to determine which of an employer’s customers would ultimately pay which share of an employer’s dues, the agreement also made choices among customers of an individual employer.

The Commission nevertheless held that the agreement did not “affect competition” because there was “no agreement” to pass the levy on to individual customers. Whether the error be deemed one of “fact” or one of “law” these conclusions are irreconcilable with reality. Terminal companies such as MTC compete with each other for the business of unloading Volks-*289wagens. The allocation agreement involved in this case increases the cost to a terminal company of unloading one Volkswagen by $2.35. This is a substantial (25%) increase in the company’s cost for handling this one product. Since the mechanization and modernization program is not expected to produce a significant (much less a compensatory) saving in the other costs of handling Volkswagens, no terminal company could, in the long run, “absorb” this cost: companies do not absorb costs that are not expected to pay dividends in the future.6 MTC’s only choice was whether to pass the $2.35 on directly to Volkswagen or to pass it on to its other customers in the form of higher charges or lower future savings from mechanization. Since the latter alternative would presumably have driven these other customers to deal with other terminal companies not bearing the Volkswagen curse, MTC was in practice compelled to pass at least a large part of the additional cost on to Volkswagen.

The statements of numerous officials of the participating companies to the effect that there were no “agreements” affecting Volkswagen (statements constituting in large part the “substantial evidence” on which the Commission is supposed to have relied, 125 U. S. App. D. C., at 291, 371 F. 2d, at 756), are at best quibbles about the meaning of words. The members of the Association must be taken to have agreed to the obvious consequences of the paper they all signed. That paper did not destroy price competition for Volkswagen’s trade; nor would a specific agreement to “pass on” the addi*290tional $2.35 charge have done so. But the allocation agreement made Volkswagen a less desirable customer to each and every terminal company unless it did pass on the $2.35 charge. How the Commission could conclude that this collective imposition, by the terminal companies on themselves, of a heavy tax for handling one kind of product did not “affect” competition among them for the trade of shippers of that product I simply cannot understand.

Commission review of the fairness of the agreement allocating the cost burden of mechanization does not mean Commission review of a labor agreement and does not imply consequences in conflict with national labor policy. Whether to mechanize, or otherwise modernize, and what provision should be made for displaced workers, are obviously matters of union concern, and negotiations about these things should be governed by the law of collective bargaining. Resolution of such questions by a decision to create a “Mech Fund” gave rise to a subsidiary “allocation” question. The union was concerned that the question receive some answer, but had no proper interest in which of the possible cost allocation plans was adopted, so long as any such plan raised the amount promised. On the other hand, in the present case no one has suggested that Maritime Commission review of a particular method of cost allocation may properly reach the question whether the obligation necessitating the allocation should have been entered into, or that the Commission may reject an allocation plan when there are no preferable alternative routes to collection of the necessary amount. Review of the fairness and propriety of a taxing scheme is not the same thing as reviewing the fairness and propriety of the uses to which the tax money, once collected, is put. When the Court notes that only the assessment agreement must be filed and examined, it seems clear that it contemplates a Com*291mission examination starting from the premise that the obligation to collect the Mech Fund will be fulfilled; at issue will be only the propriety of the choice of the route to that objective.7

II.

With respect to the §§16 and 17 issues, I consider that the Commission’s approach to those questions rested, as indeed the Court’s opinion now intimates, upon an erroneous understanding of the “assessment agreement” necessitating reconsideration of those matters on remand of the case.

The agreement that was before the Commission was, so far as appears, quite unlike any agreement that body had considered before. It dealt neither with a charge for particularized services in the carrying, handling, or storage of goods, nor with how such services would be provided. Rather, the agreement levied a “tax” on Association members, a tax which (insofar as the modernization program did not directly “pay its own way”) would be passed on to members’ customers and ultimately to the public. The tax would be used to pay for a general benefit to the shipping industry, but the allocation of that tax bore no direct relationship to benefits received by customers.

The Court holds that it was error for the Commission to reject challenges to this agreement under § 16 simply because there was no showing that the tax was discriminatory as between competitive customers. It declares that such a rule may be sound in cases involving rates *292for sea carriage of goods because of the “particularized economics” resulting from the finite capacity of ships, but that it is not sound elsewhere, including this case, for unspecified reasons.

