03-28643. Periodic Reporting Rule

AGENCY:

Postal Rate Commission.

ACTION:

Final rule.

SUMMARY:

The Commission is adopting a final rule that updates the periodic reporting regulations. These regulations identify the data and information the Postal Service is to file with the Commission on a regular, ongoing basis. The final rule differs from the proposed rule in several important respects. The Commission has narrowed or eliminated some filing requirements and has incorporated some flexibility in meeting other requirements. Adoption of these changes should facilitate the public's ability to more readily grasp the quantitative basis and support for Postal Service proposals.

DATES:

Effective December 19, 2003.

FOR FURTHER INFORMATION CONTACT:

Stephen L. Sharfman, General Counsel, 202-789-6820.

SUPPLEMENTARY INFORMATION:

Regulatory History

68 FR 2272 (Thursday, January 16, 2003)

Executive Summary

When the Postal Service requests a general rate increase it supports the request with estimates of how much each of its products costs, and how much revenue it needs. The validity of these estimates are central issues in the hearings that the Commission holds to review the request. Tens of thousands of pages of economic testimony and documentation, most of it highly technical, are offered to support raising almost $70 billion in annual revenue from over 200 postal products.

The Postal Service and interested members of the public have the right to present a case in support of the rates that they advocate, and the right to challenge the cases presented by others. The Commission must develop recommended rates based on the record within a ten-month statutory deadline. Most participants agree that this severely compressed process strains their resources to the limit. The Commission also reviews Postal Service requests for experimental rates or classifications in even more compressed time frames.

The Commission has a Periodic Reporting Rule to facilitate this process. It requires the Postal Service to provide certain relevant financial and operating reports prepared for Postal Service management. The process will be further streamlined by promptly providing the Commission and the interested public with access to the standard, routinely prepared cost and revenue data that serve as the basis for rate and classification requests.

One of the key reports that the Postal Service currently submits each year is the Cost and Revenue Analysis (CRA). It contains the Postal Service's cost, volume, and revenue estimates for the most recently completed fiscal year, both in total, and by individual product. Because the postal system has a high proportion of shared costs, it is difficult to estimate the costs that each product causes. The CRA that the Postal Service submits under the current rule estimates unit costs caused by each product, but does not provide documentation showing how those estimates were derived. Consequently, it is of limited value in identifying trends in product costs, or in analyzing their causes, which are core issues in rate and classification cases.

Under the existing Periodic Reporting Rule, the Postal Service has been providing some information about how it derives unit product costs (primarily in the mail processing and transportation areas.) The updated Periodic Reporting Rule adopted in this order asks the Postal Service to do this for all of its 20 cost segments.

For each cost segment, the updated Periodic Reporting Rule asks the Postal Service to provide the basic datasets that it uses to estimate unit product costs, and identify any new estimating technique it applies to those data to derive the unit cost estimates in the CRA. Having this information filed each year, rather than waiting for the Postal Service to provide it in a general rate case, should produce the following benefits:

• When the Postal Service files a general rate case, litigants and the Commission will already be familiar with the standard cost and revenue reports on which much of the case is based. This should reduce the need for discovery, and make it possible to shorten hearings.
• When the Postal Service files requests for experimental classifications, market tests, or negotiated service agreements, litigants and the Commission should be able to evaluate them more quickly.
• Between general rate cases, the Commission and the public will be able to analyze the accuracy of the cost, volume, and revenue projections on which current rates are based.
• Between general rate cases, if the CRA shows that cross-subsidy or other rate inequity exists, affected parties will have a basis for asking the Commission to hold a hearing to investigate the matter and fashion a remedy under § 3662.
• Between general rate cases, if the CRA shows that costs are shifting in ways that call current classifications into question, the Commission will have a basis for initiating a classification hearing to investigate the matter and fashion a remedy under § 3623.
• Between general rate cases, parties looking to propose alternative models of postal cost behavior in future rate cases will be able to analyze data that reflects current postal operations.

Seven parties filed comments in response to the Commission's proposal to update the Periodic Reporting Rule. All but the Postal Service support the rule. They note that a general rate filing typically consists of thousands of pages of highly technical testimony and computer material. They complain that the Postal Service takes whatever “lead time” it needs to prepare such filings, whereas they have no lead time to react to it and prepare alternatives. They argue that this makes it extremely difficult for them to comprehend the Postal Service's case and to present alternatives in the brief hearing time available. They argue that having access to a more thoroughly documented CRA will help to “level the playing field” in rate litigation.

The Postal Service objects that by requiring it to provide some supporting documentation for the CRA reports that it issues between rate cases, the updated Periodic Reporting Rule would impose a burden equivalent to a rate case presentation every year. It also argues that by requiring it to disclose the costs of its various products each year, the updated Rule would jeopardize the competitiveness of its products.

The update is consistent with the main purpose of the current rule, which has always been to expedite the processing of general rate cases, and to allow research into postal cost, volume, and revenue behavior to continue between rate cases. The information required by the updated rule also will facilitate the Commission's statutory duties to hear complaint cases and to initiate classification cases, regardless of whether a Postal Service rate request is pending.

The Commission finds that complying with the updated rule should not add significantly to the Postal Service's regulatory burden, since the Postal Service annually prepares almost all of this material for its own purposes. The Postal Service's own estimates indicate that under normal circumstances complying with the updated rule would involve only a tiny fraction of the resources that it devotes to preparing a general rate filing.

With respect to potential competitive harm, the Commission believes that history shows that there is little cause for concern. The Postal Service annually disclosed the information required by the updated rule for six years (FY 1995 though FY 2000) with no indication that competitive harm resulted.

The revised rule also is consistent with recent reports by the President's Commission on the Postal Service and the General Accounting Office that conclude that greater Postal Service financial transparency is necessary.

I. Introduction

On January 8, 2003, the Commission issued a notice of proposed rulemaking (NPR) in this docket proposing to update its Periodic Reporting Rule (39 CFR § 3001.102). PRC Order No. 1358; 68 FR 2272-2275, Thursday, January 16, 2003. Rule 102 contains a list of reports and documentation that the Postal Service is required to provide on an ongoing basis to the Commission on its financial condition and operating results. Since its inception in 1976, the objective of the Periodic Reporting Rule has been to ensure that the Commission and the interested public have access to current financial data and operating results that are routinely reported to Postal Service management. This brings a number of important benefits to the ratemaking process. These benefits were mentioned briefly in the NPR, and are discussed in more depth in this notice.

The NPR observed that the list of financial reports covered by the Periodic Reporting Rule has not changed since the mid-1980s, even though the nature of the reports that the Postal Service routinely produces to inform management of its financial and operating results have evolved substantially over that time. This final rule updates the list of required periodic reports and documentation to reflect the increasingly sophisticated financial information ^[1] regularly produced by the Postal Service and the improved ability of the Commission and the public to understand and benefit from such reporting. The final rule is consistent with the goal of greater financial transparency for the Postal Service that has recently been recommended by the General Accounting Office and the President's Commission on the Postal Service.

Joint comments on the NPR were received from the American Bankers Association and the National Association of Presort Mailers (ABA/NAPM). Comments were also received from American Business Media (ABM), the Greeting Card Association (GCA), the Office of the Consumer Advocate (OCA), United Parcel Service (UPS) and the Postal Service. The OCA and UPS proposed additional changes that the Commission has decided not to include in the final rule.

The Purposes of the Final Rule

The final rule calls for the periodic submission of financial information that is routinely prepared for postal management between omnibus rate cases. This information does not relate directly to a particular revenue requirement or set of proposed rates, and none of the information that it seeks about the Postal Service's financial estimation process relates to the justification for or merits of that process. The data and the estimation-process information that the rule requires shed light on the ratemaking process in a generic sense, and will improve the ability of the Commission to process future rate, classification, and complaint cases within the tight deadlines imposed by the Postal Reorganization Act and the Commission's own administrative rules. The information sought will also help the affected public to participate more meaningfully in such cases. The rule seeks routinely generated reports that disclose the Postal Service's current financial condition, allow operating and financial trends to be identified as they unfold, and allow the Commission and the public to test the validity of methods by which the Commission estimated the costs, volumes, and revenues upon which current rates are based. If the Commission and the affected public have evaluated this financial background information prior to the filing of a case, it is likely that they will not have to spend a substantial part of the brief time allotted for litigating the case trying to “get up to speed” on the issues related to the Postal Service's routine financial reports.

The rule also seeks to permit the Commission to stay informed on the “state of the art” procedures by which the Postal Service currently attributes costs. Under the current regulatory scheme, the Commission is the expert body with the ultimate say on what methods should be used to attribute postal costs to classes of mail. See National Association of Greeting Card Publishers v. USPS (NAGCP IV), 462 U.S. 810, 833 (1983). As a practical matter, however, most of the methods used to attribute postal costs to the classes of mail originate with the Postal Service. This is because it controls the cost, volume, and revenue data and determines for itself what estimating techniques it will use to compile its periodic financial reports. It also controls almost all of the data that will be used in rate, classification, and complaint cases. Between cases it decides for itself what techniques will be applied to the data and incorporated into its Cost and Revenue Analysis (CRA) report. Its CRA has become a massively intricate, partially documented, automated cost attribution engine that most interested participants cannot fathom, duplicate, or develop realistic alternatives to, in the narrow litigation window available to them. The difficulty in deciphering the CRA in the time allotted has profound due process implications, since the CRA inevitably provides the methodological baseline for Postal Service rate and classification requests.

