dissenting.
I.
Competition, experimentation and creative business policies designed to better serve the consumer are highly valued in American society. The traditional folklore has been that one beats his competition by “building a better mousetrap,” or, in the context of this case, by providing better delivery services. The majority rejects these traditional values — so they say, because of a statutory mandate. Thus, we have the paradox of the United Parcel Service (UPS) one of the United States Postal Service’s (Postal Service) major competitors, preventing the Postal Service from undertaking on its own a modest and limited experiment which is a relatively small threat, if any, to the plaintiffs, involving no more than twenty selected shippers in no more than five metropolitan areas. As I view the applicable statutes, the Postal Service need not be shackled and prevented from pursuing such modest experimentation. It has some latitude to experiment, without indulging in further bureaucratic processes, and to determine whether a plan can be devised to provide the American public with better service and its major competitor with more robust competition. The learned trial judge below, in his customarily thorough fashion, noted that “as a matter of pure policy the result advocated by the Postal Service is more desirable.” United Parcel Service v. United States Postal Service, 455 F.Supp. 857, 863 (E.D. Pa.1976). However, he and the majority have concluded that the statute precludes even the modest experiment proposed by the Postal Service in this case. I agree that the result advocated by the Postal Service is far more desirable. But, I disagree with the majority, and the court below, because their decision, from my view, makes no sense when considered against the relevant legislative history. Thus, I dissent.
II.
As the majority’s opinion observes, the Postal Service and the Postal Rate Commission (Commission) were established under the Postal Reorganization Act of 1970, 39 U.S.C. §§ 101-5605 (the Act), which created a new and comprehensive scheme for the management of the nation’s postal system,1 so that the nation’s mail could be handled more efficiently and with less expense.
Acting in accord with this Congressional design, the Postal Service has attempted to undertake a very small and temporary experiment in order to develop a new technique to deliver parcel post more efficiently. The experiment was to involve only twenty shippers in five metropolitan areas, who were to handle about 907,000 items of mail. Defendant’s Brief Before Dist. Ct., Exhibits 1(A)-1(R). When compared with the approximately 92.2 billion items of mail and 387 million items of parcel post that the Postal Service processed in 1977, the limited nature of this experiment is self-evident. Department of Commerce, Statistical Abstract of the United States 585, 587 (1978); Joint Appendix, at 29.
In this appeal we are presented with the questions of how Congress intended for the new postal management to modernize the postal system and where Congress placed the responsibility of supervising rate and classification experiments.
*1383In order to decipher the Congressional scheme and resolve this question, the Act and its extensive legislative materials must be reviewed. I am convinced that a review of the legislative materials in their totality is especially important here where the legislation was designed to effect a comprehensive reorganization.
The majority focuses on the language of sections 3622 and 36232 and urges that these provisions place the responsibility for supervising rate and classification experiments in the hands of the Rate Commission. To be sure, the challenged experiment involves a “change” in the way the limited number of program participants send their parcel post. It is also evident that the “rates” and “classifications” used in this experiment are different from those normally applied to the Postal Service’s customers. However, it is not self-evident whether the temporary and small-scale “changes” which occur as a result of an experiment are the “changes” Congress was referring to when it drafted these sections of the Act. The decision to make a national or permanent rate or classification change is a management decision of a different quality from a decision to undertake an experiment. At the very least, different criteria of evaluation may be used and different results may flow from the decision. Therefore, before these sections are read to cover experiments it is necessary to unravel the management scheme Congress envisioned.
The majority does so by analysing whether the ordinary, plain, dictionary, or regulatory definitions of the words “rates,” “classifications” and “change” are to be used. In doing so, the majority misses the essential question: Whether these sections were intended to give the Commission the responsibility of supervising rate and classification experiments. As Learned Hand said:
[Ijt is one of the surest indexes of a mature and developed jurisprudence not to make a fortress out of the dictionary; but to remember that statutes always have some purpose or object to accomplish, whose sympathetic and imaginative discovery is the surest guide to their meaning.
