United Parcel Service, Inc. v. United States Portal Service

455 F. Supp. 857 (1978)

UNITED PARCEL SERVICE, INC., a New York Corporation and United Parcel Service, Inc., an Ohio Corporation
v.
UNITED STATES POSTAL SERVICE.

Civ. A. No. 77-3620.

United States District Court, E. D. Pennsylvania.

July 19, 1978.

*858 *859 Irving R. Segal, Robert L. Kendall, Jr., John E. McKeever, Philadelphia, Pa., for plaintiffs.

Robert N. DeLuca, Asst. U.S. Atty., Philadelphia, Pa., John L. DeWeerdt, Associate Gen. Counsel, U. S. Postal Service, Washington, D.C., for defendant.

OPINION AND ORDER

EDWARD R. BECKER, District Judge.

I. Preliminary Statement

This case raises the question whether the United States Postal Service may, without notice, hearing, or prior submission to the Postal Rate Commission, conduct limited and temporary experiments in which proposed new mail classifications, offered at heretofore unavailable rates, are tested in the marketplace or, conversely, whether the institution of such experimental postal services is subject to the procedures governing changes in permanent rates and classifications.[1] The question appears to be one of first impression.

The United States Postal Service and the postal procedures at issue in the case at bar find their common genesis in the Postal Reorganization Act of 1970, Pub.L. 91-375, 84 Stat. 719 (codified at 39 U.S.C. §§ 101-5605) [hereinafter "Act"]. Passage of the Act signalled a major change in the structure and accountability of the postal bureaucracy. Critics of the Post Office Department claimed that the Post Office was at once over-political and under-efficient, see generally H.R.Rep. No. 91-1104, 91st Cong., 2d Sess., reprinted in [1970] U.S.Code Cong. & Admin.News, pp. 3649, 3651-54, 3660-61; S.Rep. No. 91-912, 91st Cong., 2d Sess. 1-3 (1970); President's Commission on Postal Organization, Towards Postal Excellence (1968), and that the Postmaster General was generally powerless to take the kind of extensive remedial action necessary to affect significantly the Department's all too apparent ills, see S.Rep., supra, at 3.[2]

Congressional response to these postal maladies was far-reaching. The Post Office was taken out of the Cabinet and established as the United States Postal Service, an independent unit of the executive branch. 39 U.S.C. § 201. Under the new system, the Postmaster General is no longer appointed by the President; rather, the *860 President appoints the nine Governors of the Postal Service, no more than five of whom may be from the same political party. The Governors, who serve for rotating nine year terms, appoint the Postmaster General and Deputy Postmaster General and set the latters' tenure and compensation; the Governors and their two appointees become the Board of Governors of the Postal Service. Id. § 202.

The Postal Reorganization Act gives to the Postal Service increased control of and responsibility for postal operations and management, free of the formal and direct supervision of the legislative and executive branches. Thus, the Postal Service is empowered to sue and be sued in its own name, to adopt rules and regulations, to enter into and perform contracts, to keep its own system of accounts, to acquire and dispose of property, etc. Id. § 401. To professionalize and depoliticize the Postal Service even further, a postal career service was established, id. § 1001(a), provision was made for hiring executives, id. § 1001(c), and otherwise upgrading postal management, id. § 1004, and political recommendations on personnel matters were prohibited, id. § 1002. As to employee-management matters, collective bargaining was made applicable to postal service employees, with provision for binding arbitration in the event of lack of agreement, id. §§ 1202-1207, and postal employees were granted the right to work, id. § 1209(c).

Reform in the area of setting postal rates and classes was at least as thoroughgoing as the other reforms just referred to. Congress abandoned the business of postal rate-making and established a quasi-regulatory agency, the bipartisan Postal Rate Commission. The Commission, which is independent of the Postal Service and the members of which serve for rotating six year terms, was given primary responsibility for overseeing changes in postal rates, classes, and services. Id. §§ 3601-3604. Although the Governors of the Postal Service are charged with the establishment of fair and equitable classes of mail and postal rates and fees, id. § 3621, before instituting such rates or classes the Postal Service must request recommended decisions from the Postal Rate Commission, id. §§ 3622 & 3623. The Commission is to hold evidentiary hearings under 5 U.S.C. §§ 556 & 557 (Administrative Procedure Act) pursuant to regulations that it may establish, 39 U.S.C. § 3624, to evaluate the proposed rates or classes of mail according to criteria established by Congress, id. §§ 3622 & 3623. Upon receiving the recommended decision of the Commission, the Governors of the Postal Service may approve, allow under protest, reject, or modify the decision in accordance with specific statutory provisions. Id. § 3625. If no decision is forthcoming from the Commission within a specified period of time, the Postal Service may, upon ten days notice in the Federal Register, establish temporary changes in rates or classes in accordance with the proposed changes under consideration by the Commission. Id. § 3641.[3] Proposed *861 changes in the nature of postal services that will generally affect service on a nationwide or substantially nationwide basis must also be submitted to the Commission for hearing and an advisory opinion before such changes may be instituted. Id. § 3661.

Responding to its congressional mandate to "plan, develop, promote, and provide adequate and efficient postal services at fair and reasonable rates and fees," id. § 403(a), in mid-October of 1977 the Postal Service instituted an experimental parcel delivery service that is scheduled to last approximately twelve months. Under this test program, which is not available to the public generally, about twenty selected shippers[4] in five metropolitan areas[5] entered into service agreements with the Postal Service. These agreements provide that parcels mailed by the test participants, when suitably packaged for bulk mail processing,[6] will be assessed postage based upon the average weight of such packages. For example — and subject to the distance limitations set out in note 7 infra—regardless of the actual weight of any given package, if the average weight of the pieces tendered, based on sample weighings, is five pounds or less, postage for each parcel is $.87; if the average weight of the packages is more than five but not more than twenty pounds, postage for each parcel is $1.15. These charges are significantly less than the generally available parcel post rates of the Postal Service. Further, the postage charges under the test program do not, for the most part, vary with distance, as they do under regular parcel post.[7] Participants in the program must provide a minimum volume of 50 pieces per mailing *862 and 250 pieces per week. They also must present all parcels to the Postal Service in containers and cooperate with the Postal Service in data collection and research activities.[8]

The plaintiffs, large nationwide providers of parcel delivery services that compete with the Postal Service's parcel post program and that would suffer adverse competitive impact were the current Postal Service test instituted on a permanent and systemwide basis,[9] contend that the Postal Service's experiment is in violation of the policy and procedures established by the Postal Reorganization Act. Although the particular experiment at issue in this case poses a relatively small threat to plaintiffs, see notes 9 supra & 11 infra, the plaintiffs fear that other less restricted experiments will be deleterious. Choosing to strike before significant harm is done rather than after, they attack the precedent of marketplace experiment sans Rate Commission approval which the Service Test Plan establishes. The primary focus of this attack on the Service Test Plan is grounded on the procedures mandated by 39 U.S.C. §§ 3622-3624 for rate and mail classification changes.[10] Plaintiffs contend that, insofar as the test takes place in the marketplace, because the rates charged to participants in the experiment are previously unavailable rates and because the plan represents a new class of mail, the Postal Service is required to submit the plan to the Postal Rate Commission and await the Commission's recommended decision before instituting the test. Plaintiffs argue that the Postal Service may not lawfully circumvent the statutorily required procedures for postal rate and classification changes merely by unilaterally denominating the changes "experimental." Since the statutory procedures, which contain no exceptions for market "experiments," were not followed, plaintiffs argue that the Service Test Plan should be enjoined.[11]

The Postal Service rejoins that, although the procedures cited by plaintiffs would indeed be applicable if the Service Test Plan were instituted on a permanent and systemwide basis, those procedures are not triggered where, as here, a reasonable, limited, and temporary experiment is conducted as a research and development activity pursuant to the Postal Service's duty to "plan [and] develop . . . adequate and efficient *863 postal services." 39 U.S.C. § 403(a). The experiment, the Postal Service maintains, is designed to assist the Service in determining what, if any, rate and/or classification changes it should make in order to better utilize the Bulk Mail System, and is not itself such a change.

For reasons that will at length appear, although we believe that as a matter of pure policy the result advocated by the Postal Service is more desirable, construction of the statute compels the conclusion that the plaintiffs must prevail and that continuation of the "experimental" parcel delivery service at issue must be enjoined.

II. Discussion

A. The Statutory Language; Some General Principles of Statutory Construction and the Principle of "Plain Meaning"

We begin our inquiry by focusing on the statutory language that governs the Postal Service's submission of changes in rates and classifications to the Postal Rate Commission. Section 3622(a) of the Postal Reorganization Act provides:

From time to time the Postal Service shall request the Postal Rate Commission to submit a recommended decision on changes in a rate or rates of postage or in a fee or fees for postal services if the Postal Service determines that such changes would be in the public interest and in accordance with the policies of this title. The Postal Service may submit such suggestions for rate adjustments as it deems suitable.

39 U.S.C. § 3622(a). Similarly, the next section provides:

Following the establishment of [an initial] mail classification schedule . . . the Postal Service may from time to time request that the Commission submit, or the Commission may submit to the Governors on its own initiative, a recommended decision on changes in the mail classification schedule.

Id. § 3623(b).

Our objective as we seek to determine the meaning to be given this statutory language is, of course, to give effect to the congressional intent. Philbrook v. Glodgett, 421 U.S. 707, 713, 95 S. Ct. 1893, 44 L. Ed. 2d 525 (1975). We think it reasonable to assume that in expressing its intent through the Postal Reorganization Act, Congress sought clarity rather than obfuscation and, unless the contrary appears, that Congress therefore used statutory words in their ordinary and usual sense and with the meaning commonly attributed to them. Caminetti v. United States, 242 U.S. 470, 485-86, 37 S. Ct. 192, 61 L. Ed. 442 (1917); Institute for Scientific Information, Inc. v. United States Postal Service, 555 F.2d 128, 131 (3d Cir. 1977).