Because the data sources and estimating techniques that the Postal Service incorporates into its CRA change unpredictably, the “state of the art” is a moving target to the outside world. Neither the Commission nor the interested public can competently interpret the results presented in the Postal Service's routine financial reports, because they have no way to distinguish between what appear to be changes in cost, volume, and revenue behavior, from changes in the methods that the Postal Service uses to measure that behavior. Changes in data sources, or changes in estimation technique can have large impacts on the attributable cost estimates in the CRA, as the recent history of carrier street time attribution demonstrates. The Commission's rules require the Postal Service to provide detailed documentation in rate cases of the changes that it proposes in cost estimating procedures, and allows for discovery. Between rate cases, however, there is no way for the outside world to know what the state of the art in cost attribution is.

To overcome this problem, the Periodic Reporting Rule will now require the Postal Service to identify all changes made since its most recent omnibus rate request to the data sources and estimation techniques used to produce the CRA, and to provide enough supporting material to allow the Commission and the affected public to understand these changes. This provides at least some basis for understanding and evaluating the estimates summarized in the CRA, and provides some insight into the likely causes of the trends discerned in these summary figures.

The Commission will be able to use the information contained in adequately documented periodic reports to decide whether it should institute a classification case. Likewise, a customer or a competitor will be able to use information gained from adequately documented periodic reports to determine whether rates are in violation of the policies of the Act and whether a complaint should be filed with the Commission. Both are functions that the Act authorizes the Commission and the public to perform whether or not the Postal Service is litigating an omnibus rate request. The Commission, customers, and competitors of the Postal Service cannot make adequately informed decisions to invoke these provisions of the Act between omnibus rate cases if they cannot competently interpret or evaluate the Postal Service's routine financial reports.

More important, the partial documentation of periodic reports that the rule requires facilitates the processing of future rate, classification, or complaint cases because it gives the Commission and the affected public some hope of keeping current with the “state of the art” by which the Postal Service attributes costs to the classes of mail. The rule does not require the Postal Service to explain or justify the changes that it has made to its cost attribution engine, but it requires the Postal Service to disclose changes to the mechanical process by which that engine attributes costs. This will allow the Commission and the interested public to identify what the Postal Service's current CRA does, if not why it does it. By staying informed of what the Postal Service's current cost attribution engine actually does, the Commission and the interested public will be able to respond more quickly to an omnibus rate request, or to a case filed by the Postal Service under expedited rules, because they will not have to spend a substantial portion of the available litigation window reacquiring this necessary expertise.

The Periodic Reporting Rule requires the Postal Service to provide the basic datasets that it uses to produce its financial estimates in the CRA between cases. Public access to current datasets between rate cases is needed because they are the raw material that others must have if they are to develop their own cost attribution, volume forecasting, or revenue forecasting techniques that reflect current operations. Providing these datasets between omnibus rate cases facilitates the functioning of the Act because it allows potential intervenors sufficient time to develop alternative techniques.

In omnibus rate cases, and in the various proceedings that are litigated under the Commission's expedited rules, intervenors are generally unable to develop alternative models of postal cost or volume behavior within the compressed litigation window provided. Because the Postal Service has custody of virtually all of the relevant data, it has an unlimited opportunity to develop such models. This imbalance is a basic flaw in the functioning of the Act, which presumes that all interested parties to a hearing will be afforded adequate due process and procedural fairness. If intervenors do not have the basic data with which to develop models between rate or classification cases, their right to present an opposing case in a rate hearing is unnecessarily limited and more theoretical than real. By requiring that the basic datasets used to produce the Postal Service's routine financial reports be made available to others between rate cases, the updated rule helps restore basic due process rights to intervenors who wish to develop alternative cost, volume, and revenue estimating procedures on which to base rates.

Differences Between the Proposed Rule and the Final Rule

The Postal Service argues that releasing enough information between omnibus rate cases to allow its routine financial reports to be evaluated, or enough data to allow others to develop alternative models of cost or volume behavior, subverts, rather than facilitates the intended functioning of the Act. It seems to interpret the Postal Reorganization Act to grant it immunity from such activity between omnibus rate cases. The Postal Service is primarily concerned about documenting its annual CRA report, complaining that the rule would require it to provide as much supporting documentation for it as it provides in support of an omnibus rate case. In order to meet the Postal Service objections to the additional burden that the updated rule would impose, the Commission has pared back its requirements to the minimum that will still serve the basic purposes of the rule. The Commission has also incorporated some additional flexibility in meeting the requirements of the rule.

The proposed rule required the Postal Service to provide three general classes of files used to produce the current year's CRA—(1) all input datasets, (2) all processing programs used to attribute mail processing costs, and (3) all other processing programs that have changed since the most recently completed omnibus rate case. See proposed Rule 102(a), Attachment to the NPR, at page 1 of 4. The final rule narrows the first general class of files to input datasets that have changed since the last general rate case. It deletes the second general category of files, and retains the third general category. Under the final rule, datasets that have not changed, such as those from already-documented special studies, need not be provided. Similarly, under the final rule, processing programs used to attribute mail processing costs that have not changed need not be provided. See final Rule 102(a)(1).

The final rule is more flexible than the proposed rule with respect to datasets and processing programs that have been used for the first time to produce the CRA. If the Postal Service uses an estimation technique based on a new special study, the Postal Service may, under the final rule, choose to provide the Commission and the interested public with a short written narrative, or an oral briefing on that technique. The narrative or briefing should describe the data, the variables, and the analytical method used (such as the regression equation used). The purpose of the presentation would be to explain how the method was applied. The Postal Service may reserve its right to discuss the merits of its new method relative to alternative methods in the context of a formal hearing. After the written or oral presentation, the Postal Service may request a waiver of the requirement that input data and processing programs used to apply the new method be provided. If the presentation is sufficient to allow others to understand how the estimates affected by the new method were obtained, the requirement that the Postal Service provide the input data and processing programs used may be waived. See final Rule 102(a)(1)(ii).^[2]

In its NPR, the Commission noted that there are significant differences between the methods used by the Commission and the Postal Service to attribute mail processing costs to subclasses of mail. It also noted that the methods that the Postal Service uses to attribute mail processing costs are in greater flux than in other segments. The Commission felt that this made it harder to determine whether the Postal Service has correctly applied PRC-approved methods in updating the Cost and Revenue Analysis-PRC Version. Accordingly, in its NPR, the Commission proposed that the Postal Service include with the CRA all of the processing programs that it used to attribute mail processing costs. NPR at 4-5. Despite these complicating factors, the Commission has decided not to require more complete documentation of the attribution of mail processing costs than of other costs, in order to reduce the Postal Service's burden in complying with the Periodic Reporting Rule.

In addition to requiring the Postal Service to provide input data and processing programs that have changed since the most recently completed general rate case, the final rule requires that the Postal Service provide the spreadsheet workpapers (the “B workpapers”) that show how the CRA was developed. See final Rule 102(a)(1)(i). There are so many links and interactions built into these spreadsheets that an effort to separately identify portions that have changed from portions that have not is impractical. They are so essential to understanding how the summary estimates in the CRA were obtained that they need to be provided as an integrated whole.

Therefore, under the final rule, given current costing methods, documentation of the CRA should include the following:

(1) Spreadsheets supporting the CRA. (The “B” workpapers. In Docket No. R2001-1 these were found in USPS-LR-J-57. These should include the workpapers for Segment 14, and the Alaska Air Adjustment, that have customarily been provided under the rule.)

(2) The CRA model. This should include the files usually provided during an omnibus rate case to allow for the replication of all of the operations used by the Postal Service's COBOL CRA/Rollforward programs. These include the Manual Input Matrix, the “A” report matrix, and the “C” report matrix. The files that contain the operating “control strings,” that is, the instructions to the computer model that distribute the indirect costs to classes and subclasses of mail, should be included. These files are usually named “A,” “B,” “C,” “D,” and “F.” They represent the instructions to the model for the development of the “A” report and the “B” report (Factor Development Report). Title files containing the categories of mail and special services that are reported in the CRA, and the titles of all 1,600 components in the USPS CRA cost matrix, should be provided. (All of this material was provided in just eight of the over 100 files that made up USPS-LR-J-6 in Docket No. R2001-1.) The printouts of the CRA and the Cost Segment and Components report should be included, as has been customary under the rule.

(3) The output data file for the In-Office Cost System (IOCS). (This was found in file PRC00.SD2 in USPS-LR-J-10 in Docket No. R2001-1).

(4) The Segment 3 accrued cost pools. (These were found in USPS-LR-J-55, Table 1, in Docket No. R2001-1).

(5) Equipment and facility-related cost spreadsheets. These three spreadsheets show the equipment variabilities for equipment maintenance labor costs, equipment parts and supplies, and capital interest costs by type of mechanized operation. The spreadsheets also develop the inputs for the components that determine the space and space-related separations for some facility-related costs, such as custodial, fuel and utilities, and rents. The inputs are data from the special facility studies and other maintenance databases. (In Docket No. R2001-1, these files were identified as FY00equip.xls, Facilt.xls, and equipvar.xls, and were sponsored by witness Smith.)

(6) Output data file for the City Carrier Cost System (CCCS). (This was identified as “cityz.sd2” in USPS-LR-J-12 in Docket No. R2001-1.)

(7) Output data file for the Rural Carrier Cost System (RCCS). (This was provided as the “z” folder in USPS-LR-J-13 in Docket No. R2001-1).

(8) The National Mail Count for rural carriers.

Procedural History of the Rule

This rulemaking has had an unusual procedural history. The NPR was issued on January 8, 2003. It allowed approximately a month for filing public comments and two weeks to file reply comments—the standard period for these procedures. The NPR proposed that part of the required information be provided in a PC-compatible format, and suggested that an informal technical conference be held if the Postal Service anticipated problems complying with this requirement. At the Postal Service's request, a technical conference was held on March 11, 2003. Afterward, the Postal Service was given three weeks to file substantive comments on the proposed update that reflected the information gained at the technical conference. See PRC Order No. 1363. At the end of that period, the Postal Service then requested an additional five weeks to file its substantive comments so that its staff could confer with its Board of Governors.