On the surface, it might appear that the argument should be the other way around: it makes some sense to speak of an “undue or unreasonable preference or advantage” to, say, watermelons over automobiles when they are “competing” for a finite amount of shipping space; it becomes much more difficult to find anything that can be called a “preference” between such products with respect to any services that are available to both in unlimited quantities.

My Brother Douglas states that the Commission has consistently adhered to its insistence upon a competitive relationship between the product preferred and the product disadvantaged, except where “there are services that are not dependent upon the nature of the cargo and the various charges therefor.” Post, at 314, n. 30. Yet, if ever it was clear that “the nature of the [products]” was not the basis for a difference in rates, it is in this case.

The true solution of the matter, it seems to me, is that in each situation the problem has been to devise some workable basis for determining whether rates are fair vis-á-vis other rates. It simply would not be feasible, as the Second Circuit has noted,8 to assess the fairness of charges for shipping heavy industrial equipment by comparison with the cost of shipping bananas. The notion of a “preference” for bananas over heavy equipment is simply too elusive to be implemented. At the same time, when the service rendered is, for example, procuring insurance or arranging for cartage, *293the nature of the product has very little to do with either the value to the customer of the services rendered or the cost of supplying them; in such cases the Commission has quite reasonably held that charging different classes of shippers different amounts for equivalent services may be preferential.9

In the present case, the problem before PMA was the allocation of a pre-specified total cost among its various members and their customers. Since this was very much a case of first impression, the Commission would have done well to go back to the language of § 16, which proscribes any “undue or unreasonable preference or advantage to any . . . description of traffic in any respect whatsoever.” 10 Certainly, since a “modernization tax” on any one group of customers lowered, by an equivalent amount, the cost of modernization to others obligated to pay for it, an unfair allocation of the burden could properly be described as a “preference” between that “description of traffic” bearing a heavy burden and that “description of traffic” whose burden was correspondingly lightened.

The real difficulty in this case is to formulate a workable definition of whether the burdens have been “unfairly” allocated. Obviously, as the debates in the PMA indicate, there was no “perfect” way to apportion the costs. Any analysis of the present problem must leave room for the implementation of some uniform, practical, general rule of assessment even though it have some features that are less desirable than some alternative imperfect rule. The difficulty with the method of assessment adopted by PMA is that it was not uniform and general but made special provision for automobiles. The fact that all automobiles are treated alike should *294not have prevented the Commission from inquiring whether special treatment for this class of goods was necessary- under the circumstances and, if so, whether the special rule adopted was the fairest that could be devised.

The Commission’s interpretation of § 17 was also erroneous. The Commission held that since petitioner received substantial benefits from the modernization program it would not make minute inquiry into whether petitioner’s benefits precisely corresponded to the costs imposed. The first difficulty is with the conclusion that petitioner received “substantial benefits.” Petitioner apparently is not in a position to profit appreciably from maritime modernization. ' Petitioner will, of course, benefit from any lessening of labor disputes in shipping and related services; but the only disruptions that are avoided by the labor agreement reached here are those that would otherwise have resulted from the efforts of other shippers and of maritime employers to institute the very modernization practices that will not benefit petitioner. It may be that those who will directly benefit from modernization and those who will benefit only from increased stability during the course of a modernization program in which they have no interest (and which others have imposed on them) should both pay part of the cost of the Mech Fund. However, the existence of such a categorical difference between the benefits received by different groups should at least invite inquiry whether charges are as appropriately proportioned as would be feasible.

In fact, the tax assessed is not “equal” as between Volkswagen and other shippers who will benefit more. The charge to MTC per Volkswagen was figured on a different basis from the assessments for handling other products; the figure reached was a substantially higher percentage of existing costs and charges; and the figure was so high that the additional cost apparently could *295not be absorbed and debited against future sayings from modernization but had to be passed on to the customer. Of course charges need only be “reasonably” related to benefits, and not perfectly or exactly related, Evans Cooperage Co. v. Board of Commissioners of the Port of New Orleans, 6 F. M. B. 415, 418, but in this case inquiry ceased before it had reached even that nearer point.

Finding no disagreement in principle between myself and the Court, I join the Court’s opinion upon the premises stated in this opinion.

dissenting in part.