Reading the statute according to this general principle, there does not on the surface appear to be anything inherently difficult or ambiguous about the language involved. The common meaning of a "rate of postage" is simply the fee the Postal Service charges for processing, transporting, and delivering a particular type of mail. Similarly, the ordinary understanding of a mail classification is nothing more complex than a grouping of mail based on established and shared criteria. See National Retired Teachers Ass'n v. United States Postal Service, 430 F. Supp. 141, 146 (D.D.C.1977). The usual interpretation of "change" is alteration or making different. Although the Postal Service argues vigorously that, in the context of Postal Rate Commission supervision, "changes" in "rates of postage" and "mail classifications" signifies only permanent and systemwide rate and classification changes rather than the limited experimental adjustments in rates and classifications at issue in the instant case, the statute itself does not at any point set out or specifically allude to any exceptions to the general scheme,[12] nor *864 do the particular provisions on which we have focused expressly limit their applicability to changes of a nationwide, systemwide, or permanent scope. On the face of the statute, therefore, there appears no reason to restrict the ordinary and usual scope of the statutory terms, which, we think, would generally be read to include any change in rates or classifications.

Turning our attention to the particular actions of the Postal Service that are challenged in this case, there can be no doubt that the Postal Service's experiment does involve rates of postage and classifications of mail within the ordinary meaning of those terms. Fees are being paid to the Postal Service to transport and deliver a certain type of mail that is grouped according to established and shared criteria. Indeed, as we have already observed, the Postal Service concedes that the very changes at issue would have to be submitted to the Postal Rate Commission pursuant to §§ 3622 and 3623 were these innovations permanent and systemwide. Further, it is clear that as to the twenty participants in the Service Test Plan, postal rates and mail classifications have in fact been changed, for the participating shippers now mail their packages under previously unavailable rates as part of a previously non-existent class of mail. So stated, where the "experiment" does not consist of mock packages mailed at hypothetical rates but instead requires the payment of real postage by real shippers for real parcel post service, there have been changes, however limited, in rates of postage and classifications of mail that affect these shippers and that fall within a "plain meaning" reading of the statutory language.

There is a possible counterpoint to our "plain meaning" analysis, which can be constructed by analogy to 39 U.S.C. § 3661;[13] that is, by comparing, in the postal context, rate, classification, and service changes.[14] When one thinks of a rate or classification change, the argument goes, one naturally tends to think in systemwide terms; i. e., when one thinks of a change in the rates, for example, of parcel post, it is the more "common" understanding that such a change would be across the board, and not a change affecting only twenty mailers and only for a predetermined length of time. The "common" or "ordinary" conception of a classification change might be said to be similar. In the case of a service change, however, the argument continues, one might more naturally think in less sweeping terms. To be sure, a service change could be systemwide. But it commonly also could be purely local — for example, the decision to discontinue home delivery in a single town when the mailboxes are moved farther away from the curb. Grover City v. United States Postal Service, 391 F. Supp. 982 (C.D.Cal.1975). See also Bradley v. United States Postal Service, 554 F.2d 186 (5th Cir. 1977) (decision to discontinue door-to-door delivery in one town not a service change of nationwide or substantially nationwide scope); Martin v. Sloan, 432 F. Supp. 616 (W.D.N.C.1977) (consolidation of two rural postal routes not a service change of nationwide or substantially nationwide scope). Thus, in § 3661 Congress had to modify the ambiguous "plain meaning" of a "service change" to make it clear that Postal Rate Commission involvement was to be only in the broader, systemwide situation. Under this theory, no such modification of the plain meaning of rate or classification changes was necessary in § 3622 or § 3623 because the ordinary usage of such terms already includes a "systemwide" component, but modification would have been necessary to expand Rate Commission jurisdiction to the less-than-systemwide situation. The argument concludes *865 that the plain meaning reflects, consistent with § 3661, a congressional desire to reserve Rate Commission review for the "big" case.

The effect of the foregoing argument, if successful, would be to reverse the presumption under which we otherwise would be laboring. For, if the "plain meaning" of the statute is as we initially articulated it, it is incumbent on the Postal Service to demonstrate that imputing such a meaning to the words of the Act would be unreasonable in terms of the Act's purpose and structure. See discussion infra. If, however, the "plain meaning" is as the Postal Service posits it to be, the presumption reverses and it becomes the burden of the plaintiff to show unreasonableness.

Although the Postal Service's argument has some appeal, we ultimately are unconvinced by it. There is nothing "plainer" than the fact that postal rates and classifications have been changed as to the participants in the Service Test Plan. When the rates and classifications applicable to certain users of the Postal Service have in fact been changed, we find it both conceptually and linguistically bizarre to say that although certain postal rates and classifications have been changed, there has not been a change in the rates or classification of postage. We do not say that such a statutory definition is impossible; the ultimate length of this opinion is testimony to its possibility. We do, however, say that it strains "plain meaning" to impose such a definition on the language as a presumptive matter.

Moreover, this argument ultimately hinges on the assumption that Congress wanted the approach to rate, classification, and service changes to be identical vis-á-vis Rate Commission involvement. Congress' linguistically different treatment of the different kinds of changes, however, may well indicate that the changes were thought of in different categories because, for example, of the different areas of expertise of the Postal Service (service management) and the Postal Rate Commission (ratemaking). See National Association of Greeting Card Publishers v. United States Postal Service, 186 U.S.App.D.C. 331, 358, 569 F.2d 570, 597 (1976) [hereinafter "NAGCP"]. Thus, comparing rate changes and service changes is like comparing apples and oranges. Even without drawing an inference adverse to the Postal Service from § 3661, we therefore are unable to employ § 3661 to provide a restrictive gloss on the facially comprehensive language of §§ 3622 and 3623.

Rather than lowering the curtain on our inquiry into the meaning of the statute, however, we think that the foregoing actually only sets the stage for further analysis. Although it is received wisdom that when a statute's plain meaning is clear "the duty of interpretation does not arise and the rules which are to aid doubtful meanings need no discussion," Caminetti v. United States, supra, 242 U.S. at 485, 37 S.Ct. at 194, it is also an endorsed caveat to this rule that "[w]hether or not the words of a statute are clear is itself not always clear." Barbee v. United States, 392 F.2d 532, 535 n.4 (5th Cir. 1968). Thus, what might appear to be the plain meaning must yield if it is inequitable, unreasonable in context, or manifestly contrary to the statutory purpose or legislative intent. See, e. g., Center for National Policy on Race and Urban Issues v. Weinberger, 163 U.S.App. D.C. 368, 372, 502 F.2d 370, 374 (1974). Our goal, after all, is to vindicate Congress' intent, and we therefore must read statutory language as Congress used it — which may or may not be precisely as Webster defined it. In order to determine whether the "plain meaning" of a particular provision is in fact unreasonable in light of the totality of the statute, the statutory purpose, or the legislative intent, a court must in the first instance engage in at least some measure of further statutory analysis. See 2A Sands, Statutes and Statutory Construction §§ 46.02.04 (4th ed.) Additionally, since the terms with which we are concerned are used in the Postal Reorganization Act against the particular background of regulatory activity, and, indeed, against the even more particular background of *866 postal regulatory activity, we must discern whether such contexts imparted to the operative terms some modifying quasi-technical meaning at variance with the terms' "plain meaning." NLRB v. Coca-Cola Bottling Co., 350 U.S. 264, 76 S. Ct. 383, 100 L. Ed. 285 (1956); Barber v. Gonzales, 347 U.S. 637, 74 S. Ct. 822, 98 L. Ed. 1009 (1954).

Thus, the "plain meaning" of the statutory language raises no more than a presumption, which may be rebutted, as to the true "statutory meaning" of that language. The meaning of a federal statute is for the federal courts to decide. Enochs v. Smith, 359 F.2d 924, (5th Cir. 1966). When a court is faced with a situation in which Congress could "reasonably" have meant two different things, i. e. when one expressed statutory purpose might lead to one interpretation but a different clear purpose leads to another, and when on its own terms the statute makes a coherent whole whichever of two ways particular provisions are interpreted, a more complicated situation is present. But, in such a case, we must remember: (1) that we are a court, not a legislature, and are charged with interpreting the words that Congress used, not with choosing them; and (2) though our own normative preference might be otherwise, if the plain and ordinary meaning of the language is itself reasonable in light of the broad purposes of the statute and in light of the structure and context of the legislation as a whole, we must uphold that meaning. Arkansas Valley Industries, Inc. v. Freeman, 415 F.2d 713 (8th Cir. 1969); Sea-Land Service, Inc. v. Federal Maritime Commission, 131 U.S.App.D.C. 246, 404 F.2d 824 (1968).

Our modus procedendi in pursuing the further statutory analysis which is necessary will be to analyze a number of considerations asserted by the Postal Service to militate against application of the "plain meaning principle." First, we shall consider the Postal Service's research and development role and the congressional goal of managerial flexibility. Second, we shall address the possible analogy to research and development provisions of other federal regulating schemes. Third, we shall consider the alleged cumbersomeness and futility of APA type hearings and the alleged unsuitability of the statutory criteria governing evaluations of rate changes to experimental rates (and the question whether these are parameters of congressional intent contra subjecting experimental rates and classifications to Postal Rate Commission review). Fourth, we shall further explore the Service's argument based on 39 U.S.C. § 3661. As will be seen, when all is said and done, the plain meaning principle will survive the assault.