Toward the end of that period, the Postal Service urged that it not be required to file its comments until well after July 31, 2003, the date on which the report of the President's Commission on the Postal Service was due. It asserted that the proposed update had major ramifications for the Postal Service and Commission, its prerogatives as a litigant in rate cases, and the confidentiality of its commercially sensitive information. It argued that these issues should not be addressed until after the recommendations of the President's Commission were made public. The report of the President's Commission, it said, would provide the appropriate context for discussing these issues. Motion of the United States Postal Service for Further Extension of Time to File Comments, June 6, 2003, at 2-3.

General Views of the Commenters

Apart from the Postal Service, all of those commenting on the NPR have participated as intervenors in omnibus rate cases. They agree that the proposed updates would improve their ability to understand and respond to an omnibus rate filing in the time allotted. They argue that with the increasing complexity of the Postal Service's omnibus rate filings, the “playing field” has become tilted heavily in favor of the Postal Service. They argue that they have so little time to understand and react to the tens of thousands pages of data and documentation that support the Postal Service's rate filing that they cannot digest it all and respond to it in the time allowed. They say that by having access to enough data and documentation to understand how the Postal Service's routine financial reports are put together each year, they will be less likely to be overwhelmed when the Postal Service files an omnibus rate request that employs similar types of information. The intervenors also point out that under the regulatory structure of the Postal Reorganization Act, the Commission is required to make independent recommendations on postal rates under the severe time pressures imposed by the Act. To be adequately prepared to process an omnibus postal service rate filing under these difficult conditions, they argue, the Commission must maintain its expertise on postal cost systems and financial forecasting between rate cases. In order to do this, they say, the Commission must have enough information to determine what data and methods the Postal Service is currently using to produce its routine financial reports.

The Postal Service does not deny that providing the routinely compiled financial information called for by the updated rule would give the Commission and the intervenors the ability to respond to Postal Service rate filings more quickly and more competently. In fact, this appears to be its main objection to the updated rule. It contends that this would upset the statutory “balance” between it and the remaining players in the postal regulatory scheme.

The Postal Service understands that when it withholds this information until it files a rate case, participants must spend so much of the 10-month period that is available to litigate a rate case reading and comprehending it that they have little time to prepare alternative rate proposals in response. The Postal Service does not consider this to be inconsistent with the Act. It emphasizes that the Act makes it the sole initiator of rate cases. In its view, this allows it to take all the time that it needs to prepare for litigation, and allows the Commission and the intervenors none. It insists that this procedural advantage is intended by the Act, and that it may withhold whatever financial information it wishes between rate cases in order to preserve it. It denies that it has any obligation to provide information between cases that would facilitate the Commission's performance of its functions during those cases, or would make it easier for intervenors or the public to comprehend or respond to its filings within the time constraints imposed by the statute. Substantive Comments at 15-17, 33-36.

The Postal Service argues that since the Act does not give the Commission any meaningful function to perform between rate cases, the Commission has no legitimate need for financial information during these interim periods. Therefore, it reasons, the Commission has no legitimate motive for seeking access to financial information between rate cases. It concludes that the Commission can only have ulterior motives for seeking information between rate cases, e.g., to conduct annual audits and investigations of the Postal Service, to gain “oversight responsibility,” and to indulge in “day-to-day monitoring of [its] detailed operations and finances.” Id. at 7, 11, 19, 22.

Besides serving ulterior motives, the Postal Service complains that the updated rule would force it to prepare rate-case style documentation between rate cases. It argues that this will infringe upon management's statutory right not to concern itself with rate issues between rate cases, and will infringe on management's duty to manage. Id. at 21-23. It contends that most of the CRA documentation required by the rule has commercial value which the Commission would be unwilling or unable to protect from public disclosure. Id. at 31-32. Finally, it argues, by seeking basic information needed to understand and analyze the CRA, the Commission is seeking to preempt the legislative reform process that the President entrusted to the Commission on the Postal Service. Id. at 20.

As the Postal Service now interprets the Act, between the rate cases that it files, the intervenors must avoid actions or thoughts that might relate to future rate or classification cases. Otherwise, the argument goes, they will nullify the litigation advantages that the Postal Service enjoys under the Act. Similarly, the Postal Service argues, the Commission must refrain from actions and thoughts that might help it prepare for future rate or classification cases. Otherwise, its collective mind will become contaminated. The updated rule, the Postal Service contends, seeks to circumvent these constraints that it infers from the structure of the Act. Id. at 15-17.

The Postal Service's portrayal of the updated rule as a newly-hatched plot by the Commission to circumvent the Act disregards the history of the Periodic Reporting Rule. As explained in more detail below, the updated rule meets the same standards, and is designed to accomplish the same objectives, as the original rule adopted 27 years ago by the Commission. At that time, the Commission explained that the rule had two main objectives: (1) to accelerate the discovery process during future rate and classification hearings, and (2) to enable all those in the postal regulatory arena, including the Commission and the intervenors, to study postal cost behavior between rate cases in order to improve the attribution of costs during rate cases.

What is novel with this rulemaking is the Postal Service's interpretation of the Act as mandating that the flow of financial information cease between rate cases. This runs counter to the Postal Service's historic view that periodic reporting of financial data between rate cases, while not mandated by the Act, is a legitimate way to make the processing of future cases more efficient by reducing the need for discovery. Docket No. RM76-5, Postal Service Response to PRC Order No. 141 (December 7, 1976) at 2. For 27 years, the Postal Service supplied the type of information called for by the rule without suggesting that its objectives were in violation of the Act.

The vast majority of the information previously required by the Periodic Reporting Rule has been willingly provided by the Postal Service. It accepted the modest additional burden of providing such information if it was likely to provide useful background information for processing future cases. It raised objections to only a few items, sometimes arguing that an item should not be included in the rule because it was not information “which could be employed for rate purposes.” Id. at 4. The views that the Postal Service has expressed in this rulemaking are the converse of its traditional view of the rule. Where it had traditionally considered items to be appropriate for inclusion in the rule only if they “could be employed for rate purposes,” it now considers items to be inappropriate for inclusion in the rule precisely because they “could be employed for rate purposes.”

The Commission did not include an elaborate justification for the update of the Periodic Reporting Rule in its NPR because it did not think that one would be necessary. The Commission assumed that the additional burden on the Postal Service of complying with the rule would be minor because the updated rule asks for only a small fraction of the information that the Postal Service provides with an omnibus rate filing. Of that small fraction, most is prepared each year by the Postal Service anyway, either to produce its own CRA, or to comply with rule 103. The Postal Service's right to litigate the merits of its procedures in a formal hearing is not infringed by the rule because the information required by the updated rule does not address the merits of, or justification for, the procedures that the Postal Service uses to produce its CRA. The Commission assumed that the commercial sensitivity of the information would not be a significant issue because the updated rule would require the types of information that the Postal Service has, in the past, freely disclosed to the public, both during and between omnibus rate cases. Because the Postal Service now challenges these assumptions, and misinterprets the Commission's motives for proposing these updates, the Commission will provide a detailed justification of the updated rule.

II. The Relationship of the Updated Periodic Reporting Rule to Legislative Reform

The Relevance of the President's Commission on Postal Service

The Postal Service has chosen to depict the update to the Commission's Periodic Reporting Rule as an attempt to fundamentally alter the “balance of power” that the Act strikes between the Postal Service and the Commission, focusing almost all of its arguments on the CRA documentation that the rule requires. Substantive Comments at 20. The other commenters share the Commission's view that the updated Periodic Reporting Rule is a legitimate, restrained exercise of its § 3603 authority under the Act whose effect is to only modestly increase the regularity with which the Postal Service would otherwise disclose this information.

Based on the false premise that the rule attempts to rewrite the Act, the Postal Service criticizes the Commission's decision to go ahead with the update, rather than wait for the dust from the President's Commission on Postal Service to settle. The President's Commission was organized in December, 2002, and was charged with recommending a solution to what appeared to be the Postal Service's stagnating volumes and mounting losses. The Postal Rate Commission proposed to update its rule in January 2003, one month after the President's Commission began its work. At that time, nothing was known about the direction that the recommendations of the President's Commission might take.

Due to a series of Postal Service requests for extensions of time to comment on the proposed updates, the rulemaking was still pending in June of this year. At that time, the Postal Service asked the Commission to suspend this rulemaking until after the report of the President's Commission was due to be issued, apparently so that the recommendations in that report could guide the deliberations of this rulemaking. With the unanimous support of all of the other commenters, the Commission declined the Postal Service's request for such a lengthy, additional delay, and ordered the Postal Service to file its substantive comments. The Postal Service responded with some indignation, accusing the Commission of, in effect, attempting to trump the legislative reform effort. Id. at 19-20.

One of the dominant themes in the report of the President's Commission is the need for greater transparency of Postal Service operations and finances. Its proposed regulatory scheme would allow the Postal Service to retain its monopoly over letter mail, at least initially, and give it great flexibility to set rates for competitive products. The President's Commission, however, made it clear that the price for combining monopoly power with pricing discretion over competitive products would be greatly strengthened regulatory oversight and accountability, entrusted to a new Postal Regulatory Board. First on the list of duties entrusted to the new Board would be the duty to “ensure the financial transparency of the Postal Service.” Report of the President's Commission on Postal Service, issued July 31, 2003, at 53. The report elaborates, at page 66:

The Commission believes that the Postal Service has a responsibility to the public to be transparent in its financial reporting. Given its important public mission and central role in the nation's economy, changes in Postal Service economic health should not come as a surprise to those responsible for or impacted by its performance.