I believe the Court has misconstrued § 15 of the Shipping Act, 1916;1 and I fear that its erroneous construction will cause serious disruption in the process of *296collective bargaining in the maritime industry. If the tariff exacted from petitioner is discriminatory or unreasonable, §§16 and 17 of the Shipping Act2 provide a remedy. If it violates the antitrust laws, there is also a remedy, as I shall indicate. But to require the funding part of maritime collective bargaining agreements to receive prior approval from the Maritime Commission is to use a sledge hammer to fix a watch. I cannot read § 15 so as to attribute to Congress such a heavy-handed management of sensitive labor problems.

The collective bargaining agreement involved in this case, with its Mechanization and Modernization Fund (Mech Fund), cannot be evaluated properly without an understanding of maritime labor relations and technological developments in the shipping industry.

The history of maritime labor relations in this country has been punctuated with lengthy major strikes and *297continuous minor disruptions.3 The maritime industry has long been faced with problems of instability — economic and managerial. Employment for maritime workers is generally both irregular and insecure.4 That condition lies behind the large number of major strikes and work stoppages on our coasts.

Because the shipping industry is vitally important both to our national commerce and national defense, the Federal Government has maintained a special interest in trying to promote its growth and stability. The Shipping Act, 1916, is one example of this concern.5 With respect to maritime labor relations, however, the activities of the Federal Government were, until our entry into World War I, primarily devoted to laws protecting or disciplining seamen as individual workers. The war years saw the Government actively encouraging collective bargain*298ing in the maritime industry, its efforts resulting by 1920 in a significant expansion of collective bargaining. There followed a general retrogression, with wages and working conditions reaching low levels. That condition prevailed until the highly disruptive and violent Pacific Coast strike of 1934.

That strike was the product of deep-seated grievances of maritime employees regarding low wages and poor working conditions.6 The situation on the Atlantic Coast was not much better. Although an agreement was reached in late 1934 for Atlantic Coast workers, labor relations remained unstable and work stoppages were rampant. On both coasts, intra-union and inter-union disagreements, coupled with employer-union hostility, made agreement highly difficult. Quickie strikes dotted the ports, and another general strike followed in 1936. On the Pacific Coast, the employers and the ILWU (which had achieved recognition after the 1934 strike) were in constant conflict through 1948, when still another general strike erupted. This period, from 1934 to 1948, has been *299aptly described as something like “class warfare.” 7 As one commentator put it:

“The ILWU (then a part of the AFL International Longshoremen’s Association) gained formal employer recognition as a result of the general strike of 1934, which followed years of exploitation and abuse of longshoremen by their employers. The bitterness which had characterized the industry carried over into the subsequent employer-union relationship. The employers did their best to break the union, and the union retaliated just as militantly. The years which followed were among the stormiest in U. S. labor history. Between 1934 and 1948, the West Coast had over 20 major port strikes, more than 300 days of coastwide strikes, about 1,300 local 'job action’ strikes, and about 250 arbitration awards.” 8

During the stormy 1930’s, the Federal Government was greatly expanding its role in labor relations. The NIRA and NLRA greatly revived unionism among both seamen and longshoremen in addition to workers in other industries. Those Acts guaranteed the right of collective bargaining and offered a means for recognition of unions; the unions gained members and strength. And with stronger unions, collective bargaining became more widespread. But the explosive situation in the maritime industry was not solved by these general enactments, and Congress passed a series of laws to deal with the labor problems in that industry. First was the Merchant Marine Act, 1936, 49 Stat. 1985, creating a United States Maritime Commission to investigate conditions of seamen on ships and to determine minimum wage scales and working conditions on vessels that were receiving govern*300ment subsidies. Despite the 1936 Act, labor relations did not improve significantly; and Congress in 1938 amended the Act, creating a Maritime Labor Board (MLB) with the duty of encouraging collective bargaining and assisting in the peaceful settlement of labor disputes through mediation. A provision of the 1938 amendment, § 1005, 52 Stat. 967, required all maritime employers to file with the MLB within 30 days a copy of every contract with any group of its employees covering wages, hours, rules, and working conditions. Any new contract or change in an existing contract also had to be filed with the Board. The contracts did not require approval by the Board, but were to be used to assist the Board in its mediation activities and in its promotion of peaceful settlement of labor disputes.9