B. Considerations Alleged to Militate Against the Application of the Plain Meaning Principle

1. The Postal Service's Research and Development Role and the Goal of Managerial Flexibility.

The argument most vigorously pressed by the Postal Service against an inclusive reading of the statutory jurisdiction of the Postal Rate Commission — and the one that is apparently at the core of the Service's position — is based on § 403(a) of the Act. This section makes it the responsibility of the Postal Service to "plan, develop, promote, and provide adequate and efficient postal services at fair and reasonable rates and fees." 39 U.S.C. § 403(a). When coupled with § 401(10), which provides that the Postal Service shall "have all . . . powers incidental, necessary, or appropriate to the carrying on of its functions or the exercise of its specific powers," the Postal Service contends that the duty to "plan, develop, promote, and provide" efficient postal services means that the Service has the authority to conduct marketplace experiments like the Service Test Plan.

In support of its position, the Postal Service cites to the history of the Post Office in general and of the Postal Reorganization Act in particular. The Service points especially to the explicitly articulated congressional purpose of permitting—indeed mandating —increased managerial initiative and flexibility. See, e. g., Buchanan v. United States Postal Service, 508 F.2d 259, 263 (5th *867 Cir. 1975). Thus, the Senate Report accompanying the Act found it "obvious" that "postal management must now be given the unfettered authority and freedom it has been denied for years to maintain and operate an efficient service." S.Rep. No. 91-912, 91st Cong., 2d Sess. 2 (1970). The Postal Service argues that "unfettered authority and freedom" to experiment with new rates and new classes of mail is essential to fulfilling this congressional expectation. Without experimentation, it is said, the Service would risk either operational stagnation or premature implementation on a systemwide scale of innovations that might turn out to be counterproductive or, at least, not trouble-free. The Postal Service reminds us that the old Post Office was considered to have been deficient in the degree to which it engaged in service experimentation, see President's Commission on Postal Organization, Towards Postal Excellence 37 (1968) ("[N]o service experimentation program [currently] exists. As a result, though the service ideal is deeply imbedded in postal employees, they serve a market whose needs are but dimly perceived."), and that Congress clearly intended that the new Postal Service would remedy this shortcoming, see H.R.Rep. 91-1104, 91st Cong., 2d Sess. 9, reprinted in [1970] U.S.Code Cong. & Admin.News pp. 3649, 3657 ("The Postal Service is empowered to engage in research and development programs directed toward the expansion of present postal service and the development of new services responsive to the evolving needs of the United States.").

Furthermore, argues the Postal Service, even under the predecessor provisions of § 403(a) in the old postal legislation, e. g., Pub.L. No. 86-682, 74 Stat. 578 (1960), which was replaced by the Postal Reorganization Act, the Post Office was permitted to engage in marketplace experiments. Thus, the old § 504(a) of Title 39 of the United States Code provided:

The Postmaster General shall maintain in the Department a research and development program, including investigations and studies, for the purpose of introducing or improving equipment, supplies, methods, procedures, means, and devices used in the Department in order that its business may be more efficiently and economically operated.

In United Telegraph Workers v. FCC, 141 U.S.App.D.C. 190, 436 F.2d 920 (1970) the D.C. Circuit upheld the Post Office's participation, jointly with Western Union, in the Mailgram experiment. Mailgram was conceived of as a hybrid of the traditional telegram and first-class mail, between the two in speed and cost. The experimental service was to be available to Western Union subscribers in 12 selected cities for a test period of two years. It was projected that the subscribers would be able to send messages through Western Union communications network to one of the 110 post offices nationwide that would have teleprinters installed. Postal workers were to scan the printed messages for completeness and legibility and then insert them in window envelopes for placement in ordinary first class mail. The Post Office charged Western Union a handling fee; postage was the normal first-class rate.

In finding that § 504(a) of the postal statute permitted the Post Office's participation in the experiment, the D.C. Circuit relied on its earlier decision in Atchison, T. & S. F. Ry. v. Summerfield, 97 U.S.App. D.C. 203, 229 F.2d 777 (1955), cert. denied, 351 U.S. 926, 76 S. Ct. 779, 100 L. Ed. 1456 (1956). In Summerfield, the Postmaster General had instituted an experiment whereby ordinary first-class mail was to be carried by air on a space available basis, without payment of additional airmail postage. In all other respects (e. g., handling and distribution) the first-class mail procedures were unchanged. The district court recognized the Post Office's power to experiment, but held that the experiment was unduly prolonged and so not a true experiment. 128 F. Supp. 266, 274 (D.D.C.1955). The D.C. Circuit reversed, stating that "we are unable to find any prohibition of the present experimental program, while we do find statutory provisions broad enough to encompass it within their terms." 97 U.S. App.D.C. at 208, 229 F.2d at 782. However, *868 the statutory provisions to which the Summerfield court referred included apparent approval of the experiment by the Congress in an appropriations bill, Act of May 28, 1954, 68 Stat. 144, 148, see S.Rep. No. 1286, 83d Cong., 2d Sess. (1954), and the Postmaster General's authority to provide for the transportation of the mail, 52 Stat. 1027 (1938). The opinion in fact made no mention of the Postmaster General's research and development responsibilities under 63 Stat. 608 (1949), which was the direct line ancestor of the old § 504(a), on which the United Telegraph Workers court and the Postal Service here rely.

On the basis of this rather brief history of the Postal Reorganization Act and also of the Postal Service's research and development power, as subsumed within the cited cases, the Service would have us approve the Test Plan as a legitimate exercise of that power. The Service further asserts that requiring submission to the Postal Rate Commission of proposals like the Service Test Plan would thwart the goal of managerial freedom and would impede implementation of a managerially ordered undertaking designed to improve the efficiency of the Postal Service.

The reliance of the defendant Postal Service on cases that antedate the Postal Reorganization Act is misplaced and its evaluation of the Act's legislative history one-sided. For the Act produced broad changes in postal responsibility that went beyond the simple transfer of the responsibilities of Congress to the Postal Rate Commission and the responsibilities of the Post Office to the Postal Service. It is an incomplete characterization of the changes worked by the Act to assert that because the Post Office once took certain actions — e. g., initiation of experimentation — without input from Congress, the Postal Service may now take those actions without input from the Postal Rate Commission. This point was at issue in NAGCP, supra. The Postal Service there asserted that it was authorized under the Postal Reorganization Act to set fees for special services,[15] primarily because it had had the power to set such fees under prior law. The Service also relied on § 404(a)(6) of the Act, a general power provision that authorizes the Service "to provide, establish, change, or abolish special nonpostal or similar services." Yet, relying on the "plain language" of the new Act, 39 U.S.C. § 3622, and referring to the postal restructuring rendered by that Act, the D.C. Circuit rejected the Postal Services' claim of fee-setting power.

[Any] reasonable examination of the purposes of the Act discloses Congress' implicit design that the distinct functions of service provision and rate adjustment be divided between the Postal Service and the Rate Commission. The expertise of the Postal Service supposedly is in management, and its authority therefore reasonably extends to basic decisions pertaining to the provision of special, non-postal and other services. The Postal Rate Commission, however, was created specifically to oversee the ratemaking process. Its expertise is in the setting of rates and fees that are fair and equitable, and its authority therefore reasonably extends to all aspects of such decisions, including review of budget estimates, allocation of postal costs, establishment of rates for postage, and, it would seem plainly, the setting of fees for those special services which management decides should be provided.

NAGCP, supra, 186 U.S.App.D.C. at 358, 569 F.2d at 597.

We find the perspective of the NAGCP court helpful here, although the specific question that court faced was, of course, much different then the one before us. The elaborate divisions of responsibility between Congress and the Post Office that had *869 evolved over decades of congressional regulation gave way in the Postal Reorganization Act to a more predictable and principled division of responsibility. Management was vested in the Postal Service, rate and classification supervision in the Postal Rate Commission. We recognize and weigh heavily the congressional goal of greater managerial flexibility, but also recognize another congressional purpose that finds its incarnation in the Postal Rate Commission. The Commission's existence insures that an agency independent of the Postal Service will provide for public notice and hearing — input of those affected by the proposed action — and full and on the record, see 39 U.S.C. § 3624(a), consideration of pertinent factors and congressionally imposed goals before certain types of decision are made.

The bill provides that the Postal Rate Commission shall be a body fully independent of the Board of Governors and fully independent of any influence whatsoever of the Postmaster General or of members of his staff. The Commission's independence is contemplated as being complete from the other arms of the postal service, subject to no subordination within the postal service either expressed or implied. The limited circumstances under which a recommended rate decision made by the Commission may be modified by the Governors are carefully spelled out in the language of the bill.

S.Rep., supra, at 13. Congress was, after all, relinquishing the bulk of its control over the post office and it had to be sure that legitimate public interests were protected, for it would no longer be able to provide that assurance itself. Thus, the Postal Rate Commission, in addition to providing expertise on the subject of ratemaking, also was designed as a sort of sunshine mechanism to avoid undue political influence and to assure that the public is heard from and the public interest represented before rate, classification, and significant service changes are made.

It is ineluctable that the power of the Postal Rate Commission and the power of the Postal Service will intersect at some point. At that point the interests served by management freedom and public exposure and input come into potential conflict, for whenever by operation of the statute the Postal Rate Commission is interposed between the Postal Service's decision that a rate increase is needed and the implementation of that rate increase, managerial autonomy is in some measure limited. Reference to older cases like Summerfield and United Telegraph Workers is thus suspect, because there was no such thing as the Postal Rate Commission when these cases were decided. The policies and purposes represented by the Commission thus were not factors in those cases. Without that critical element in their calculus, the precedential value of those cases is sharply limited. The very existence and function of the Postal Rate Commission bespeaks a limitation on postal management's freedom. Moreover, at least some of the factors that, when a permanent rate increase is proposed, come into play and energize the involvement of the Postal Rate Commission would also appear to be relevant when the issue is marketplace experimentation — e. g., possible managerial favoritism, political or otherwise, avoidable and harsh impact on that part of the private sector that competes with the Postal Service, etc. Thus, although requiring that the Postal Service go to the Postal Rate Commission with marketplace experiments might impede one congressional purpose, it would promote another.