* * * * *

As a unifying force in American commerce and society, and as a customer-financed government endeavor, the Postal Service should be setting the standard for financial transparency by which all other Federal entities are judged. While the Postal Service does, in many respects, conduct financial reporting over and above what is required of Federal agencies, it remains behind the level of disclosure offered by its corporate peers. [Emphasis in original.]

And at page 68:

In addition to SEC-like reporting, the Commission recommends that the Board of Directors be required to submit annually a detailed report to the Postal Regulatory Board on the financial viability of the institution, providing both significant financial insights as well as adequate explanation of related trends. The report should adhere to the 'no surprises' rule, ensuring that any major changes to the fiscal health of the institution are widely understood in advance, so appropriate responses can be anticipated and generated. The Commission further recommends that this report be made available to the public.

The new Postal Regulatory Board would be empowered to set baseline rates and price caps for non-competitive services, and empowered to review rates for competitive products to ensure that they are not cross-subsidized by non-competitive products. The Board would complete reviews of competitive product prices in 60 days. In order to make this streamlined rate regulation possible, the report recommends, at page 69:

For the Postal Regulatory Board to ensure financial transparency and make fully informed determinations on issues from rate ceilings to cross-subsidies, it must have access to the most reliable and current information possible. For this reason, the Commission recommends that the Postal Regulatory Board have the authority to request accurate and complete financial information from the Postal Service, including through the use of subpoena powers, if necessary, to obtain a thorough and reliable snapshot of Postal Service operations.

At page 67, the report concludes:

Where the Postal Service participates in markets also served by private industry, effective oversight is essential to ensure that monopoly revenues are not manipulated to the benefit of the Postal Service's competitive offerings. For this reason, the Commission recommends that the Postal Service periodically report on the allocation of costs among all products and services in accordance with form, content and timing requirements determined by the Postal Regulatory Board. [Emphasis added.]

The guidance provided by the President's Commission could hardly be more direct in its support of the approach taken by the updated Periodic Reporting Rule. Indeed, the report of the President's Commission recommends going much further in mandating transparency both in general, and in the area of rate regulation. The Commission, however, is aware that the report of the President's Commission is only relevant to the extent that its recommendations are enacted into law. This may never happen, or may take years to accomplish. The Postal Rate Commission's proposal to update its Periodic Reporting Rule was conceived independently of the President's Commission, whose recommendations had not yet been formulated. The Commission's motive in persevering with its proposal was not to steer the legislative reform effort in any particular direction, or to trump it. The Commission motive was, and is, to make the existing regulatory regime more effective in achieving the objectives Congress set out for it.

The Report of the General Accounting Office

On November 13, 2002, the General Accounting Office issued a report entitled “U.S. Postal Service Actions to Improve Its Financial Reporting.” GAO-03-26R Postal Financial Reporting. The report found that the Postal Service's financial reporting lacked the necessary transparency in general, and that its periodic reporting, in particular, was inadequate. At page 12 of the report, it states:

* * * it is clear from recently publicized problems in financial reporting that more detailed information and transparency are called for by both Congress and the public. Such transparency is critical for the Service because of the importance of its financial situation and the implications for stakeholders in making their own financial plans. These factors help support stakeholders' need for timely, accurate, and complete financial information that is provided on a consistent basis.

At page 14, it continues:

We acknowledge that the Postal Service provides a significant amount of information in its rate case filings; however, this information is provided only for rate-setting purposes, and rate cases are not filed on a regular cycle. Thus, rate case information does not provide stakeholders timely information about the Service's current financial condition and changes to its expected outlook.

* * * * *

As we noted, however, these periodic financial reports do not clearly explain changes in its financial condition, outlook, and results of operations, and have not always been readily available to the public.

Apart from the Postal Service, there appears to be a public consensus on the need for more complete periodic financial disclosure.

III. The Need for Updating the Periodic Reporting Rule

A. The History of the Rule

Historically, during omnibus rate cases, the Postal Service has attempted to support its rate requests with input data, spreadsheets, and documentation that is sufficiently detailed and complete to allow the behavior of postal costs, revenues, and volumes to be evaluated and understood by the interested public. Typically, several years elapse between rate cases. During those intervals, the Postal Service has provided the Commission and the public with summary financial reports that it generates at regular intervals for use by postal management. It has included documentation of significant portions of those reports in response to the Periodic Reporting Rule. The portion provided has been a small subset of the kind of documentation that the Postal Service provides during an omnibus rate case.

As noted, most of the Postal Service's objections to the updated rule are to the requirement that the Postal Service provide the input data and documentation that it uses to prepare its annual CRA report. Substantive Comments at 2. Each year this report summarizes, at the most general level, the results of the Postal Service's procedures that estimate the amount of costs caused by each subclass of mail, and the amount of revenue that each subclass earned. The process that produces the estimates in the CRA takes dollars from hundreds of subaccounts in the Postal Service's Books of Account and assigns them to one of hundreds of “functional” cost components. (Functional costs are viewed as economic costs). Costs in the various functional components are analyzed to see how they vary with mail volume. The volume variable part is then distributed to subclasses according to piece counts or other “distribution keys” that imply subclass causation.

The Postal Service's estimates of the costs and revenues generated by each subclass of mail are derived from the intricate rules that it uses to convert its accounting costs to functional costs, apply variability percentages to functional costs, and distribute the variable portion to subclasses. When submitted in rate cases, these are the baseline estimates underlying the rate proposals made by the Postal Service. With some adjustments, CRA estimates also provide the basis of the rate proposals of the intervenors, and the rate recommendations of the Commission.

When the Postal Service files an omnibus rate request, it includes spreadsheets and computer programs that contain the CRA's conversion, attribution, and distribution rules. This is its “cost attribution engine” described earlier in this order. These rules and their interactions are exceedingly complex. The input data, and the processing programs and spreadsheets showing how such rules are applied to the data, occupy the equivalent of tens of thousands of printed pages. Documentation of the myriad component parts of the process by which the CRA is produced is fragmentary. There is no single, coherent narrative explanation of the process to which an outside analyst could turn to understand, evaluate, and offer alternatives to the Postal Service's CRA. An outside analyst must rely primarily on a detailed study of processing program code and spreadsheet algorithms in order to discover how the CRA is developed. The analyst must make test runs replicating this largely automated process to confirm that his or her preliminary understanding of it is correct. The expertise necessary to evaluate the methods by which the Postal Service produces the CRA, or to develop alternatives to it, must be accumulated over many years. Despite attempts over the more than 30 years that the Commission has been processing omnibus rate requests, no outside party has been able to replicate the full CRA upon which the Postal Service bases its rate requests, or develop a comprehensive alternative to it, within the 10-month window that the Act provides for litigating a rate case. Under these circumstances, even with the voluminous documentation provided by the Postal Service during omnibus rate cases, much of the process by which it produces its CRA has, of necessity, been accepted on faith by intervenors and the Commission.

Since the methods that the Postal Service uses to produce the CRA continually evolve in minor, and sometimes major, ways, the CRA presents a “moving target” for outside analysts. Each time an omnibus rate request is filed, those on the outside of the CRA development process (the Commission and affected public) struggle to grasp these procedures and track their evolution. Even though this task can consume a majority of the 10-month period that the Act allots for processing a rate case, it must be completed before the intervenors and the Commission can effectively evaluate or respond to the Postal Service's rate proposals.

To mitigate this problem, the Commission implemented its Periodic Reporting Rule in 1976. At that time it explained the purposes of the rule and the policies underlying its decision about what the rule should contain as follows:

Currently, a majority of the data which the Commission receives from the United States Postal Service (Postal Service) is obtained only when a rate request is pending before the Commission. The present requirements of rule 54 (39 CFR 3001.54) and the Commission's regulations relating to interrogatory procedures and the discovery process have enabled the Commission and the participants in rate proceedings to obtain much of the data required to evaluate a request for increased postage rates and fees. However, the existing method of obtaining data, especially as regards discovery and the interrogatory process, is necessarily conducted on an ad hoc basis and is subject to all the pressures and exigencies of a rate case environment. If the Commission is to better fulfill its statutory responsibilities-particularly with respect to the Postal Reorganization Act's directive that we expedite our proceedings consistent with procedural fairness to the parties appearing in them [39 U.S.C. 3624(b)]—it must be continually and fully familiar with these data. To do this the Commission believes that it must improve the present method of obtaining data from the Service.

At the present time, the Commission is aware of the existence of a number of reports routinely compiled by the Postal Service. The Postal Service also compiles manuals and handbooks which are necessary to understanding and analyzing such reports. It would appear that these documents would be useful for the purpose of evaluating Postal Service operations which are the subject of cost analyses presented in proceedings before the Commission. If these documents were to be obtained by the Commission as they were completed (and were made publicly available at the Commission's offices) it is anticipated that the Commission and the numerous interested parties appearing in our formal proceedings would then have an opportunity to evaluate the data contained in the documents on an ongoing basis rather than solely during a rate proceeding.

In addition to providing the Commission with a better opportunity for keeping abreast of the changing factors which will affect the execution of its regulatory functions, other benefits are likely to result if these reports were to be made available to the Commission. Since these data are necessary for evaluating a rate request, their early accessibility may aid in expediting rate proceedings. Relying solely on interrogatories and the discovery process to obtain information consumes time, both because data must initially be requested of the Service and, thereafter, additional time is expended while the Service responds. If the data which are the subject of this rulemaking were on file with the Commission, the time needed by the Commission and the parties would likely be reduced because of the ready availability of information.

Notice of Proposed Rulemaking issued April 5, 1976, in Docket No. RM76-5 at 2-4.