The Board was instructed in the 1938 Act to submit to Congress by 1940 its recommendations for establishing a permanent federal maritime labor policy ensuring stable labor relations. The Board in its 1940 report concluded that conditions in the industry were still uneasy, and recommended a permanent federal body with wide jurisdiction over questions of maritime labor— including representation10 and settlement of disputes. *301The 1938 Act provided that the Board was to be discontinued in 1941; but in 1940 Congress extended its life until mid-1942 to permit further studies by the Board and Congress. Nothing more was done until 1955 when Congress again turned its attention specifically to the problems of maritime labor relations.11 In the meantime, the MLB had expired. Although several bills were introduced providing for specialized federal control over maritime labor relations, no special machinery was established; and the maritime industry remains subject to the various provisions of federal labor laws.12

In 1948 another general maritime strike rocked the Pacific Coast. Following that strike, which lasted about 100 days, there was a “period of relative calm.” 13 The 1948 strike had led to a change in employer leadership, a less hostile attitude on the part of the union leadership, *302and a consequent lessening of tension along the Pacific Coast. Both sides recognized that the reduction of strife was desirable since a substantial amount of trafile had been diverted from the Pacific Coast to other ports or to other means of transportation on account of chronic maritime labor difficulties and work stoppages. But despite the reduction in hostility between labor and management, solutions to problems were not readily forthcoming. Business was bad for the shipping companies — foreign competitors had cut heavily into the market, and a decline in business meant less work for both seamen and longshoremen. Modernization was sorely needed, but it was also greatly feared, for mechanization would cut out jobs. But without improved techniques and facilities, the employers could not regain a strong competitive position.14 In addition to lack of modern *303equipment, employers were further hampered by highly restrictive work rules that had been in effect since the 1930’s, such as multiple handling,15 sling-load weight,16 and gang-size restrictions.17

*304It is only against this background of chronic strikes and restrictive labor practices that the tremendous impact of the Mech Fund can be appreciated. That was the heart of the 1960-1961 settlement. As noted by one commentator intimately acquainted with the negotiations of the parties, “[t]his agreement did not spring full-blown from the brow of Zeus, or from the brain of Bridges.”18 Rather, “[t]he agreement, which was hammered out in 5 months of negotiations ending in October 1960, culminated 4 years of discussion between the PMA and the ILWU.” 19

Earnest bargaining began in 1957. PMA wanted to obtain a guarantee from the ILWU that strikes and work stoppages would not result from the introduction and use of mechanization and other labor-saving devices. In return, the union wanted its workers to share in the cost savings resulting from modernization, and desired assurances that changes in work methods would neither create unsafe working conditions nor accelerate the productivity required of individual workers. After two years of preliminary negotiations, an agreement was made in August 1959 which provided for a further study of the problems of mechanization and for the establishment by PMA of a fund of $1,500,000 for the benefit of union workers.20

Negotiations beginning in May 1960 led to a “Memorandum of Agreement on Mechanization and Moderni*305zation,” concluded in October 1960, and providing for a $29,000,000 trust fund to be financed by PMA. The fund was to consist of the $1,500,000 due under the 1959 agreement plus another $27,500,000 to be accumulated over a five-and-one-half-year period at the rate of $5,000,000 per year. The fund was to be used to protect longshoremen and marine clerks from the consequences of reduced employment caused by mechanization. The agreement was to enter into force upon approval by the members of PMA and the ILWU, and was to expire on July 1, 1966.21 The agreement also provided management with the relatively free rein it had sought to eliminate restrictive work practices. The former practice of multiple handling was eliminated, and the minimum size of a gang for loading and unloading operations was specified. The sling-load limit for loads was to remain unchanged if the manner of operation was the same as when the limit was first negotiated; otherwise, the employer could set the weight, provided that he acted “within *306safe and practical limits and without speed up of the individual.”

Thus, the agreement satisfied the desire of employers to modernize and eliminate outmoded and restrictive work rules, and at the same time provided a measure of security for the workers whose jobs would be affected by the use of the new devices. The agreement, however, left open the question of how the employers’ contributions of $5,000,000 a year would be raised. The question of a proper method of assessment had been discussed by the union and management during the preceding negotiations; several suggestions were offered by the parties. But in return for a commitment from the PMA members obligating themselves individually and collectively to the payment of the fund, the ILWU agreed to permit PMA to establish the method of payment.