Mere invocation of the need for and purpose of managerial discretion in experimentation, therefore, cannot prevail, for Congress has unmistakably put limits on that discretion. We are inclined, as a normative matter, to agree with the Postal Service that the point of limitation on managerial discretion should not be reached at the point of implementation of the Service Test Plan. A limited experimental change is not likely to have the same impact on the public as a whole or even one segment of the public (e. g. parcel shippers) that a permanent systemwide change would have. Our task, however, is, by reference to the Act, to locate the limits on managerial discretion *870 and not, by reference to our own preferences, to create them.

The Postal Service would resolve the tension between the dual purposes of managerial discretion to experiment and public interest supervision of rates by saying that when a rate change is experimental, it is no longer a rate change for purposes of Commission review. Plaintiffs, on the other hand, argue that when an experiment involves a rate change it is no longer an experiment for purposes of unreviewed managerial discretion. The question, under the Act, is on which side of the line between experiment and rate this hybrid called an experimental rate belongs for purposes of Commission review.

The Postal Service's position in effect gives statutory primacy to § 403, which articulates the Service's duty to plan, develop, promote, and provide postal services. It is this section that most embodies the principles of management's authority to experiment, and the Postal Service reads it to mean that when rate or classification changes are still in the planning or developmental stage, even if such changes have been made experimentally, Rate Commission involvement is not required under the Act. We observe in this regard that § 504(a) of the old title 39, the research and development provision on which, together with its antecedents, the United Telegraph Workers (and putatively the Summerfield court, but see discussion of Summerfield supra) relied to uphold marketplace experimentation, was specifically rescinded by the Postal Reorganization Act. See H.R.Rep. 91-1104, 91st Cong., 2nd Sess., reprinted in [1970] U.S.Code Cong. & Admin.News pp. 3649, 3697. Thus, although the general subject matter of the new § 403(a) and the old § 504(a) overlap to some extent, § 504(a)'s apparent grant of authority to conduct marketplace experiments cannot be said to have survived the change, for § 504(a) itself did not.

Furthermore, even if we were to assume that United Telegraph Workers and Summerfield somehow survived the repeal of the statutory authorization for marketplace experimentation on which they relied, it is at least arguable that those older cases did not involve rate or classification matters at all, which is the particular focus of this litigation, but rather dealt with service changes. The Postal Service apparently so concedes, at least as to the Summerfield type of experiment. See Reply of the United States Postal Service to Plaintiffs' Trial Memorandum 5. Service charges are now governed by 39 U.S.C. § 3661, and Rate Commission involvement in such changes is explicitly restricted to cases of meaningful impact on a nationwide or substantially nationwide scale. No such restrictions appear in §§ 3622 and 3623. See Buchanan v. United States Postal Service, 508 F.2d 259 (5th Cir. 1975); text accompanying notes 24-25 infra.

The Postal Service's argument, then, can only be that the general responsibility to plan, develop, promote, and provide postal services by itself imposes a limitation on Rate Commission jurisdiction to review rate and classification changes that is not apparent from the sections that establish that jurisdiction — §§ 3622 and 3623. The ordinary rule of statutory construction, however, is that a general duties provision like § 403(a) may not override specific provisions like §§ 3622 and 3623. NAGCP, supra, 186 U.S.App.D.C. at 358, 569 F.2d at 597. Indeed, the Act itself demonstrates that § 403(a) does not operate as a brake on Commission jurisdiction for, as we have noted, the duty imposed by § 403(a) to provide postal services is clearly subject to Commission jurisdiction under §§ 3622 and 3623 if providing the services involves rate or classification changes. Otherwise, the comprehensive and elaborate ratemaking and classification scheme set out in the Act would be meaningless.

The Postal Service does not seem to quarrel with the proposition that providing services, at changed rates and classifications, implicates Commission review. Yet the Service offers no convincing explanation why, under the Service's reading of § 403(a), the duty to provide postal services is subject to Commission jurisdiction while *871 the duties to plan, develop, and promote postal services are not. Moreover, we are hard-pressed to understand why, as to the participants in the Service Test Plan, services are not in fact being provided at changed rates and under a new classification. It may be that the services are being provided on an experimental basis, but, once again, the Act does not on its face exclude experiments from Commission jurisdiction. In the absence of such an exclusion, internally consistent application of the terms of the Act demands that when rates or classifications are implicated in the marketplace, Commission review under §§ 3622 and 3623 is triggered.[16]

2. Analogy to Research and Development Provisions of Other Federal Regulatory Schemes

At the conclusion of the final hearing, we invited the parties to augment their original briefs with in depth analysis of the teachings of the cases dealing with other federal regulatory systems. They have done so, though with equivocal results.

Looking beyond the perimeters of the Postal Reorganization Act, the Postal Service argues that the more general regulatory context imposes a gloss on the statutory terminology that alters what we have characterized as the "plain meaning" of the rate sections of the statute. The Service cites cases involving the Federal Communications Commission, the Civil Aeronautics Board, and the Interstate Commerce Commission. See Network Project v. FCC, 167 U.S.App.D.C. 220, 511 F.2d 786 (1975); Saturn Airways, Inc. v. CAB, 157 U.S.App.D.C. 281, 483 F.2d 1284 (1971); Delta Air Lines, Inc. v. CAB, 147 U.S.App.D.C. 272, 455 F.2d 1340 (1971); United Telegraph Workers v. FCC, 141 U.S.App.D.C. 190, 436 F.2d 920 (1970); Comm. Against Pay TV v. FCC, 112 U.S.App.D.C. 248, 301 F.2d 835, cert. denied, 371 U.S. 816, 83 S. Ct. 28, 9 L. Ed. 2d 57 (1962); General Increases — Eastern Central Territory, 316 I.C.C. 467 (1962); River Terminals Corp. Class and Commodity Rates, 14 M.C.C. 542 (1939).[17] The Postal Service reads these cases to stand for the proposition that when federal agencies have research and development responsibilities, the agencies will be upheld in the reasonable exercise of those responsibilities, even when that exercise takes place in the marketplace. Since the Postal Service has such a responsibility under 39 U.S.C. § 403(a), and since the experiment at issue is demonstrably reasonable, the conclusion urged upon us is that on the strength of the cited cases we should uphold this reasonable marketplace exercise of postal research and development.

Curiously, plaintiffs cite the very same case to support their position. They maintain that reference to the general regulatory framework as represented by these cases establishes that in the ordinary regulatory context Congress and the courts understand general regulatory ratemaking authority to include authority over experimental rates charged to members of the public. Therefore, urge the plaintiffs, any "technical" connotations assumed by the Postal Reorganization Act's ratemaking language by virtue of prior usage in the regulatory context militate in favor of the statute's "plain meaning," not against it.

*872 The fact that both parties cited the same cases to support opposite conclusions raises initial suspicions about the degree to which the cases might be dispositive in this litigation. Upon consideration of the citations, we do not think that the cases are particularly helpful to either side, although they may offer some small measure of support for plaintiffs. We will not set out and analyze each of the cited cases, for they are sufficiently similar in general legal analysis and conclusions to warrant treatment by example. We therefore will address directly only United Telegraph Workers v. FCC, supra, which is, on its facts, perhaps the closest to the instant case of the cases cited. Our comments, however, are, unless otherwise indicated, generally applicable to these cases as a whole.

United Telegraph Workers was the Mailgram case. Western Union, before instituting the program, filed with the FCC a tariff for the experiment. The tariff was approved without hearing. In reviewing the FCC's decision, a critical issue before the D.C. Court of Appeals was "whether approval of experimental Mailgram is beyond the Commission's authority." 141 U.S.App.D.C. at 193, 436 F.2d at 923. Answering this question in the negative, the court did not mention the generalized FCC authority over rates for common carrier services offered to the public, although the general requirement that new tariffs be filed with the agency, see 47 U.S.C. §§ 203-204, was apparently the reason the issue was before the FCC in the first place, 141 U.S.App.D.C. at 191, 436 F.2d at 921. Rather, the court cited broad agency responsibility to provide for a rapid and efficient communications system, 47 U.S.C. § 151, wide agency discretion and authority in effectuating its responsibilities, id. § 154(i), and the agency's duty to keep itself informed "as to technical developments and improvements in wire and radio communication and radio transmission of energy to the end that the benefits of new inventions and developments may be made available to the people of the United States," id. § 218.

The Postal Service reads all this to mean that just as the FCC could authorize and supervise reasonable experiments, so too can the Postal Service. Even assuming, however, the reasonableness of the postal experiment at issue in the case sub judice, the Postal Service's argument is fatally flawed in analogizing the Postal Service to the FCC and other agencies of that genre. Such agencies are regulatory agencies, whereas the Postal Service is, under the supervision Postal Rate Commission, analogous to a regulated agency. It is true, as the Postal Service points out, that like the FCC, the CAB, or the ICC, the "regulated" in this case — the Postal Service — is an independent agency within the executive branch of the federal government, whereas the regulated entity in the cited cases were members of private industry. This governmental/private enterprise configuration, however, is hardly sufficient to justify inverting the regulatory roles. Whatever responsibilities for research and development agencies like the FCC possess, those duties are cast in the larger context of a supervision by the regulator of marketplace experimentation by the regulated. The attempt of the Postal Service to analogize the power of the regulator in United Telegraph Workers to the asserted power of the regulated in this case ignores this larger context, and so must fail.

United Parcel Service, of course, analogizes the FCC in United Telegraph Workers to the Postal Rate Commission rather than to the Postal Service. Proceeding from this analogical starting point, plaintiffs draw from United Telegraph Workers and the cases of which it is representative the lesson that regulatory authority to supervise ratemaking in general includes the authority to supervise experimental ratemaking in particular. Otherwise put, plaintiffs contend that the case teaches that when circumstances are such that the regulated would have to go to the agency for approval of permanent rate or service changes, the regulated must also do so when the proposed change is experimental in nature, and that as a general rule a regulatory agency's jurisdiction does not depend on whether the proposal of the regulated entity is for permanent *873 or for experimental tariffs. If this were not the case, say the plaintiffs, the Western Union experimenters would not have felt compelled to go to the FCC in the first place, but instead, like the Postal Service in the instant case, would have resolutely proceeded without so much as a nod to the regulatory agency.