If the data currently made available to the Commission and interested persons were made available on an ongoing basis, rather than solely during the course of a rate proceeding, substantial benefits would inure to the Commission and the parties. The proposed rules will provide current data which will assist the Commission in keeping fully apprized of developing circumstances having an effect on its regulatory functions. Additionally, continued access to the data will assist interested members of the public in more thoroughly evaluating a filing of the Service and making alternative presentations within the time constraints imposed by the statutory directive that Commission proceedings be conducted with the “utmost expedition consistent with procedural fairness to the parties.” (39 U.S.C. § 3624). The reports, documents, and other data sources which are being made a part of the Commission's periodic reporting system will aid in achieving these objectives.

The data sources which the Commission is now including in this amendment to the rules of practice have been evaluated on the basis of (1) the demonstrated utility of the data source, and (2) the burden imposed on the Service in filing the particular information. Although the information sources covered by our new rules do not include all the reports and documents proposed by the parties, the Commission is not foreclosing the possibility of the later inclusion of some or all of these items. As additional information is demonstrated to be useful, the underlying sources of information will be included in the data reporting system except where inclusion would impose an undue burden on the Service. So that interested parties may have the opportunity to analyze and experiment with additional information, even when there is no case in progress, the Service should provide access to these additional information sources. The Commission believes that where the information is available, its use on an experimental basis will be very helpful in determining its utility. [Footnote omitted.]

PRC Order No. 141 (October 21, 1976) at 3-4.

The initial version of the Periodic Reporting Rule emphasized accounting and other types of financial information that were likely to be useful in analyzing the behavior of the Postal Service's revenue requirement over time. It did not emphasize information on attributing costs to mail classes because analysis of Postal Service costs was still in its infancy. The process for attributing accounting costs to mail classes did not approach the complexity of the current CRA. Cost data collection systems and models of postal cost behavior were in considerable flux. Most of them were being developed on an arcane data processing platform that made it technically difficult for the Commission and the affected public to decipher and analyze. Primarily because attribution analysis was considered to be inadequate, the Postal Service, the Commission, and the affected public were all exploring ways to improve attribution methods. Facilitating such research with a view to speeding up the resolution of cost attribution issues in rate cases was among the primary goals of the Periodic Reporting Rule. Id. at 3-4, 15.

In contrast to its current attitude, the Postal Service's response was accommodating. It did not challenge the legitimacy of providing basic financial data and documentation to facilitate independent research of postal cost behavior between rate cases. It did not assert the commercial sensitivity of the cost data that the Commission or the intervenors proposed to include in the rule, except where data were facility-specific or customer-specific. (The Commission readily accommodated this concern in its initial version of the rule.)

From the beginning, the Commission's explicit policy has been to minimize the burden of the Periodic Reporting Rule on the Postal Service by limiting it to information that the Commission or the affected public was likely to use. With respect to cost information, the initial version of the rule asked primarily for summary-level cost reports [the precursors of the Cost and Revenue Analysis (CRA) and the Cost Segments and Components (CSC) reports] ^[3] since the technical obstacles referred to above made it difficult for the Commission or the public to use the input data and documentation underlying the Postal Service's standard cost attribution reports.

By the mid-1980s, some Postal Service cost data collection systems had matured, and cost attribution analysis had grown more complex, notably in the method by which attributable costs were distributed to mail classes in the mail processing and transportation areas. Adjusting to these developments, the Commission updated its Periodic Reporting Rule to require supporting documentation of these methods (the LIOCATT in mail processing and workpapers 31 and 57 in transportation). See PRC Order No. 697 (June 27, 1986) at 7. The rule did not seek input data in these costing areas because technical problems still prevented the Commission from using input data in the form in which it was reported. In other major cost centers, such as carrier street time costs, ongoing data collection systems had not yet stabilized. Attribution of these costs depended primarily on ad hoc studies that had already been publicly documented in the course of rate hearings, rather than on the analysis of regularly-collected data. Because regularly-collected data on carrier street time cost played a lesser role in modeling carrier costs, the rule did not require that carrier cost data be periodically reported.

The Periodic Reporting Rule has not been modified in 17 years. Over that time, the Postal Service's financial reporting systems have undergone major changes. Updating of the Periodic Reporting Rule to reflect those changes is long overdue. The sophistication of cost attribution methods has grown dramatically since the rule was last modified. The Postal Service has introduced elaborate cost variability models in the mail processing, transportation, and carrier cost areas. Also in each area, it has developed new, more complex methods of distributing attributable costs to subclasses. The ongoing data collection systems that the Postal Service used to develop these new attribution models and distribution methods were not used for these purposes, or did not exist, when the rule was last updated. These include “MODS” in the area of mail processing costs, “TRACS” in the area of transportation costs, and “CCCS” and “RCCS” in the area of delivery carrier costs. As the complexity of the Postal Service's cost attribution methods has grown, the need to document them in order to competently interpret them has grown.^[4] Because these new models and methods use new data collection systems, the rule must be updated to include the new data systems if the Commission and the affected public are to understand how they are used to produce the CRA.

A primary reason that the Commission was slow to update its Periodic Reporting Rule to include this new cost data was that the capability to use this information only became available gradually. For much of this time, the Commission wrestled with Postal Service datasets and programs developed on a mainframe COBOL platform. Despite hiring a series of service bureaus and consultants who specialized in this data processing platform, the Commission and participants in rate hearings were generally not successful in reading, understanding, and using the datasets and programs that the Postal Service developed on this platform. It was not until the mid-1990s, when the Postal Service began providing some of this information in the mainframe SAS language, that the Commission's staff and some hearing participants were able to convert such information to PC-compatible versions that they could read, understand, and use.

Although the Commission, by the mid-1990s, was beginning to acquire the technical capacity to use the data and programs that underlie the Postal Service's periodic cost reports, it did not update the Periodic Reporting Rule to reflect its technical progress. This is because the Postal Service had been providing the information needed on an annual basis anyway, apart from the rule. It provided the basic data and documentation underlying the CRA each year from FY 1995 through FY 2000. Sometimes it provided this information in support of a rate or classification request. Other times it provided it voluntarily, simply to be helpful.^[5]

Since the most recent omnibus rate case was settled, however, the prospect for continuing to get an adequately documented CRA each year has dimmed. Passage of Pub. L. 108-18, which dramatically reduces the Postal Service's contribution to the Civil Service Retirement Fund, led to a Postal Service promise not to increase rates until 2006. Consequently, the Postal Service is unlikely to file a documented version of the CRA in support of an omnibus rate request for four years—from FY 2001 through FY 2004. The Postal Service is signaling that it will not voluntarily submit such information in the future. Unless the Periodic Reporting Rule is updated to seek a moderate level of documentation of the Postal Service's CRA each year, the outside world will not be able to competently interpret the CRA for up to four years. The Commission and the intervenors do not believe that the regulatory scheme established by the Postal Reorganization Act can function as Congress intended if they are to be kept in the dark for up to four years. By updating its Periodic Reporting Rule, the Commission will eliminate long blackout periods of this kind.

B. The Postal Service's Scope and Burden Objections

The preceding discussion of the history of the Periodic Reporting Rule and the considerations that shaped the update to the rule respond to the Postal Service's speculations that the update was prompted by an array of improper motives. As the preceding discussion makes clear, the Commission's objective in adopting the rule has remained the same over the 27-year life of the rule—to help the Commission perform its statutory functions more quickly and efficiently.

To do that, the rule directs the Postal Service to provide current-year financial reports summarizing the Postal Service's financial results, with enough mid-level documentation to allow the Commission and the affected public to competently interpret them. The rule also directs the Postal Service to provide intermediate-level datasets that will allow outside analysis of postal cost and volume behavior to continue between omnibus rate cases. While the information sought is not case-specific, it facilitates the processing of future rate and classification cases by providing essential technical background for evaluating the kind of issues that typically arise in such cases.

The information provided under the rule makes the Commission and the interested public better prepared to process rate and classification cases. As explained above, the rule needs to be updated because the Postal Service has made major changes to the way it estimates its costs and revenues over the past 17 years, and the Commission and the public have developed the technical capability to interpret and use the information supplied. The rule strikes a reasonable balance between these benefits and the added burden on the Postal Service of providing this additional information. As in the past, the Commission is willing to make appropriate arrangements to protect information that the Postal Service believes to be commercially sensitive.

The Postal Service expresses concern that the Commission has ulterior motives for seeking to update the rule. These include a desire to change its institutional relationship with the Postal Service, to arrogate to itself auditing, supervisory, and managing functions reserved to others under the Postal Reorganization Act, and to hijack the legislative reform process. Substantive Comments at 2-3. This questioning of the Commission's motives proceeds from the premise that the amount of data that the updated rule would require the Postal Service to provide is far out of proportion to its needs. This premise reflects two beliefs—that the updated rule requires documentation of the CRA equivalent to that required in an omnibus rate case, and that the Commission has no need for financial information unless it is actively processing an omnibus rate request.

Scope Arguments Based on the Wording of the Rule

When the Postal Service asserts that the updated rule requires CRA documentation on the same scale and scope as it provides in an omnibus rate case it grossly mischaracterizes the requirements of the rule. It finds support for its assertion in the “very broad” wording of the rule, quoting the preamble to proposed rule 102(a)(1):

All input data, all processing programs that have changed since the most recently completed general rate proceeding, and all computer programs used to attribute mail processing costs to subclasses, if they are used to produce the Cost and Revenue Analysis Report (CRA).

Without analyzing this language, it asserts that the proposed rule could

potentially call for production of virtually all information used in the production of the CRA report, from secondary, tertiary, and lower inputs to the CRA model and its inputs to raw data collected by the Postal Service's data collection systems.

Substantive Comments at 3.

It should be made clear at the outset that the CRA deals with only half of the costing issues that are addressed in detail in an omnibus rate case. The CRA summarizes the Postal Service's estimates of attributable costs by cost segment and by subclass. Equally important to recommending a comprehensive set of rates, and equally detailed, are the Postal Service's cost avoidance estimates upon which hundreds of worksharing discounts are based. The Commission's Periodic Reporting Rule does not require any reports or documentation that relate to this half of the Postal Service's cost presentation in an omnibus rate case.