PMA then set up a Work Improvement Fund Committee to determine the best method of raising the money. That Committee considered various bases for assessing contributions — man-hours of each employer, cargo tonnage, a combination of the two, cargo tonnage moving in containers, measurement of improvements in longshore productivity. The Committee majority recommended a cargo tonnage basis; its reasons for doing so were summarized by the court below as follows:

“The Committee recommended a formula based on cargo tonnage as a 'rough-and-ready’ way to divide the cost, admittedly lacking the refinement of the productivity measurement method but also lacking its infeasibility and avoiding the inequity of the man-hour method whereby contributions are in inverse proportion to benefits received. It considered that cargo volume though not necessarily proportional was some indicator of stevedoring activities and that administrative simplicity was a cardinal consideration.

*307“The Committee recognized further that there were also objectionable features of the tonnage formula but considered these to be less weighty than the objections inhering in the other formulae. It recommended that the formula be reviewed to prevent the continuation of any hardship or inequity that might develop.” 22

In recommending the tonnage formula, the Committee noted that the same system was used for assessing a part of PMA dues. It had also been the practice of PMA to use a tonnage formula for assessments allocating other types of labor costs, such as joint maintenance of dispatch halls and the payment of arbitrators’ salaries.23 In fact, it appears that the ILWU had itself proposed a tonnage formula during the negotiations and asked that it be incorporated into the collective bargaining agreement; but PMA resisted this approach, apparently wishing to keep its options open and fearing that incorporation in the agreement might tend to commit the PMA to a fixed formula that would also be included in a future agreement. The tonnage formula recommended by the Committee was subsequently adopted by the PMA membership.

It was specifically provided in the agreement that each employer would abide by the formula adopted by the *308Association; and this promise to comply was the quid pro quo for the union’s agreement not to write any particular formula into the contract or take part in the determination of the method of assessment.24 Thus the PMA decision on the method of assessment was part and parcel of the collective bargaining agreement. Indeed, the *309modernization plan was the heart of that agreement, and the subsequent assessment plan merely implemented the employers’ duty under the collective bargaining agreement to establish a fund specifically marked to protect maritime workers against the far-reaching effects of modernization.

PMA treated the financing of the fund as an integral part of the collective bargaining process. The Committee established by PMA to recommend a funding formula was appointed by the negotiating committee which worked on the collective bargaining agreement; 25 and the PMA membership ratified both the collective bargaining agreement and the funding formula at the same time.

It is not, I submit, possible, as a practical matter, to separate the Mech Fund provision in the collective bargaining agreement from the subsequent decision of *310the PMA membership concerning how the fund was to be raised. A collective bargaining agreement is the product of negotiations. How can negotiators sitting at a table arrive at an agreement if they know that a major part of it depends on the approval of the Federal Maritime Commission? How many months — or years — will it take to get approval? What will happen meanwhile? Will not the imposition of that kind of administrative supervision bring an end to, or at least partially paralyze, collective bargaining?

The Mech Fund is a labor expense. Increased labor costs normally are passed on at least in part by increased prices. When the Auto Workers were recently negotiating with General Motors for a guaranteed annual wage, what would have been the consequence if nothing could have been decided until a federal agency had determined whether the impact on prices or on the economy was proper? I can imagine a regime of total controls where such prior approval would be required. But we have no such regime at present; and I can see no possible justification for a judicially created one in the explosive maritime field. To meet the costs increased by any collective bargaining agreement, a company might have to raise its prices and pass at least part of the added cost on to the consumer. But this happens all the time in the maritime industry, as well as in other industries, and does not constitute rate fixing of the type at which the Shipping Act is aimed. There is nothing in the legislative history of the Shipping Act which suggests that §15 gives the FMC the power or license to oversee labor negotiations. But that is the effect of what the Court does today when it decides that the employers’ agreement here must be submitted to that body for approval.

My Brother Harlan suggests that the assessment agreement can be distinguished from the collective bargaining agreement because “[t]he union was concerned *311that the question [of how the cost burden of the fund was to be allocated] receive some answer, but had no proper interest in which of the possible cost allocation plans was adopted . . . (Ante, at 290.) But to argue that the union does not care from what source the PMA gets the money for the fund is both questionable26 and irrelevant, for such an approach ignores the fact that there are two parties to a collective bargaining agreement. The PMA members do care how they will be assessed $27,500,000 for a fund dedicated to the benefit of their employees. The Mech Fund was the key provision in the agreement, and without it there may well have been no agreement at all. The parties should not be expected to wait to settle their differences while the FMC decides under § 15 whether the employers’ funding plan is in the public interest. Speedy resolution of labor disputes by collective bargaining has been the consistent federal policy.