As we earlier observed, the United Telegraph Workers opinion did not specifically address the discrete question of the general scope of FCC authority over rates; that is, it did not speak of the power of rate supervision independently of the FCC's other broad statutory duties and powers. On the face of the court's language, it would appear that the court found FCC "authority" to rule on experimental rates in the FCC's broad development responsibilities. Such a reading of the case, of course, would be of no help to UPS, for the Postal Rate Commission, to which UPS would liken the FCC, has no "development" responsibilities. The responsibilities of the Postal Rate Commission are strictly confined to relatively passive review of rate, classification, and major service changes, unadorned by the overlay of broad FCC-esque responsibility for industry guidance and of wide discretion in choosing the appropriate manner and means of pursuing its statutory mandate. To the extent that the United Telegraph Workers court relied on such broader powers to sustain the FCC's exercise of jurisdiction over experimental rates, its opinion is not applicable to the situation before us.

Plaintiffs recognize the silence of United Telegraph Workers on the issue of the reach of unembellished supervisory authority over rates, but, as we read their brief, argue that this silence only indicates that the court assumed, without the need for discussion, that FCC jurisdiction over rates in general included FCC jurisdiction over experimental rates. Such jurisdiction is so obvious, the argument runs, that just as Congress saw no need to single such rates out in the Postal Reorganization Act, the United Telegraph Workers court saw no need to dwell on the issue of FCC ratemaking authority. Plaintiffs also contend that the court's use of the word "authority" is misleading. When the court purports to find FCC "authority" over the experimental Mailgram tariff in the agency's broad promotion and development responsibilities, we are urged, the court is not, in fact, speaking of authority in a jurisdictional sense. In addressing FCC "authority" under the agency's broad development mandate, UPS submits, the court was not addressing the existence of authority — jurisdiction — but the exercise of authority — discretion. Thus, the "authority" sections of the opinion are read to mean that the broad industry-development responsibilities of the FCC gave the agency the discretionary authority to take the particular actions that it did in the procedural manner that it did. Plaintiffs argue that the reason the issue was before the agency in the first place, however — the reason FCC jurisdiction was invoked — was that a new, if experimental, tariff was being proposed by a member of the regulated industry. In like manner, urges UPS, Postal Rate Commission jurisdiction — which is mandatory under the Postal Reorganization Act — is implicated when the Postal Service proposes to test a new mail classification by charging new rates in the marketplace.

Plaintiffs reading of the case is certainly not implausible. Indeed, it may be more plausible than the alternative — that the court considered FCC jurisdiction over experimental programs and rates to depend on the agency's broad and developmental statutory responsibilities. It seems to us, however, that the United Telegraph Workers court did not analytically compartmentalize the two questions of existence and exercise of authority as discretely as have plaintiffs. Rather, the pervasive regulatory responsibilities of the FCC seem, in the court's approach, to have coalesced to some extent with the agency's jurisdictional boundaries. As a result, although the case probably provides some marginal support for plaintiffs, we do not think that United Telegraph Workers bears the weight that United Parcel seeks to put on it. It surely cannot be gainsaid that in United Telegraph Workers an agency with general supervisory *874 ratemaking authority was affirmed in its exercise of authority over experimental rates. Nevertheless, we do not read the case to stand unequivocally for the proposition that regulatory jurisdiction over permanent tariffs embraces jurisdiction over experimental tariffs such that when, by statute, a permanent tariff must be filed with the regulatory agency, so too must an experimental tariff be filed.

The other cases cited seem to us to suffer from similar ambiguity in locating the statutory ground in which agency authority over experimental tariffs and services finds root. Even so, plaintiffs argue that the non-FCC cases in their arsenal of citations are stronger support for their position because the mandate to the FCC to secure the benefits of new inventions and developments for communications users, 47 U.S.C. § 218, one of the primary factors upon which the United Telegraph Workers court relied, and which also therefore colored the opinions in Network Project v. FCC, 167 U.S.App.D.C. 220, 511 F.2d 786 (1975) and Connecticut Comm. Against Pay TV v. FCC, 112 U.S.App.D.C. 248, 301 F.2d 385, cert. denied, 371 U.S. 816, 83 S. Ct. 28, 9 L. Ed. 2d 57 (1962), does not exist in the statutory charters of the other agencies to which plaintiffs' citations refer us — the CAB and the ICC. Plaintiffs contend that these agencies, insofar as they lack the specific responsibility of promoting and directing experimentation in the regulated industry, provide a much closer parallel to the Postal Rate Commission and that when the exercise of authority over experimental tariffs or services by the ICC or CAB is upheld, the presumption that control over experimental tariff regulation flows directly from control over general tariff regulation is much stronger.

It is our perception, however, that the FCC, the CAB, the ICC, and other comparable regulatory bodies are really different types of agencies from the Postal Rate Commission. Nor do we think that the difference between them can be distilled in one or two specific provisions of the agencies' statutory charters such that the presence of absence of a particular provision is determinative. A reading of the statutes relevant to agencies like the FCC, CAB, and ICC, 47 U.S.C. § 151 et seq. (FCC); 49 U.S.C. § 1301 et seq. (CAB); 49 U.S.C. § 1 et seq. (ICC), reveals a statutory gestalt according to which the regulatory agency's control over and involvement in the affairs of the regulated industry are much more pervasive and sweeping than are the involvement and control exercised by the Postal Rate Commission. We do not find this difference surprising, for when the regulated industry is essentially private, primary governmental guidance of the industry in order to promote congressionally defined or intimated public policy goals and to develop the industry along paths perceived to be beneficial to the nation as a whole conventionally comes through the regulatory agency. In the case of the Postal Service, however, primary control is exerted directly by Congress, since the Postal Service is essentially a modified government corporation. Developmental goals and instructions could be and were included within the Postal Service's statutory charter, and the mandate to "plan, develop, promote, and provide adequate and efficient postal services," 39 U.S.C. § 403(a), is given directly to the Postal Service.[18]

Investigation, supervision, and review of publically beneficial research and development of new programs and services offered to the public was, at least to some extent, statutorily charged to the regulatory agency in United Telegraph Workers and in the other of the cited cases. In the postal context, however, these functions are charged directly to the regulated — the Postal Service — not to the regulator — the Postal Rate *875 Commission. The ultimate issue we face is whether, for purposes of the Postal Reorganization Act's internal jurisdictional structure, experimental rates and classifications are primarily experiments, for which the Postal Rate Commission apparently has no primary responsibility, or, conversely, are primarily rates, which the Postal Service may not institute without input from the Postal Rate Commission. That is, does the fact that a rate or classification is offered to a limited number of mailers only and is experimental in nature make these things not rates or classifications within the meaning of the Act? The cases cited by plaintiffs and defendants are of relatively little help in this regard, for in them the courts found situations in which the regulatory agency had at least some significant measure of stated regulatory jurisdiction over both marketplace experiments and general rates. In such a situation, it was not necessary to define the scope of authority granted by either of these components in isolation. Insofar as the courts were not presented with regulatory schemes such as the one we confront, we are not surprised that they do not provide unambiguous answers to the questions before us.

Thus, the Postal Service's proposed application of the cases like United Telegraph Workers, whereby the Postal Service is analogized to the FCC because they both have research and development powers, suffers from a defect similar to that suffered by plaintiffs' citation of the same cases. Though not an absolute dichotomy, the two responsibilities of supervision of industry development and review of tariffs — both of which are responsibilities of agencies, like the FCC — seem in large measure to be split between the Postal Service and the Postal Rate Commission with research and development primarily the realm of the Postal Service and rate review the initial domain of the Postal Rate Commission. The dissimilarities already noted between the FCC and the Postal Rate Commission thus find their correlates in complementary dissimilarities between the FCC and the Postal Service (these latter dissimilarities being magnified by the fact that the Postal Service is not a regulatory agency at all). Uncritical application of these cases, in the fashion urged upon us, would put us somewhat in the position of the blind man examining the elephant. The proponents of such an approach err, in that they selectively focus on part of the typical regulatory framework and, finding that reflected in part of the postal framework, then, depending on the proponent, conclude that either the Postal Service or the Postal Rate Commission is just like the FCC, the CAB, and the ICC. Finding ourselves on too uncertain ground when we attempt to wrench the holdings of the cited cases from their own context and impose them on ours, we decline to do so.

The result dictated by the "plain meaning" principle has thus survived the Postal Service's invocation of the alleged research-development exception and of precedent from other regulatory fields. We turn now to consideration of still another factor alleged by the Postal Service to militate against application of the plain meaning principle, this one emanating from the textual fabric of the Postal Reorganization Act.

3. The Alleged Cumbersomeness and Futility of APA-Type Hearings and the Alleged Unsuitability of the Statutory Criteria Governing Evaluation of Rate Changes to Experimental Rates (Barometers of the Congressional Intent?)

The Postal Service's next argument is, in effect, that Congress could not have intended that experimental rates be subjected to Rate Commission review because of the alleged cumbersomeness and futility of APA-type hearings and the alleged unsuitability to experimental rates of the statutory criteria governing evaluation of rate changes. Before addressing these contentions, we state them more fully.