Cabell v. Markham, 148 F.2d 737, 739 (2d Cir.), aff’d, 326 U.S. 404, 66 S.Ct. 193, 90 L.Ed. 165 (1945) quoted in, United States v. Cerilli, 603 F.2d 415, 429 (3d Cir. 1979) (Aldisert, J., dissenting). Arguably Congress might have intended for sections 3622 and 3623 to cover experiments, although they make no reference to experiments. The majority argues that the absence of such a reference supports their view of the Act. It seems equally plausible to me that it indicates a possible limitation on the scope of the Commission’s authority. The ease with which contrary arguments can be made about the importance of the absence of a single word in the statute strengthens my conviction that resort to the legislative history is essential.
In this case, our inquiry into Congressional intent is made difficult by the fact that the legislative materials are also silent about rate and classification experiments. However, we are obliged to find an answer and therefore must resort to materials from which the Congressional intent can only be inferred.
III.
A.
After reviewing the Act in its entirety and its legislative history, I am convinced that Congress intended to permit the Postal Service to undertake rate and classification experiments and that it did not intend for the Postal Service to seek Commission approval before it undertook such an experiment. I reach this conclusion because of the overwhelming evidence contained in the legislative history that shows that Congress went to considerable effort to give the Postal Service full management authority. Congress wished to strip away and not create layers of managerial control so that new and more efficient modes of operation *1384could be developed and could become a part of our nation’s postal system. Commission review of experiments would be inconsistent with that Congressional purpose. Moreover, I have found nothing in the legislative materials that shows that Commission review was nevertheless intended. Indeed, the references to the Commission contained in both the Act and the legislative history show that the Commission was established and structured so it might address concerns that are not implicated in the experimentation process.
B.
In this case, as in many others, the legislative history is ambiguous in many regards. But arising above all ambiguity is the Congressional conviction that the Postal Service should be more efficient, utilizing modern managerial techniques rather than sinking further into its maelstrom and that the Postal Service should have, in most instances, full authority to undertake that effort.
The Congressional debates, reports, and the other legislative materials are replete with references to the inefficiencies of the pre-1970 postal system. For example, the House Report stated:
For several years there has been a growing awareness . . . that our vast sprawling postal complex is heavily overburdened and in deep trouble. Delays, breakdowns, errors, damage, and other inconvenience to the public have become more and more frequent.
H.Rep.No.91-1104, 91st Cong., 2d Sess. (1970) reprinted in [1970] U.S.Code Cong. & Admin.News 3652 [hereinafter H.Rep.] See also S.Rep.No.91-912, 91st Cong., 2d Sess. 2 (1970) [hereinafter Sen.Rep.]; Remarks of Rep. Dulski, 116 Cong.Rec.19844 (June 16, 1970); Towards Postal Excellence the Report of The President’s Commission on Postal Organization 1-14 (1968) [hereinafter President’s Commission].
These materials are also replete with references that show that Congress believed these deficiencies were caused by ineffective management, which in turn resulted primarily from a lack of managerial flexibility and freedom and from a lack of centralized control. For example, the Senate Report which accompanied the Senate version of the Act stated:
[T]he Postmaster General is blocked or undercut at every turn by a labyrinth of postal statutes echoing every postal concern, interest, or whim expressed by Congress over a 200 year period . leaving the Postmaster General bound in his own house. Twist and turn as he may, he cannot function in the public interest as a responsible manager.
Sen.Rep., supra, at 2.
The House Report stated:
It has now become apparent that all the shortcomings of The Post Office Department are bound up in the fact that responsibility for managing the system is shared by a number of executive agencies and by several congressional committees.
H.Rep., supra, at 3653. See also President’s Commission, supra, at 33-43.
The Act was designed to correct this problem. New and more efficient systems were to be developed. The new managers would
create a lasting foundation for a modern, dynamic, and viable postal institution that is equipped and empowered at all times to satisfy the postal requirements of the future.