The Solicitor General would have us atomize the collective bargaining agreement and treat the schedule of charges that create the fund as a mere “side agreement.” But without the so-called “side agreement” there would have been no collective bargaining agreement. And it must be remembered that § 15, if applicable, requires that an agreement be filed “immediately with the Commission.” What would have to be filed is the entire agreement, not merely the proviso to which petitioner now objects. The Commission then must give notice and a hearing and “disapprove, cancel or modify” the agreement. Which persons would be entitled to participate in the hearing presents an initial problem.27 Thereafter, what provisions would become the target in the hearing *312is conjectural. The target might be small or large. But certainly no collective bargaining agreement could become operative until its underpinning — the fund— was thoroughly litigated. Meanwhile years might pass as the contest wound its way slowly through various tribunals and the labor problems continued to fester.

This is what my Brother Harlan overlooks when he suggests that advance approval of “labor-related agreements” might be more desirable from the standpoint of facilitating collective bargaining than leaving open the question whether the agreement, or parts of it, would be subject to the antitrust laws. Presumably, he means that legal uncertainty concerning the possible vulnerability of certain provisions of an agreement to attack under the antitrust laws might stall negotiations or lead some association members to decline to cooperate in carrying out the agreement, fearing a treble-damage action. To be sure, the parties to a collective bargaining pact must frame their agreement to fit within the standards of the antitrust laws or any other governing statutes. But without a requirement of advance approval of the terms of the agreement, they remain free to bargain speedily. Frustration of the collective bargaining process comes not so much from the possibility that one or more provisions in a collective bargaining pact might be found illegal at some future date under the antitrust laws, or other statutes such as §§16 and 17 of the Shipping Act, but rather from the undue and possibly lengthy freezing or stultification of solutions to troublesome labor problems while an intimate part of the proposed agreement is sent to the FMC for approval.

With all respect, the Court’s approach in requiring the funding plan to be submitted to the FMC for approval under § 15 of the Shipping Act will frustrate legitimate and speedy collective bargaining in the maritime indus*313try. Neither the Court nor my Brother Harlan is able to refer to any legislative history which indicates that Congress considered the Shipping Act to require the filing of labor agreements or provisions of those agreements under § 15.28 The Court instead takes the approach that the Shipping Act provisions were purposely drawn broad enough to encompass association agreements which have more than a de minimis effect on commerce. This rationale would require the filing of any collective bargaining provision agreed to by PMA members that raised labor costs beyond the point at which PMA members could be expected to absorb those costs without raising prices or charges.

The Court may well mean, as my Brother Harlan suggests, that the “obligation to collect the Mech Fund,” contained in the collective bargaining agreement, is not to be examined by the Commission on remand, but rather the question is to be limited to the “propriety of the choice of the route to that objective.” But that misses the mark. My point is that the latter question is as much a part of the bargaining process as the former. Commission control over either question runs substantial risk of frustrating agreement by the parties on both issues, not to mention other matters in the collective bargaining pact. For example, if an allocation formula satisfactory to PMA members and to the Commission could not be devised, the fund might never be established, requiring perhaps other changes in the agreement, such as higher wages or continuance of some or all of the restrictive work rules.

If the present practice is an abuse, there is an existing remedy. This agreement between employers could of *314course be challenged in the courts as violative of the antitrust laws.29 Moreover, §§16 and 17 of the Shipping Act afford protection to foreign commerce in cases of undue discrimination or unreasonable practices affecting that commerce. While I cannot say that the Commission erred in finding no violation of § 16,1 concur in a remand to the Commission for further findings under § 17.30 *315If the finding is for petitioner, there may be an incidental and after-the-fact effect on the provisions of the collective bargaining agreement. But it will not produce *316the paralyzing effect which will follow when prior approval is required. The application of §§16 and 17 in particular instances can indeed realistically be compared with enforcement of federal antitrust laws directed against specific practices.

concurring in the judgment.

I agree with the judgment and with Part I of the opinion herein. I do not understand that the Court’s opinion purports to determine the effect of §§ 16 and 17, and I believe that the Court certainly should not do so. I do not join Part II of the opinion dealing with these sections.