The Postal Service first argues that, regardless of the general regulatory structure of other agencies, the very provisions that establish and regulate Postal Rate Commission review of rates and classifications, *876 §§ 3622-3624, establish their own inapplicability to experimental rates and classifications. Under 39 U.S.C. § 3624(a), the decisions of the Postal Rate Commission on rate and classification matters may be made only after evidentiary hearings conducted in conformity with the Administrative Procedure Act, 5 U.S.C. §§ 556 & 557. That Act sets out specific criteria that the Commission is to consider in evaluating a proposed rate change, 39 U.S.C. § 3622(b),[19] or classification change, id. § 3623(c).[20] The Postal Service argues that these criteria only make sense in the context of permanent and systemwide rate or classification changes, and make limited experiments unsuitable for Commission review, thereby demonstrating the congressional intent not to subject experiments to the hearing requirements.[21]

*877 Turning to the cumbersomeness contention, the Postal Service concedes that, despite the presumed management prerogatives of the leaders of private industry, rate supervising agencies often review and rule on proposed marketplace experiments undertaken by the regulated industry. See Part II.B.2 supra. However, the Service observes that when these other agencies in other regulatory areas rule on marketplace experiments by members of the regulated industry, the agencies generally exercise the option of ruling on the experiment without formal adjudicative proceedings. See, e. g., Network Project v. FCC, 167 U.S.App.D.C. 220, 511 F.2d 786 (1975); Saturn Airways, Inc. v. CAB, 157 U.S.App.D.C. 281, 483 F.2d 1284 (1971); Delta Air Lines, Inc. v. CAB, 147 U.S.App.D.C. 272, 455 F.2d 1340 (1971); American Airlines Inc. v. CAB, 123 U.S.App.D.C. 310, 359 F.2d 624, cert. denied, 385 U.S. 843, 87 S. Ct. 73, 17 L. Ed. 2d 75 (1966). The Postal Service notes that the Postal Rate Commission is, unlike other agencies, without this discretion to dispense with the requirement of an evidentiary hearing. The Service sees this as an indication that Congress, familiar with the normal procedures, never intended the Postal Rate Commission to get involved in experimental, as opposed to permanent rates. Congress, continues the argument, would not have intended proposed experiments, generally handled on an expedited basis by other agencies, to get bogged down in the mires of evidentiary hearings. Additionally, the Postal Service highlights language in the foregoing cases that suggests that evidentiary hearings are not fruitful where experiments rather than permanent tariffs are at issue. Network Project v. FCC, 167 U.S.App.D.C. 220, 230-31, 511 F.2d 786, 796-97 (1975); Saturn Airways, Inc. v. CAB, 157 U.S.App.D.C. 281, 287, 483 F.2d 1284, 1291 (1971); Delta Air Lines, Inc. v. CAB, 147 U.S.App.D.C. 272, 276, 455 F.2d 1340, 1344 (1971); United Telegraph Workers v. FCC, 141 U.S.App.D.C. 190, 194-195, 436 F.2d 920, 924-925. An evidentiary hearing, the argument runs, is futile before an experiment is conducted because the experiment is designed to gather the very evidence that is to be presented at the hearings. Without the experiment, in other words, there is no evidence to hold an evidentiary hearing with. We now address these various contentions.[22]

*878 We observe, first, that the postal context differs in significant ways from the regulatory contexts dealt with in the cases cited by the Postal Service. The Postal Rate Commission has but one entity to regulate, not a multitude of smaller ones. The bureaucratic burden of holding hearings on marketplace experiments is therefore not likely to be as great as a similar burden would be if imposed on other agencies. Therefore, we reasonably believe (as could the Congress) that delays can be minimized. Further, because in the postal area the entire regulated industry is represented by one managerial unit and also because of the sheer size and economic importance of the Postal Service, the potential economic havoc that the single unit could wreak with an imprudently structured experiment is enormous. Most critically, even if no evidentiary hearings were in fact held in the cases cited by defendant, the experimental tariffs were at least filed with the regulating agency, and opportunity for comment to the agency by adversely affected parties was present. The agency and those whose interests they protect had advance warning and a chance to consider the implications of the proposal. If circumstances warranted, the agencies could have suspended the rate and held hearings; at least there was provision for some preliminary evaluation. In our case, unless the Postal Service goes to the Postal Rate Commission under § 3622 or § 3623, there is no review at all and no opportunity for notice or response, abbreviated or otherwise. The Postal Service's position allows the goal of management discretion and initiative to swallow up the goal of exposure and review in the public interest of proposed actions that impact on the public. The Postal Service's own cited cases reveal that even when evidentiary hearings are not held for planned marketplace experiments, the regulated must at least check in with the agency before going out and experimenting in the marketplace; there is some sort of review. We do not think it unreasonable, in a regulatory context where review is either all or nothing, to think that Congress intended all.

Secondly, we are not convinced that the hearing requirement is as inherently inflexible and inappropriate as the Postal Service would have us believe. Although most of the factors the Postal Rate Commission is to consider in reviewing proposed rate or classification changes, 39 U.S.C. §§ 3622(b), 3623(c), do seem out of place when applied to experiments, the review criteria are explicitly elastic, with the Rate Commission instructed to consider in general "the policies of this title and" in particular "such other factors as the Commission deems appropriate." Id. §§ 3622(b)(9), 3623(c)(6) (emphasis added). Thus the statutory criteria are not limitations on the Commission's review authority. Moreover, considerations not "appropriate" to a permanent rate change may be "appropriate" to an experiment. In any case, at least one of the specific factors relevant to rate changes is directly responsive to the competitive impact concerns expressed by plaintiffs in this litigation, for the Commission's attention is directed to "the effect of rate increases upon the general public, business mail users, and enterprises in the private sector of the economy engaged in the delivery of mail matters other than letters . . .." Id. § 3622(b)(4). The Rate Commission, therefore, would not be bound to an exercise in futility in reviewing a proposed experimental rate.

We realize, of course, that the thrust of the Postal Service's argument goes not to feasibility but to intent. That is, even assuming that the Commission could develop and apply meaningful criteria appropriate to a proposed experiment, the general inapplicability *879 of the statutory review criteria is said to be indicative of the fact that Congress never intended experiments to be reviewed under these sections. The "elastic clause" in §§ 3622 and 3623, the Postal Service might continue, signifies only Congress' recognition that it may not have thought of every factor relevant to a permanent rate or classification change, not that the sections were designed to cover experimental changes.

This is, we think, a rather wooden approach to statutory construction. It allows statutes little vitality to cover more than the paradigm. Realistically, it is entirely possible that Congress did not explicitly consider the problem of marketplace experiments at all. Neither our research nor that of the parties has uncovered any mention of such experiments in the legislative history. Legislators do not always think of everything; that is why catch-all provisions like 3622(b)(9) and 3623(c)(6) are included in statutes. Our task in such an unprovided for case is to determine how the structure and purpose of the Act applies to the situation. Our modus procedendi is to begin with the statutory language and then to measure that language against what we know of the congressional intent.

Nor do we think that the requirement of APA-style evidentiary hearings forces the Rate Commission into a Procrustean bed unsuited to experimental rate review. We must presume that the Postal Service's decision to conduct a particular experiment with particular conditions was not the product of a fit of whimsy but was based on some objective information and guided toward some articulable goal. Obviously, the Commission could not evaluate the evidence the experiment was designed to generate, but it could review the information on which the initial decision to experiment was based. An evidentiary hearing need not be a meaningless or futile exercise merely because the subject matter is an experimental rate; the policies of the Act can still provide meaningful standards. Further, the means for applying those criteria need not follow the ossified and interminable format the Postal Service fears. Both the Postal Reorganization Act, 39 U.S.C. § 3624(b) and the Administrative Procedure Act, 5 U.S.C. § 556(c)(5) & (6) contemplate streamlined proceedings to achieve "utmost expedition consistent with procedural fairness to the parties," 39 U.S.C. § 3624(b). Consistently with the APA, the Commission could, and we trust would, tailor the scope and conduct of its hearings considerably when the issue before it is not a nationwide and permanent rate change but a limited and temporary experimental rate change. Indeed, we strongly urge the Commission to formulate and adopt such procedures.[23]

It is not for us, however, to articulate the evidentiary factors the Commission might consider and to erect the procedural structure in which it would conduct its review. Our role is confined to determining whether, consistently with the statutory purposes, Congress might reasonably have intended, as the plain language suggests, that the Postal Service go to the Postal Rate Commission before any rate or mail classifications are changed, experimentally or otherwise. We are satisfied that the requirements of evidentiary hearings as set out in the Postal Reorganization Act could reasonably and consistently with statutory purpose be applied to marketplace experiments.

4. Significance of 39 U.S.C. § 3661

Finally, the Postal Service observes that under 39 U.S.C. § 3661, Postal Rate Commission involvement in service changes is triggered only when the Postal Service plans "a change in the nature of postal services which will generally affect service *880 on a nationwide or substantially nationwide basis . . .." 39 U.S.C. § 3661(b). Construing this section, the Fifth Circuit has held that, when the issue is changes in postal services, the Act does not require an advisory opinion from the Postal Rate Commission before implementation of the change unless (1) there has been a "change" in a quantitative sense, (2) there has been a meaningful impact on service, and (3) the change affects service on a nationwide or substantially nationwide basis. Buchanan v. United States Postal Service, 508 F.2d 259 (5th Cir. 1975). See also Wilson v. United States Postal Service, 441 F. Supp. 803 (C.D.Calif.1977). The Postal Service apparently contends that these requirements are also applicable to rate and classification changes under §§ 3622 and 3623 in determining whether Rate Commission intervention is mandated. Reply of the United States Postal Service to Plaintiff's Trial Memorandum, 6, 8-16. The Postal Service argues that these requirements are not met in the case of the Service Test Plan and that therefore the Commission need not be consulted.

The basis on which the Postal Service apparently transposes the requirements of § 3661 to §§ 3622 & 3623 is not clear, but presumably the point is that § 3661 provides insight into the congressional conception of the role of the Postal Rate Commission. It is only in the major matters — i.e., matters that have nationwide or substantially nationwide impact —[24] the argument continues, that Congress envisioned Rate Commission involvement. The Service's apparent contention is that what Congress made explicit in § 3661 is implicit in §§ 3622 & 3623, and further that, since an experiment of limited scope and temporary duration is not a change of significance on a nationwide scale, Rate Commission review is not mandated.