H.Rep., supra, at 3650.
And the Postal Service was to have major responsibility for the development of those new systems. The House Report stated:
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.
H.Rep., supra, at 3657. The Act itself provides that the Postal Service “shall plan, develop, promote, and provide adequate and efficient postal services.” 39 U.S.C. § 403.
Further, it is evident that Congress felt that modernization could be accomplished only if the Postal Service was provided broad authority to undertake its duties.
*1385The House Report stated that the Act was designed to
[eliminate serious handicaps that are now imposed on the postal service by certain legislative, budgetary, financial, and personnel policies that are outmoded, unnecessary, and inconsistent with the modern management and business practices that must be available if the American public is to enjoy efficient and economical postal service.
H.Rep., supra, at 3650. It also stated:
[T]he only solution to these problems is fundamental reform that puts complete responsibility in a single place, with appropriate safeguards against abuse of that responsibility.
Id. at 3653.
The Senate agreed. Its report stated: [The Act] is not a halfway measure. Its objective is to establish a postal structure and a method of operating that will entrust the management of the U.S. Postal Service, within broad policy guidelines designed to protect the ordinary mailer, to responsible public officials.
Sen.Rep., supra, at 3.
Thus modernization and managerial freedom were prime Congressional objectives.
C.
It is also evident that Congress anticipated that rates and classifications would be changed as a result of the development of more efficient systems and that the Postal Service would be involved in that process.
The interrelationship between rates and classifications changes and the modernization of postal services was discussed in the House Report which stated:
Since the Postal Service will be required to provide postal services “at reasonable and equitable rates and fees,” massive and far-reaching improvement in postal efficiency must be achieved before a self-supporting Postal Service becomes a reality.
H.Rep., supra, at 3665;
Ample evidence is available to show that Congress intended for the Postal Service to be involved with rate and classification changes.
Most dramatic is the fact that the Act very explicitly states that the Postal Service and not the Commission has the final authority to set rates and classifications.
The Act provides:
Upon receiving a recommended decision from the Postal Rate Commission, the Governors [of the Postal Service] may approve, allow under protest, reject, or modify that decision. 39 U.S.C. § 3625.3
Moreover, there is evidence that shows that Congress intended for the Postal Service to provide the Commission with information so that the Commission could make decisions about rate and classification changes.
For example, one of Congress’s concerns about the pre-1970 rate-making process was that costs attributable to a particular class of mail were often not charged to that class of mail. Under the Act, each class of mail was to bear its own costs. 39 U.S.C. § 3622. While perhaps a simple theory, it was anticipated that implementation of this mandate would be difficult because, under the accounting system used by the Post Office Department, costs could not readily be attributed to a particular class of mail. President’s Commission, supra, at 132-35. The Postal Service was thus empowered and given a Congressional mandate to develop an accounting system that permitted cost attribution so rates could be accurately set. 39 U.S.C. §§ 2008, 2009, 2401. See Association of American Publishers v. Governors of U.S. Postal Service, 157 U.S.App.D.C. 397, 485 F.2d 768 (1973).
I therefore disagree with the Court of Appeals for the District of Columbia, which implied in language the majority quotes, that rates and classifications were Commis*1386sion business and that the Postal Service was to have nothing to do with those matters.4 National Association of Greeting Card Publishers v. United States Postal Service, 186 U.S.App.D.C. 381, 569 F.2d 470 (1976), vacated in part, 434 U.S. 884, 98 S.Ct. 253, 54 L.Ed.2d 169 (1977). The Act does not provide for such a sharp delineation of duties. Indeed, the majority concedes that the Postal Service has an independent role to play in rate and classification making when it recognizes the right of the Postal Service to develop, plan and implement a “test plan” that would provide information that would be used by the Commission. Majority’s opinion, supra, at 1377.5
It is against this backdrop of Congressional concern that the responsibility for experimentation should be evaluated. Alone, it would support the Postal Service’s authority to engage in the rate and classification experiment challenged by the UPS, an experiment which was designed to develop data on new techniques for the handling of parcel post and the effect of different rates and classifications.