Focusing on the various levels of inputs into the CRA model, as the Postal Service has done, helps demonstrate how it has exaggerated the scope of the proposed rule compared to the costing documentation provided in omnibus rate cases. To organize the discussion, the Commission will characterize the Postal Service's CRA documentation as consisting of the following six layers, or levels:

Level One—the programs that derive distribution keys for indirectly attributable costs and distributes them to subclasses

Level Two—the spreadsheets that calculate directly attributable costs and distributes them to subclasses

Level Three—cost attribution models

Level Four—input datasets

Level Five—data assembling, editing, and structuring techniques

Level Six—raw data

Omnibus rate cases involve formal hearings governed by the rules of evidence. Under those rules, the Postal Service is required to “lay a foundation” for the results of statistical or scientific studies that it wants to use to support its proposed rates. To lay the required foundation, it must start with the raw data it used and describe how that information was collected, edited, and structured before an estimating technique was applied to it.

For sake of discussion, the Commission will label raw data as Level Six documentation and the various manipulations that convert raw data into input data as Level Five documentation. The Commission will label “input data” as Level Four documentation. Input data are generally understood to mean data to which an estimating technique or model has been applied, which is its intended definition in the Periodic Reporting Rule. By specifying “input data,” the rule eliminates foundational information of the kind described above (Level Six and Level Five documentation) from its scope. The final rule is further narrowed to input data that have changed since the most recent omnibus rate case was completed. This eliminates input data collected as part of special studies that have already been reviewed in an omnibus rate case.

Because Level Five and Level Six documentation are not required by the rule, it is substantially narrower in scope than full rate case CRA documentation. Level Five and Six documentation can make up a large part of the burden of documenting the CRA in an omnibus case.^[6]

In the development of the CRA, input data are fed into spreadsheet, statistical, or econometric models of postal cost behavior to identify costs that are caused by particular classes of mail. The models themselves, including the theories upon which they are based, the definitions of the variables, the equations or other analytical techniques used, and the results, may be labeled Level Three documentation for purposes of this discussion. Typically, these models attempt to find the degree to which particular segment or component costs vary with volume, estimating a volume variability percentage or “factor” for those costs. Variable costs are distributed to subclasses of mail in proportion to their relative piece volume, cubic volume, or other cost-driving characteristic. The Postal Service calls the relative subclass shares of a given cost characteristic a “distribution key.” Level Three documentation sometimes shows how distribution keys were applied to volume variable costs to distribute them to subclasses. The updated rule requires the Postal Service to provide only a small subset of the Level Three documentation that it would provide in an omnibus rate case, i.e., the processing programs that have changed since the last general rate case.

In an omnibus rate case, Level Three documentation is by far the most burdensome and time-consuming kind to produce. It usually requires a narrative explanation and defense of the theory, the variables, the equation specification, the research of alternative estimation procedures and the reasons for rejecting them, and the validity of the results. The fact that the Periodic Reporting Rule requires only input data and processing programs means that it altogether avoids the need to justify and defend any aspect of the CRA development process. Furthermore, because the final rule applies only to processing programs that have changed since the most recently completed general rate case, most Level Three documentation is eliminated from the scope of the rule. This is because most attribution models and distribution techniques do not change from one rate case to the next. These two considerations are the most important reasons that, with respect to documenting the CRA, the burden of complying with the rule is a small fraction of the burden that the Postal Service bears in an omnibus rate case.^[7]

Level Two documentation consists of a workbook called “I-Forms” and Excel spreadsheets called the “B workpapers.” In Level Three documentation, component costs are modeled to see what portion varies directly with volume.^[8] The Postal Service typically collects these variability factors in a workbook known as “I-Forms.” Excel spreadsheets known as the “B workpapers” take variability factors from the “I-Forms” and apply them, in proper sequence, to the accrued costs of the appropriate components to obtain attributable costs for those components. Typically, the B workpapers also distribute a component's attributable costs to subclasses of mail, according to subclass shares of piece volume or some other cost-causing factor.

The development steps documented in Level Two are key steps in producing the CRA. The rules that the B workpaper spreadsheets apply summarize the Postal Service's cost attribution methods, and provide insight into the causes of trends in postal cost behavior. These rules are exceedingly intricate, and are continuing to evolve. This makes it difficult for an outside analyst to remain expert on this phase of the production of the CRA without current Level Two documentation. The Postal Service prepares Level Two CRA documentation each year when it produces the CRA-USPS Version. It prepares the same Level Two documentation when it produces the CRA-PRC Version as part of its obligation to facilitate production of the international mail study. Because it prepares Level Two documentation for both versions anyway, and preparing it can be done automatically with little effort, providing Level Two documentation for the Periodic Reporting Rule should not impose an additional burden on the Postal Service of any significance. For these reasons, the Periodic Reporting Rule requires the Postal Service to provide essentially the same Level Two documentation of the CRA each year that it provides in an omnibus rate case. See Rule 102(a)(1)(i).

Level Two documentation shows primarily how the Postal Service estimates subclass shares of costs that vary directly with volume. These estimates become direct inputs (what the Postal Service calls the “Manual Inputs”) into the “CRA model.” The CRA model is a mainframe COBOL program that distributes indirectly attributable costs to subclasses in the same proportions as the Manual Inputs to which they relate.^[9] The Level One CRA documentation shows how the CRA model does this.

The subclass shares of directly attributable costs embodied in the Manual Inputs are fed into the CRA model to estimate total attributable costs and cost coverages by subclass. If outside analysts do not have access to the B workpapers that show how the Manual Inputs were calculated, they are unable to interpret or analyze the Postal Service's estimates of subclass attributable costs and cost coverages. They must simply take these summary estimates “on faith.”

In the Level One CRA documentation, the Manual Inputs perform roughly the same function that the “I-Forms” perform in Level Two. In Level One CRA documentation, “control strings” perform roughly the same function that the spreadsheets perform in Level Two. The control strings apply intricate relationship rules to the Manual Inputs to construct hundreds of distinct keys for distributing indirectly attributable costs to subclasses. The CRA model then aggregates these subclass shares of directly attributable, and indirectly attributable costs.

With respect to space-related costs, such as rent, fuel, and utilities, the CRA model does more comprehensive calculations, calculating subclass shares of directly attributable, as well as indirectly attributable costs. Directly attributable space-related costs are not calculated in the B workpaper spreadsheets. There are distinct variability factors for many finely disaggregated activities that drive space costs, and there are many keys constructed from other keys that are used to distribute these costs. Because of this complexity, the Postal Service has used computer programs, rather than spreadsheets, to perform these calculations.

Like the B workpaper spreadsheets in Level Two, the CRA model is essential to understanding how the Postal Service arrived at its estimated subclass shares of attributable costs. Because of the numerous links and interrelationships embodied in its control strings, the CRA model needs to be provided as an integrated whole. Like the B workpaper spreadsheets, the intricate rules that the CRA model applies are continually being refined. These rules are exceedingly intricate, and evolve continually in minor, and sometimes major ways. For example, in its comments, the Postal Service announced that it has incorporated the results of a special facilities cost study into the FY 2002 CRA that would take it a year to document. Substantive Comments at 22. For an outside analyst to remain expert on the CRA model, current-year Level One (and Level Two) documentation of the model must be provided.

Each year, the Postal Service produces Level One documentation (the Manual Inputs and the Control Strings) for its own purposes when it produces the CRA-USPS Version. Each year, it produces Level One documentation for the CRA-PRC Version as part of its obligation to facilitate the international mail study. The CRA model is almost entirely automated. Because it prepares Level One documentation for both versions anyway, and preparing it can be done automatically with little effort, providing Level One documentation for the Periodic Reporting Rule should not impose an additional burden on the Postal Service of any significance. For these reasons, the rule requires the Postal Service to provide Level One CRA documentation each year that is comparable to that provided for the base year in an omnibus rate case.

Of the six levels of CRA documentation that the Postal Service provides in an omnibus rate case, the Periodic Reporting Rule will, in a typical year, require Level One, Level Two, and Level Four documentation, almost exclusively. Level One and Level Two documentation is not burdensome for the Postal Service to provide, since its production is almost entirely automated, and the Postal Service produces it each year anyway, for reasons apart from the rule. The burden of providing Level Four documentation should be minor, too. The input databases have already been produced, since they are an indispensable step in producing the CRA, and their production is automated. In providing some Level Four data, the Postal Service faces the extra task of masking the identifying label for data that are facility specific or customer specific. This, however, should not be significant, since it, too, is easily automated.

The only significant burden of complying with the rule that the Postal Service would not otherwise bear is that involved in providing Level Three documentation (the processing programs used in its component cost variability models). But it only needs to provide a minority of those programs used to produce the CRA (those that have changed since the last rate case), and it need not provide narrative explanations or justifications. The Postal Service thereby avoids most of the burden that it would encounter providing Level Three documentation in an omnibus rate case.

For all of the above reasons, the Postal Service grossly mischaracterizes the Periodic Reporting Rule when it asserts that it will require CRA documentation on the “scale and scope” of an omnibus rate case. Id. at 4.

Scope Arguments Based on the List in the NPR

As noted earlier, the Postal Service's principal support for this gross mischaracterization is an entirely non-analytical reference to the “very broad” language of the rule. It also cites a list of information at page 5 of the NPR, apparently to prove that the rule would require CRA documentation equal to that submitted in an omnibus rate case. Ibid. Here, too, it offers no analysis that explains how the contents of the list might support this assertion.