We have addressed this contention in Part II.A. supra. As that discussion makes clear, we do not find this argument convincing; indeed, the citation to § 3661 seems to cut in favor of the plaintiffs rather than the Postal Service. If Congress specifically limited Commission involvement in service changes to "nationwide or substantially nationwide" changes, it is apparent that it knew how to circumscribe Commission jurisdiction. Since Congress did not also do so for rate or classification changes, we might as well be justified in taking the different formulations to indicate affirmatively that marketplace experiments involving rate or classification changes are grist for the Commission's mill, whether or not the changes have nationwide or substantially nationwide effect.[25]

III. Conclusion

The end of our inquiry brings us back to where we began — the plain meaning of the statutory language. The Postal Service asks us to hold that a bona fide experiment, although it may in fact involve at least temporarily charging a limited number of shippers new rates, does not constitute a rate change within the meaning of the Postal Reorganization Act. In essence, the argument is that a change in rates under the statute is only a systemwide change; merely because twenty or so shippers' rates *881 have been changed does not mean that there has been a statutory rate change so as to trigger mandatory Rate Commission review. The Postal Service has argued that legislative intent, postal history, the structure of the Act, and broad regulatory context all point to excluding experimental rate changes from the requirements of notice and hearing.

We confess that the Postal Service's argument is not unappealing. The statute is not — as statutes seldom are — crystal clear, and the legislative purposes seem to meet at cross-currents. In many ways it might well be preferable to permit some measure of marketplace experimentation to the Postal Service without imposing the rigors of Commission review. Perhaps, under some abstract "reasonableness" test the Postal Service would prevail; perhaps, divorced from the statutory framework, we could be convinced that the Postal Service's allocation of responsibility is the more reasonable.

The words of the statute, however, provide the framework for our musings on reasonableness, and those words are inclusive, unmodified, and mandatory. See 39 U.S.C. §§ 3622(a) & 3623(b). Compare 39 U.S.C. § 3661(b). However we align the arguments and considerations, we cannot avoid the fact that the postage rates and mail classifications applicable to the participants in the Service Test Plan have been changed from their previous configuration. An interpretation based on the plain language of the provision at issue does not distinguish between various types of rates and mail classifications. Such an interpretation is not inconsistent with the purposes of the Act or with the structure of the Act as a whole. Although we have explored at great length a number of considerations alleged to militate against application of the plain meaning principle, they have failed to dent its armor.

Moreover, interpretation based on plain meaning does, on its own terms, make sense and positively promote some of the statutory purposes. Faced with plain language that is reasonable in light of the rest of the statute and the legislative intent, the result is clear. When the Postal Service seeks to experiment with the rates postal patrons pay for sending their mail through the postal system, it must first request a recommended decision from the Postal Rate Commission. The permanent injunction requested by the plaintiffs accordingly must issue.

NOTES

[1] Postal rates and mail classifications are distinct concepts in the statutory scheme. 39 U.S.C. §§ 3622 & 3623.

[2] Illustrative of this problem is the following exchange in 1967 between then-Postmaster General Lawrence O'Brien and the Chairman of the House Post Office Appropriations Subcommittee:

MR. STEED: General . . . would this be a fair summary: that at the present time, as the manager of the Post Office Department, you have no control over your workload, you have no control over the pay rates of the employees that you employ, you have very little control over the conditions of the service of these employees, you have virtually no control, by the nature of it, of your physical facilities, and you have only a limited control, at best, over the transportation facilities that you are compelled to use — all of which adds up to a staggering amount of "no control" in terms of the duties you have to perform.

MR. O'BRIEN: Mr. Chairman, I would have to generally agree with your premise. . . . that is a staggering list of "no control." I don't know [whether] it has ever been put that succinctly to me. If it had been at an appropriate time, perhaps I wouldn't be sitting here.

H.R.Rep. 91-912, 91st Cong., 2d Sess. 3 (1970).

[3] Section 3641 of the Act, providing for temporary changes in rates and classes of mail, permits the Postal Service to institute a temporary rate if the Postal Rate Commission fails to transmit a recommended decision in accordance with § 3624(c). 39 U.S.C. § 3641(a). Section 3624(c) requires the Commission to transmit its recommended decision within ten months after receiving the request for such a decision from the Postal Service, except that if the Commission's failure to comply is caused by the Postal Service's delay in responding to any lawful order of the Commission, the ten month period is extended by the time of the Postal Service's noncompliance. The temporary rate instituted under § 3641 is to be in accordance with the rate proposal under consideration by the Commission, id. § 3641(c), and must terminate no later than 150 days after the Commission renders its recommended decision, id. § 3641(d). Section 3641(e) permits the Postal Service to institute a temporary change in mail classification if a recommended decision is not forthcoming from the Commission within 90 days of submission of a request by the Postal Service. Such a temporary classification change must be in accordance with the proposal made to the Commission and must terminate no later than 30 days after the Commission transmits its recommended decision.

The procedures for temporary rate changes were established as stated above in 1976 by Pub.L. 94-421, 90 Stat. 1303 (1976). The original 1970 Act had permitted temporary rate increases to go into effect 90 days after submission to the Commission, but such increases were not to exceed 33% of the existing permanent rates. Further, they expired 30 days after the Commission rendered its decision. There was no statutory provision requiring the Commission to render a decision within a specified period of time. Temporary classification changes were unaffected by the 1976 amendment.

[4] Shippers were selected from among the Postal Service's regular parcel post patrons whose mailing patterns made them likely prospects for the service represented by the test plan.

[5] The metropolitan areas involved are: Seattle, Washington; Dallas-Fort Worth, Texas; Omaha, Nebraska; Boston, Massachusetts; and Washington, D.C. One test participant is in fact located in Baltimore, Maryland, but, in accordance with past practice, mails its parcels at the Washington, D.C. Bulk Mail Center.

[6] The Postal Service's Bulk Mail System came into full operation in 1976 in an effort to automate and expedite the mailing of parcels. It consists of 21 Bulk Mail Centers and 9 auxiliary facilities. As a result, postal operations involving parcel post and third-class parcels are now organized around the Bulk Mail System facilities, rather than around standard post offices and the general mail network. Packages mailed through the Bulk Mail System must meet certain packaging, size, weight, shape, and content restrictions. Basically, these limitations seek to ensure that the packages are "machineable" by standardizing to some extent parcels mailed through the System. A minimum of 300 pieces per mailing must be tendered in order to qualify for bulk parcel post. Rates are based on the average weight per piece of all parcels in the mailing and vary with distance.

[7] Distance is limited to the extent that the delivery destination of the parcel cannot be outside the geographic area covered by the Postal Service Bulk Mail Facility that serves the shipper. Thus, the Boston Bulk Mail Center service area is Maine, Vermont, New Hampshire, Massachusetts, Rhode Island, Connecticut, and a part of upper New York; packages mailed from Boston as part of the test plan are limited to that area. As a further example, the Dallas-Fort Worth Bulk Mail Center service area is most of Texas and parts of Louisiana, Arkansas, Oklahoma, and Kansas.

Within such service areas, which cover five parcel post zones — a local zone and four numbered zones — a five-pound package mailed under normal parcel post would cost from $.91 to $1.36, depending on distance; under the test plan, packages averaging five pounds would each cost $.87, regardless of distance within the service area. Similarly, a twenty-pound package would regularly cost anywhere from $1.58 to $2.93; the test plan standardizes the cost at $1.15. Comparable rates for parcels shipped via plaintiff United Parcel vary by distance from $1.05 to $1.28 for five-pound parcels and from $2.10 to $3.02 for twenty-pound packages.

A further limitation on distance under the current plan is that regardless of the size of the service area, no more than 25% of the test volume may be delivered beyond parcel post zone number two, which is approximately 150 miles from the parcel's point of origin.

[8] As an additional aspect of the test, a service feature known as Day Certain Delivery, not being generally promoted by the Postal Service during the experiment, is being promoted to the shippers participating in the test.

[9] Although plaintiffs, the United Parcel Service companies, now have only limited contract carrier rights to deliver parcels for designated retail stores, U.P.S. currently has a proposal before the Interstate Commerce Commission that, if approved, would grant U.P.S. the unlimited right to deliver for retail stores as a common carrier. N.T., Dec. 30, 1977, at 9. The service presaged by the Postal Service's experiment would also be aimed in large measure at retailers. See Stipulated Exhibits 4, 5. The test itself however, has relatively little competitive impact on U.P.S. since distance of delivery and participant limitations have been imposed on the experiment by stipulation of the parties. See N.T., Nov. 21, 1977, at 3-4, 5.

[10] Plaintiffs originally asserted that the Postal Service's experiment violates both the mandate of 39 U.S.C. § 403(c) that "[i]n providing services and in establishing classifications, rates, and fees under this title, the Postal Service shall not . . . make any undue or unreasonable discrimination among users of the mails, nor shall it grant any undue or unreasonable preferences to any such user" and the requirement of § 3661 that any service change "which will generally affect service on a nationwide or substantially nationwide basis" must be submitted to the Postal Rate Commission for an advisory opinion before it is instituted. In briefing and argument, however, plaintiffs placed primary emphasis on the argument based on §§ 3622 & 3623. Because of our disposition of the latter argument, we need not address plaintiffs' alternative contentions.

[11] The complaint was filed on October 20, 1977, and a motion for preliminary injunction was filed on October 28. A hearing on the motion was scheduled for November 21. On that date, but prior to the scheduled hearing, plaintiffs and defendant entered into a stipulation restricting the scope of the test. See note 9 supra. As part of that stipulation, plaintiffs withdrew their motion for a preliminary injunction. On the scheduled trial date of December 30, 1977, a stipulation of facts was moved into evidence and the matter was submitted on final hearing as a case stated.