D.
The majority argues that regardless of this history, the existence of the Commission demonstrates a Congressional intent to restrict the power of the Postal Service to undertake experiments. I do not agree. It is my belief that the legislative materials discussed above compel any court to exercise caution before it decides that Congress intended to place a restraint on the Postal Service’s management authority. Indeed, I believe that this history, which demonstrates an overwhelming concern with deregulation, almost compels us to hold that management flexibility exists, unless contrary evidence is provided. Cf. United Telegraph Workers v. F.C.C., 141 U.S.App.D.C. 190, 194, 436 F.2d 920, 924 (1972) (“This expression of congressional desire that the Commission encourage technological innovation requires us to demand a compelling showing of legislative prohibition before we strike down an experiment . which is designed to furnish the information input that makes such innovation possible.”).
It is my view that the legislative materials do not provide the evidence necessary to reject the presumption of flexibility and to hold that Congress intended to have the Commission supervise experiments. Moreover, these materials provide positive support for my conclusion that supervision of experiments was not one of the duties envisioned for the Commission.
As I note above, the descriptions of the Commission’s responsibilities contained in the legislative documents are somewhat lacking as they contain no reference to rate and classification experiments. However, read in their totality they suggest that the Commission was given the responsibility of setting national policy on rates and classifications and acting as an overseer to ensure that these policies were properly implemented. See H.Rep., supra, at 3666, 3667-68, 3692-95 and Sen.Rep., supra, at 13-17. To be sure, at times this responsibility would include the supervision of permanent rate or classification changes that would have a limited impact, for example, a subsidized rate for sending a particular type of periodical. However, the duty to supervise these changes is derived from the broader duties. In the case of an experiment, such as the one challenged here, whose object is the testing of new systems of handling mail, these broader duties are not implicated. This suggests that the duty to supervise experiments was not included among the Commission’s responsibilities.
*1387Sections 3622 and 3623 also make no reference to rate and classification experiments. Again, read in their totality they give us some assistance. For example, the statutes outline a specific process which the Commission must follow when it makes rate and classification decisions. Section 3624 further provides that the Commission can ap - prove a request by the Postal Service only after an opportunity for a hearing conducted pursuant to the Administrative Procedure Act has been granted.6 This process is simply not a productive one when a managerial decision as to the necessity for the development of data on a system change is being made. Moreover, the information that would be necessary to have a meaningful hearing on the desirability of the change would not be available until the experiment is conducted. Cf. Delta Airlines v. CAB, 147 U.S.App.D.C. 272, 276, 455 F.2d 1340, 1344 (1971) (“The CAB in reality is not able to investigate without testing how this experimental tariff actually works.”) and American Airlines v. CAB, 123 U.S.App. D.C. 310, 319, 359 F.2d 624, 633 cert. denied, 385 U.S. 843, 87 S.Ct. 73, 17 L.Ed.2d 75 (1966) (“A month of experience will be worth a year of hearings.”).
In addition, I envision that during the process of an experiment the Postal Service might wish to modify some portion of the experimental design on rates or classifications. Under the majority’s reading of the statute, hearings would be necessary. I cannot believe Congress intended so absurd a result. For example, during the course of the experiment challenged here the Postal Service might decide that packages were falling in different weight groups than originally anticipated and may wish to change the weight classifications in the experiment. It would simply make no sense to have a hearing to make the modification.
Another indication that Congress did not intend to have the Commission review experiments is also found in sections 3622 and 3623. These provisions, section 3622(b) and section 3623(c), state that the Commission’s decisions are to be made in accordance with the policies of the Act and with a number of itemized factors.7 These factors, which include factors such as the relative value of the kind of mail transmitted and the effect of rate increases upon the general public are not generally applicable to experiments. E.g., 39 U.S.C. § 3622(b)(4) and § 3623 (c)(2). Moreover, some of the factors involve cost evaluations that cannot be made before an experiment is complete. 39 U.S.C. § 3622(b)(5) & (6).