The 11 items on the list do not support the Postal Service's assertions that the requirements of the rule are open ended and comparable to an omnibus rate case in scope and scale. The list illustrates the Level One documentation required by the proposed rule (the “CRA Model” itself) with item 2. It illustrates the Level Two documentation required by the proposed rule (B workpaper spreadsheets and their equivalent) with items 1, 5, 10 and 11. It illustrates the Level Four documentation required by the proposed rule (input datasets) with items 3, 6 and 9. As noted above, the Postal Service would have to produce all of these items anyway for reasons apart from the rule, and their production is almost entirely automated, and therefore not a significant burden to provide.

The Level Three documentation required by the rule (analytical studies supporting variability estimates or distribution techniques) potentially would require some significant additional burden in the rare case that a special study was used to produce the CRA before being vetted in an omnibus rate case. Even then, the documentation would be much less burdensome to provide than the documentation required in an omnibus rate case because the Postal Service need not include a narrative explanation and defense of the study, such as that which it would submit as sponsoring testimony in an omnibus rate case. The final version of the rule gives the Postal Service additional flexibility that it would not have in the context of an omnibus rate case. The Postal Service may provide an abbreviated written or oral description of the study, which should include a description of the theory, the data, and estimation technique used. It may then ask for a waiver of the requirement that it provide the underlying dataset and processing programs. If the description is sufficient to allow others to evaluate the resulting estimate at a general level, the Commission could grant a waiver. If the Postal Service concludes that an abbreviated narrative description would be burdensome to provide, it may provide the input data and the processing programs, and let them speak for themselves. In its length and its scope, the narrative presentation might resemble the kind of informal technical conference that is occasionally used in a rate case to acquaint litigants with the basic outlines of a complex new study. It would not, however, involve testimony, discovery, or cross-examination, which are the significant burdens associated with litigating a rate case. It should be borne in mind that the additional burden of documenting a special study this way for the Periodic Reporting Rule would be zero, since the study would eventually have to be fully explained in an omnibus rate case.

Three items on the list on page 5 of the NPR (items 4, 7 and 8) illustrate the kind of Level Three CRA documentation that the proposed rule would require. These are special studies whose theory, variable definitions, and estimating techniques haven't changed from the most recent general rate case, but the accuracy of their results requires the use of updated input from routine data collection systems.^[10] Therefore, the Commission needs documentation related to these special studies that allows it to determine whether current data have, in fact, been used to produce the current-year CRA. This need relates primarily to the CRA-PRC Version, since the Postal Service does not audit that version as carefully as it audits the CRA-USPS Version. The Service's outside auditors do not review this document.

Item 7 on the list indicates that under current circumstances, where Docket No. R2000-1 serves as the most recent fully-litigated rate case, the proposed rule would extend to the spreadsheets and programs used to estimate load-time variabilities in the CRA. To get accurate estimates of variable load-time costs, it is necessary to calibrate the Load Time Variability model with current-year data from the City Carrier Cost System (CCCS). The Postal Service purported to estimate variable load time costs according to Commission-approved methods in the FY 2001 International CRA-PRC Version, which it provided to help the Commission produce its FY 2001 international mail report. (In such reports, the Commission must determine if the Postal Service's attributable cost estimates for international mail categories accurately reflect Commission-approved methods.) Because the Commission could not replicate the estimates using the methods approved in the last full rate case, it asked for the processing programs to see if it could determine why. The Postal Service provided new spreadsheets in place of the SAS programs it had been using in prior rate proceedings.

The Commission could not decipher the undocumented spreadsheets. Because the Commission did not have access to current-year CCCS data, it could not run the established programs to diagnose the problem itself.^[11] Consequently, the Commission could not determine whether the Postal Service had used current-year data to produce the current-year CRA. Later, through cumbersome trial and error procedures, the Commission was able to decipher the Postal Service's new spreadsheets, and determine that the Postal Service had simply plugged in an obsolete variability factor instead of using current data to update the load time variability model. This illustrates why the Commission needs access to input data and processing programs that have changed since the most recently completed rate case, if it is to be able to evaluate the CRA.

Item 8 on the list indicates that the Postal Service should provide “the underlying route-type data” needed to produce the in-office worksheets in the B workpapers. In the past, the Postal Service used the IOCS tally information compiled in the LIOCATT to distribute mixed mail sorting costs incurred at delivery units to subclasses. Because the Postal Service changed the processing programs that it uses to perform this task, this item was included in the list. The Commission recognizes, however, that the changes were documented in the most recently completed rate proceeding. As a result, the Periodic Reporting Rule need not extend to this item. If the Commission receives the IOCS input data, it will not need these processing programs to competently evaluate the distribution of in-office mail sorting costs to subclasses.

Item 4 on the list would require the Postal Service to provide the MODS-based costing spreadsheets needed to produce output for the B workpapers. While the processing programs used to attribute mail processing costs to subclasses were specifically required under proposed Rule 102(a)(1), they are not included in the final version of the rule. Many of those programs, however, change from year to year due to additions or deletions of activity codes or finance numbers. Also, because some programs use hard-coded numbers to compute distribution keys, they need to be updated each year. Consequently, the final rule would still require the Postal Service to provide many of the MODS-based programs in item 4. It may be more practical for the Postal Service to submit them all, rather than to attempt to segregate out the ones that have changed. The final rule allows the Postal Service this option.

The relatively recent switch from the LIOCATT-based system to the MODS-based system is a fundamental shift of methods governing a major portion of the Postal Service's overall operations. The Postal Service is apparently still making adjustments to the estimation methods that it uses to produce the CRA in related areas to bring them into conformity with the fundamental shift. For these reasons, the Commission is likely to need the programs implementing the MODS-based method for distributing mail processing costs to subclasses if it is to stay abreast of these developments in cost attribution, and retain its ability to competently interpret the CRA.

Scope Arguments That Assume Improper Commission Motives

The Postal Service seems to be aware that neither the wording of the Periodic Reporting Rule, nor the list of examples of what it would cover, supports its assertion that it would require as much CRA documentation as would be required in a full-blown rate case. The main support that it offers for this assertion is its hypothesis that the Commission has an array of improper motives for updating the rule (e.g., to supplant the Postal Service's auditors, to take day-to-day monitoring away from postal management, to conduct rate cases off the record, to upset the legislative balance, etc.). Given such motives, it claims, it is “inevitable” that the Commission will ignore the limits of its rule and seek the full-blown rate case documentation. Substantive Comments at 22.

As an example of the full-blown rate case documentation the Postal Service says the rule will require, it cites two special studies that it says it has relied on to develop the CRA, neither of which has been publicly disclosed. One estimates attributable carrier costs, the other facility-related attributable costs. It argues that “[u]nder the proposed rule, any such study would need to be extensively documented in order for its new data and methods to be understandable to, and usable by the Commission.” Id. at 22. It estimates that it would take 6 person-months to adequately document the carrier cost study, and 12 person-months to adequately document the facility costs study. Ibid.

The Postal Service does not attempt to explain why it would take this much additional time to document such studies, and no plausible explanation is readily apparent. These studies, presumably, have already been documented sufficiently by their authors to convince upper management that they provide a sound basis for one of its most important routine financial reports. The rule, however, only requires that the Postal Service provide the input data and processing programs used to perform the study. It is not plausible that a study could have received this level of scrutiny and acceptance unless there was already in existence a set of input data and processing programs that the author could locate and provide in less than six (or twelve) months. Similarly, it is not plausible that it would take six (or twelve) months for the author of the study to prepare a morning's briefing on the study for interested parties, if the Postal Service were to choose that option.

In order to have any credibility, these burden estimates have to assume that the Commission will ignore the limits of the rule and seek full rate case documentation of the study, including detailed narrative testimony that establishes a foundation for study results and defends the theory, the estimating techniques, and the robustness of the results. It can be seen that, with respect to Level Three CRA documentation especially, there is an enormous gap between the relatively insignificant additional burden of complying with the rule and what the Postal Service spends its time and energy opposing.

The Postal Service contends that if it did not use any new special studies in the CRA, complying with the rule would require 781/2 additional person-days. It does not explain why it would take this much effort, since it already prepares this documentation for reasons apart from the rule, and its preparation is (or could be) almost entirely automated.

Even if this estimate were accurate, however, it should be kept in perspective. When the Postal Service prepares an omnibus rate request, by its own account, it produces tens of thousands of pages of documentation, data, and testimony, most of which is devoted to explanation of its cost, revenue, and volume estimates. The burden of producing this information is so heavy, according to the Postal Service, that it must begin its preparation approximately six months in advance in order to file by its target date. Substantive Comments at 21. Over that period, the Postal Service assigns a host of consultants and its own professional staff to this task. The hours that the Postal Service says it needs to comply with the CRA documentation requirements of the Periodic Reporting Rule (in the normal circumstance where it is not based on new special studies) is a tiny fraction of the burden of documenting an omnibus rate filing.^[12]

Yet the Postal Service's comments are replete with assertions designed to leave the impression that the rule would impose a burden that is comparable to the burden of preparing additional omnibus rate filings. Id. at 4, 7, 15, 21. The Postal Service's arguments about the balance of powers between the Postal Service and the Commission being “fundamentally altered” due to a massive increase in the Postal Service's “regulatory overhead,” interference with postal management's ability to focus, conducting rate cases off the record, etc., evaporate when its gross mischaracterization of the burden of complying with the rule is exposed.

The Postal Service becomes particularly apocalyptic about the prospect of answering informal questions about the way it produces the CRA. It asserts that comments were made at the technical conference held on April 3, 2003, indicating that the Commission staff “envisioned the establishment of a process” whereby the Commission, and others, could direct questions to the Postal Service concerning the documentation that it provides under the rule. Id. at 7. According to the Postal Service, this raises “the possibility of an open-ended, ‘perpetual’ rate-case.” Id. at 15. In the Postal Service's mind, this possibility then morphs into the specter of

year-round rate-making style data-production, documentation, and perhaps more significantly, ongoing inquiries by the Commission, Postal Service competitors, and any other interested party with the time and resources necessary to pursue such activities.