[12] There is provision for temporary rates and classes in 39 U.S.C. § 3641, see note 2 supra & accompanying text, but such temporary changes are not exceptions from the requirement that rate and classification changes be filed with the Postal Rate Commission. Temporary rates and classes are permitted when changes, already filed with the Postal Rate Commission, are not acted upon within statutorily prescribed time periods.

[13] Under § 3661, Postal Rate Commission involvement in service changes is triggered only when the Postal Service plans "a change in the nature of postal services which will generally affect service on a nationwide or substantially nationwide basis." 39 U.S.C. § 3661. See generally Part II.B.4 infra.

[14] This argument is hinted at by the Postal Service in its brief, though not clearly articulated.

[15] The services in question included: (1) the furnishing of mail list corrections; (2) the privilege of prepayment of postage without stamps; (3) the forwarding or returning of undeliverable mail; (4) the registry of mail; (5) the insurance of mail; (6) the provision of COD mail; (7) the certification of mail; (8) the securing of a signed receipt upon the delivery of mail and the returning of it to the sender; (9) special delivery; (10) the special handling of mail; (11) the provision of money orders. See 40 Fed.Reg. 43233-34 (Sept. 19, 1975).

[16] We observe that, even under this reading of the statute, Commission jurisdiction over what we have termed "marketplace experiments" would not entail such jurisdiction over all experiments. The Postal Service would remain free unilaterally to engage in whatever experimentation it saw fit so long as the experiment did not impact upon the actual rate someone paid for his or her postal service. Dummy packages could be mailed to test sorting machines, costs of processing, efficiency of operations, etc. Market surveys could be taken to determine shipper receptivity to proposed new services. None of these things would raise the spectre of Rate Commission interference with management's institution or conduct of experiments.

[17] Defendant also cites American Airlines v. CAB, 123 U.S.App.D.C. 310, 359 F.2d 624, cert. denied, 385 U.S. 843, 87 S. Ct. 73, 17 L. Ed. 2d 75 (1966). American Airlines, however, involved rulemaking authority rather than individual tariff or service review, and therefore is not as relevant to the case at bar as are the cases noted in the text. In any case, to the extent American Airlines is relevant, its analysis adds nothing of significance to that which is offered by the other cases.

[18] Section 403(a) also instructs the Postal Service to perform its duties at fair and reasonable rates and fees. This responsibility is, of course, shared with the Postal Rate Commission. 39 U.S.C. § 3622. But the Postal Rate Commission does not explicitly share in the developmental mandate, except insofar as the necessity for ongoing planning, development, and promotion impacts on what would be "reasonable and equitable and sufficient" rates and fees. Id. § 3621. See id. § 3622(b).

[19] 39 U.S.C. § 3622(b) provides:

(b) Upon receiving a request [from the Postal Service for a recommended decision], the Commission shall make a recommended decision on the request for changes in rates or fees in each class of mail or type of service in accordance with the policies of this title and the following factors:

(1) the establishment and maintenance of a fair and equitable schedule;

(2) the value of the mail service actually provided each class or type of mail service to both the sender and the recipient, including but not limited to the collection, mode of transportation, and priority of delivery;

(3) the requirement that each class of mail or type of mail service bear the direct and indirect postal costs attributable to that class or type plus that portion of all other costs of the Postal Service reasonably assignable to such class or type;

(4) the effect of rate increases upon the general public, business mail users, and enterprises in the private sector of the economy engaged in the delivery of mail matter other than letters;

(5) the available alternative means of sending and receiving letters and other mail matter at reasonable costs;

(6) the degree of preparation of mail for delivery into the postal system performed by the mailer and its effect upon reducing costs to the Postal Service;

(7) simplicity of structure for the entire schedule and simple, identifiable relationships between the rates or fees charged the various classes of mail for postal services;

(8) the educational, cultural, scientific, and informational value to the recipient of mail matter; and

(9) such other factors as the Commission deems appropriate.

[20] 39 U.S.C. § 3623(c) provides:

(c) The Commission shall make a recommended decision on establishing or changing the [mail classification] schedule in accordance with the policies of this title and the following factors:

(1) the establishment and maintenance of a fair and equitable classification system for all mail;

(2) the relative value to the people of the kinds of mail matter entered into the postal system and the desirability and justification for special classification and services of mail;

(3) the importance of providing classifications with extremely high degrees of reliability and speed of delivery;

(4) the importance of providing classifications which do not require an extremely high degree of reliability and speed of delivery;

(5) the desirability of special classifications from the point of view of both the user and of the Postal Service; and

(6) such other factors as the Commission may deem appropriate.

[21] The Postal Service expressly declined to argue the point that sections 3621-3628 of the Postal Reorganization Act, which include the review provisions with which we are here concerned, are under a heading which reads: "SUBCHAPTER II—PERMANENT RATES AND CLASSES OF MAIL." In light of Congress' particular instructions in the Public Law version of the Act that "[a]n inference of a legislative construction is not to be drawn by reason of a chapter in Title 39, United States Code, as enacted by Section 2 of this Act, in which a section is placed nor by reason of the caption or catchline," Pub.L. No. 91-375, § 11(b), 84 Stat. 785 (1970), the Postal Service was correct in not making the argument. Divisions made and titled for organizational clarity are not always reliable indicators of substantive intent, and Congress has expressly forbidden us from treading this uncertain path of statutory interpretation. We are aware of the fact that the D.C. Circuit, at least, has referred to the subchapter heading to support the proposition that the affected sections of the Act are for permanent rates only. Direct Mail Advertising Ass'n, Inc. v. United States Postal Service, 147 U.S.App.D.C. 394, 458 F.2d 813 (1972). Even so, we feel ourselves constrained by § 11(b) of Pub.L. No. 91-375. Furthermore, the D.C. Circuit was only comparing permanent rates under subchapter 2, §§ 3621-3628, with temporary rates under subchapter 3, § 3641. We have already noted the role of temporary rates, which may be instituted if the Postal Rate Commission does not issue a decision in a specified period of time, in the structure of the Postal Reorganization Act. See text accompanying note 3 supra. The question of where, if at all, experimental rates fit into the statutory scheme was not at issue before that court. See also Direct Mail/Marketing Ass'n, Inc. v. United States Postal Service, 163 U.S. App.D.C. 157, 501 F.2d 717 (1974); Association of American Publishers, Inc. v. Governors of United States Postal Service, 157 U.S.App.D.C. 397, 485 F.2d 768 (1973). In each of these cases, although broad statements were made by the courts concerning the scope of subchapter 2, the only issues before the courts were the relative requirements of permanent and temporary (not experimental) rates and the applicability of provisions contained in one subchapter to rates instituted under the other.

We emphasize that although temporary, an experimental rate is not a "Temporary Rate" within the meaning of the Postal Reorganization Act. Temporary rates under the Act are instituted only after the Postal Rate Commission has failed to act on proposed permanent rates that have been filed with it, and may be instituted only under limited and circumscribed conditions and terms. Beyond temporal limitedness, experimental rates and temporary rates have little in common. Temporary rates have standards, safeguards, ceilings, and, since they have been filed with the Postal Rate Commission and in the Federal Register, see 39 U.S.C. § 3641, exposure built into the procedures applicable to them. On the other hand, experimental rates, under the Postal Service's reading of the Act, have no such circumscriptions, even though they could have comparable impact on the public or at least segments of it, depending on the scope and duration of the experiment. The sunshine and public interest purposes of the Postal Reorganization Act are thus represented in the temporary rate provision.

[22] We note at the outset of this discussion that we do not find compelling the Service's citations to cases in which courts have upheld agency approval, without hearing, of marketplace experiments. See Network Project, supra; Saturn Airways, Inc., supra; Delta Air Lines, Inc., supra; United Telegraph Workers, supra. As an initial matter, we observe that the cases do not stand for the proposition that the relevant agencies would have abused their discretion had they held hearings on the proposals before them. We do not miss the point that those courts apparently believed the non-hearing route to be preferable; nor do we necessarily disagree. At most, however, these cases establish only that it was reasonable under the facts presented not to hold a full-blown evidentiary hearing for the limited experiments proposed. That is far from holding that it necessarily and in all circumstances would be unreasonable or an abuse of discretion to hold a hearing. More significantly, although we might vote otherwise were this a legislative rather than judicial forum, for the reasons that follow we think that Congress could reasonably have intended the hearing mechanisms of the Postal Reorganization Act to be triggered by proposed marketplace experiments, and therefore are not disposed to upset, on these grounds, the conclusion that we tentatively reached by a plain reading of the statute.

[23] In fact, the Postal Rate Commission did conduct expedited proceedings under 39 U.S.C. § 3661 when the Postal Service began to provide service to all first-class mail comparable to that previously provided airmail only. See 40 Fed.Reg. 20866 (1975); P.R.C. Docket No. N75-2; Reply of the United States Postal Service to Plaintiff's Trial Memorandum 5 n.1. In any case, the provisions for temporary rates would be applicable, 39 U.S.C. § 3641, putting an upper limit on the time a proposed experiment could be blocked by bureaucratic footdragging. See note 3 supra & accompanying text.

[24] It seems to us that the matter of "substantially nationwide impact" is the only issue under the Buchanan rubric even arguably in question here, and then only if "nationwide" be construed in a postal user rather than geographic sense, for the Service Test Plan does have broad geographic effect. See notes 5 & 7 supra. We have no doubt that as to the shippers participating in the Service Test Plan there has been a quantifiable "change" in rates and mail classifications that are "meaningful" in impact.

[25] From all that appears, § 3661 would require Commission review of a nationwide experimental change in services. Indeed, the Postal Service appears to concede that "preparatory activities" of the Postal Service, though affecting "the Postal Service's development function" are subject to hearing procedures if "substantially the entire postal system" is affected. Taken at face value, this seems to mean that the issue is not the experimental nature of the change, but the scope. As the text of the opinion shows, however, there is no textual limitation on the scope of changes subject to review under §§ 3622 & 3623, as there is under § 3661.