Finally, an examination of the reasons that were given to support the Commission’s creation leads me to conclude that the Commission was not designed to supervise experiments. These reasons are, for the most part, relevant only when major policy decisions are made, when the integrity of the Postal Service could be threatened, or when one of the Act’s underlying purposes could be frustrated. These are not concerns which arise during an experiment.
First, many members of Congress were concerned that while they wanted the Postal Service to operate as an efficient business enterprise, they were concerned that the
Postal Service in fact and shall be operated as a service to the American people, not as a business enterprise, designed to provide excellent postal service.
Sen.Rep., supra, at 4.
It was felt that the Commission would be able to strike the necessary balance, and would ensure that postal rates reflected not only management objectives but also served broader public purposes, e. g., certain nonprofit organizations might be given special rates. These broader public purposes are not implicated by an experiment.
*1388A second impetus for the establishment of the Commission was the sentiment that some distance from the day-to-day operations of the Postal Service might be necessary to make sound business decisions on rates. See Majority’s Opinion, supra, at 1373-1374. Again an experiment would not tamper with this concern, and would, if anything, permit the Postal Service to give the Commission information it needs to perform this function.
Congress was also concerned about the impact political pressure had had on rate and classification determinations. Under the reorganization, Congress wished to design a system that would be as free from political influence as possible, where experts could make decisions based on the technical problems of running the postal system and not the effectiveness of particular lobbyists. See, e. g., Remarks, Sen. McGee, 116 Cong.Rec. 26955 (1970).
• The Commission was seen as a solution to this problem. The House report stated:
The Commission provides an invaluable buffer between the management of the Postal Service and the possible influence of partisan politics. Insulation from partisan politics is one of the principal purposes to be served.
H.Rep., supra, at 3660.
It is conceivable that partisan politics could influence the Postal Service’s decision to select particular participants in an experiment which used cheaper rates. But there is no showing of partisan politics in the pleadings or the stipulated facts in this case. To the contrary, it is clear that this “Local Parcel Seryice Test Plan” was created for managerial purposes.8 The authority to undertake limited and temporary experiments does not permit the return to the days when every interest group imaginable had special permanent rates established for them. Moreover, political influence on the decision to undertake an experiment or to choose a participant is simply not as dangerous as on decisions where national policy is made.
In our zeal to protect the public we must remember that the public is not always protected by more regulation and more supervision. That is the overwhelming message of the Postal Reorganization Act. Indeed, in this case, the effort to impose regulation upon the Postal Service and to require it to seek Commission review is spearheaded not by a representative of the general public but by one of the Postal Service’s major competitors, the United Parcel Service.
The majority rejects the present plan because it hypothesizes that the authority to approve an experiment for twenty shippers in five metropolitan areas “would • permit unregulated changes in rates and mail classification at any time and under any circumstance whereby less than all members of the public were entitled to benefits stemming from such changes.” Majority Opinion, supra, at 1377. This “logic” is the worst example of the parade of horribles. If the Postal Service, under the rubric of experimentation, attempted to set up permanent or specialized rates for any individ*1389ual or group presumably a court would not be misled by a label marked “experiment.” Moreover, if the Postal Service abused the limited authority that I believe Congress intended for it to have, the Commission has the power under 39 U.S.C. § 3662 to review and prohibit Postal Service actions that circumvent the policies of the Act.
One final point raised by the majority. Í believe 39 U.S.C. § 403, which proscribes the “unreasonable discrimination” “preferences,” provides no assistance in resolving the question in this appeal. Majority’s Opinion, supra, at 1377. First, any experiment will discriminate against other users of the postal system whether or not the Commission approves it. And second, the section only prevents “unreasonable discrimination.” A properly conducted marketplace experiment is certainly not an “unreasonable” type of “discrimination” or “preference.”
IV.
I believe the experiments are a permissible exercise of Postal Service authority. I do not believe, as the majority concludes, that the Congress, in its attempt to free the postal system of constraints so that it might operate as a modern and efficient enterprise, intended to require the Postal Service to undertake the time-consuming process of hearings and Commission review in order to develop and provide data for the Commission’s rate-making judgments.
One of the nation’s leading scholars on administrative law, Dean James 0. Freedman, in his classic treatise, Crisis and Legitimacy: The Administrative Process and the American Government, 261-62 (1978), has stressed:
The legitimacy of the administrative process cannot turn, then, upon its nonconformity to a simplistic version . Rather, it must be tested pragmatically, by the responsiveness of administrative institutions to the most fundamental principles of a democratic society and by the degree to which administrative institutions meet the nations highest aspirations for justice and effective government.
Unfortunately in this case, despite the Congressional dream and design to bring flexibility, creativity and managerial efficiency to the Postal Service, the public’s aspirations have once again been dashed. The public has been precluded from having within the Postal Service “justice and effective government.”
I therefore dissent.9
. The Act also established a fourteen member Postal Service Advisory Council to advise the Postal Service on “all aspects of postal operations.” 39 U.S.C. § 206. The role of the Advisory Council is not raised in this appeal.
. See Majority opinion, supra, at nn. 4 & 5 for text of statutes.
. The full text provides that in the event that the Postal Service wishes to reject or modify a decision, the Postal Service must resubmit its request and if still dissatisfied it may modify or reject the decision if it is inconsistent with the Act or if it will produce insufficient revenues.
. I do not mean to suggest by this point that the Commission does not have substantial and dominant authority in the area of rate and classification making. My point is simply that Congress intended for the Postal Service to have a significant involvement in this process, especially with regard to the development of data to be considered by the Commission.
. The majority has reserved for another day the question of whether the Postal Service may conduct, even with Commission approval, marketplace experiments. However, the majority does not seem disturbed by the Postal Service’s involvement in rates and classifications; it is merely the type of involvement it questions.
. 39 U.S.C. § 3624 provides:
(a) The Postal Rate Commission shall promptly consider a request made under section 3622 or 3623 of this title, except that the Commission shall not recommend a decision until the opportunity for a hearing on the record under sections 556 and 557 of title 5 has been accorded to the Postal Service, users of the mails, and an officer of the Commission who shall be required to represent the interests of the general public.
. See Majority opinion, supra, at 1374-1375 n.4, & n.5 for the text of the statute.
. The Postal Service outline of the experiment indicates the following:
Test Objectives
A. The first objective is to determine the feasibility of the new operations configuration affecting:
1. Acceptance procedures.
2. Local Holdout and BMC processing.
3. MVS operations.
B. The second objective is to determine the costs associated with reduced handlings which result from the new operations configuration.
C. The third objective is to determine the service and cost impact on the mailer and to measure mailer acceptance of this impact.
D. The fourth objective is to analyze the service within the context of our continuing review of the structure of traditional services. Issues to be explored are:
1. Whether nearer Parcel Post zones are in need of an adjustment?
2. Whether volume requirements for Bulk Parcel Post are in need of adjustment?
3. Whether pick-up (not plant load) offered as an option or offered as a feature (no option) of the service is needed?
4. Whether further refinements are needed in the single piece third class and zone-rated, fourth class classifications and rate design?
. The majority has decided this appeal without determining whether the Postal Service has the authority to undertake the experiment even if it seeks Commission approval. At oral argument the Postal Service indicated that it still had an interest in conducting this specific experiment. As a result of the majority’s decision, the Postal Service will have to submit the proposed experiment to the Commission and participate in hearings with the possibility that it will be determined after a second challenge, a second decision by a district court and a second appeal to this court, that the Postal Service cannot undertake marketplace experiments even with Commission approval. (I would of course disagree with that conclusion). I believe this would be a waste of the resources of the Postal Service, the Commission and the judiciary. The majority should reach this issue.