Id. at 18. This leads the Postal Service to warn

[i]f such pseudo-discovery were similar to that encountered by the Postal Service in omnibus rate cases, one would expect the burden associated with responding to questions on new cost studies to be very large indeed. The very open-endedness of such extra-record questioning not only raises serious concerns regarding the potential burden involved, but reinforces the fundamental objection that the Postal Service and the Commission should not be spending their time and resources devising ratemaking procedures that not only are unsupported by our governing statue, but actually conflict with that statute.

Id. at 23-24.

Never, in recent memory, has the Postal Service tried to make such a grandiose mountain out of such an insignificant molehill. The Commission's staff entered the technical conference without having discussed, let alone taken a position, on the question of whether there should be informal questioning of the Postal Service staff under the rule. In the memory of the Commission's staff, it was another attendee who asked whether such questioning would be compatible with the rule, and the response of the Commission staff was that it saw no incompatibility with the rule, and it had no objection.

The Commission sees no problem with continuing the same helpful practice that has been followed for decades by the staffs of both the Postal Service and the Commission. Between rate cases, on rare occasion, a member of one staff, for example, would spot what appears to be error in a spreadsheet, make a call to the other staff, and ask if it was, in fact, an error. Someone on the other staff would typically investigate and respond informally with a corrected spreadsheet, or some brief explanation of the apparent error. The Postal Service staff occasionally does this after reviewing the workpapers that the Commission provides to explain the technical aspects of its Recommended Decisions. The Commission's staff occasionally does this after the Postal Service provides a periodic report. The Commission's staff saw no reason not to continue this practice. The Commission has never suggested launching general “procedures” for “pseudo discovery” between rate cases under the auspices of the Periodic Reporting Rule. The option of providing a public briefing on special studies that the final rule provides in paragraph 102(a)(1)(ii) is one that the Postal Service is free to decline if it wishes.

C. The Need for Mid-Level Documentation of the CRA

Recognizing When There Are Grounds for Initiating §§ 3623 and 3662 Hearings

The Postal Service devotes the majority of its comments to impressionistic descriptions of the CRA documentation that the rule would require, followed by conclusory statements that the documentation “far exceeds” the Commission's “legitimate” needs. The Postal Service's conclusion that the CRA documentation required by the rule exceeds the Commission's needs rests heavily on the Postal Service's mischaracterizations of the documentation required as all information that the Postal Service uses to develop the CRA from raw data on up. Substantive Comments at 3, 21.

As explained above, the rule does not extend to raw data, the design of Postal Service data collection systems, or the processing programs that edit and structure data into input datasets (“Level Five” and “Level Six” documentation). As the most recent fully-litigated rate case demonstrates, the design of data collection systems and the structure and editing of raw data into input datasets can be of fundamental importance in evaluating the soundness of a study. Even with the updated rule, analysts in interim years would have to assume that these aspects of the CRA are valid, and wait for an omnibus rate case for an opportunity to investigate them.

With respect to Level Four documentation, the revised rule requires that only input datasets that have changed since the last general rate case be provided. More significantly, the rule only requires minimal Level Three documentation (the input data and processing programs that implement new analytical models). The rule, therefore, is a balanced compromise, falling well short of everything that would help the Commission understand and evaluate the CRA results.

Narrative explanations of new studies are not required, so it may be difficult for the Commission and the affected public to evaluate their soundness. With the input datasets and the processing programs, however, the Commission and the interested public at least can run the CRA model with the established method, and then with new study inputs, to see the impact that the new study has on estimates of subclass attributable costs. Having the Level One and Level Two documentation required by the rule makes this crude level of diagnosis possible, as they are necessary to run the CRA model. Level One and Level Two documentation also make it possible to gain some insight into why the new study has the effect that it has on subclass attributable costs, because its effect on other inputs, and its effect on intermediate outputs, can be observed.

The Level One, Two, Four, and partial Level Three documentation that the rule requires falls well short of what will be needed in a rate case to fully evaluate the merits of a new study. But without it, the Commission and the affected public would have to simply accept the estimates of total subclass attributable costs reported in an interim-year CRA on faith. There would be no way to know if shifts in subclass attributable costs reflect true underlying economic effects, changes in data sources, or changes in estimation techniques. Similarly, if costs have not shifted, there is no way to tell if this reflects underlying economic stability, or the failure to update Level Three models with current-year data.

The Postal Service apparently believes that the regulatory scheme established under the Postal Reorganization Act functions perfectly well when this level of public ignorance prevails between rate cases. But the Act clearly anticipated that the hearings that the Commission conducts when the Postal Service files a rate case are not, by themselves, enough to ensure that the policies of the Act are carried out. That is why § 3623 of the Act authorizes the Commission to initiate classification hearings on its own. That is also why § 3662 of the Act empowers the Commission to review public complaints that current rates or classifications are in violation of the policies of the Act, in order to determine whether they warrant a hearing.

The CRA is the Postal Service's most important and fundamental report on subclass attributable costs, volumes, and revenues. Without any documentation, its estimates are effectively unreviewable. Under these circumstances, the Commission's ability to make an informed decision on the need to initiate classification reform or to hold a hearing on a complaint that current rates or classifications violate the Act is severely circumscribed. If, in the period between rate cases, the Commission cannot thoroughly interpret and understand the Postal Service's routine financial reports that bear on ratemaking, these remedies that the Act provides are undermined.

The following may help illustrate this point. In its comments, the Postal Service mentions the need for flexibility to respond to the “fast moving markets in which it competes.” Id. at 19. Over four years, the volume patterns and cost structure of these markets might change enough to invalidate the assumptions upon which current rates were based. If the affected public has access to the partial CRA documentation required by the rule, it would have a way to identify when key assumptions underlying rates are no longer valid, and a greatly improved opportunity to learn whether a petition for relief is warranted. If the public were to go four years without documented CRA estimates, and therefore could not learn how they were estimated or what influences they reflect, it might seriously misestimate the basis for a petition for adjusted rates or classifications.

An example of how key assumptions underlying a set of rates could become invalid over time without detection would be in Standard Mail. In the Enhanced Carrier Route (ECR) subclass of Standard Mail, seven IOCS-dependent discounts are offered. The size of the discounts is determined by IOCS tally data. There are instances in the past where appropriate cost-based discounts, based on IOCS tally data, have abruptly and substantially increased or decreased in some of these presort levels but not others. If these abrupt, substantial shifts were sustained over several years, the disparity between the cost differences among density levels on which the discounts were based, and the cost differences reflected in current data, could become large, making the actual passthroughs of avoided costs so disparate that it might violate the fairness and equity criteria of the Act. See 39 U.S.C. § 3622(b)(1).

A mailer of presorted Enhanced Carrier Route mail could not know of the passthrough disparities that had emerged unless it had access to the kind of CRA documentation required by the rule. It could not make effective use of its right to ask for a hearing under § 3662 to show that ECR rates violate the policies of the Act, because it could not demonstrate to the Commission that the cost assumptions underlying these discounts were no longer valid. Although the Act provides the public with this alternative way to secure its rights under the Act, this alternative is of little value when facts that are basic to ratemaking and classification are unavailable to the public.

Being Adequately Prepared To Process Rate Requests

Beyond indicating when there are grounds for initiating § 3623 and § 3662 hearings, the Commission and the public need mid-level documentation of the CRA to be adequately prepared to process cases brought by the Postal Service. These include omnibus rate cases under § 3622, which must be processed within the severely compressed 10-month window allowed by the Act. They also include “experimental,” “market test,” “negotiated service agreement” and other special kinds of rate and classification cases for which the Commission has created even more compressed hearing procedures, at the Postal Service's request.

The reasons that such documentation is needed for these kinds of cases were previously explained in describing the history of the Periodic Reporting Rule. The first of these reasons is the fact that the Postal Service's CRA, which is the starting point for analyzing any set of proposed rates, is exceedingly complex, continually changing, and has proved to be extremely difficult to comprehend in the few months allotted for discovery in a general rate case. Providing partial documentation of the most recent versions of the CRA between rate cases is the minimum necessary to make this task feasible in the hearing time allotted.

In addition to keeping its technical expertise current in order to quickly process a rate case brought by the Postal Service, the Commission described a second reason that partial documentation of the CRA is needed between rate cases. See Docket No. RM76-5, Notice of Proposed Rulemaking, issued April 6, 1976, at 2-3. This is to narrow the scope of discovery and Presiding Officer's Information Requests needed in rate cases, and shorten this phase of the proceeding. The objective is to avoid, as far as possible, inquiries into technical costing issues that are background, or generic in nature, rather than tied to a specific set of proposed rates. This objective has not changed since the Periodic Reporting Rule was first issued 27 years ago.

Researching Cost Behavior Between Omnibus Rate Cases

A third reason that partial documentation of the CRA is needed between rate cases is to gain access to the basic datasets needed to develop models of cost behavior that can be presented in an omnibus rate case, so that less of the litigation window is consumed with such development work. See PRC Order No. 141 at 3. This objective, also, has not changed since the Periodic Reporting Rule was first issued. What has changed since the Periodic Reporting Rule was last updated are the new sources of data that the Postal Service uses to develop the CRA, and the increased capability of the Commission and the intervenors to work with those data.

To develop models of postal cost behavior, it is necessary to have two things—relevant data, and the time and resources to analyze the data. The Postal Service is in a unique position among all stakeholders in postal ratemaking in this regard. It has exclusive control of almost all of the data that could be used to model postal cost behavior. When it decides to study a particular area of postal cost behavior, it has well over 100 in-house analysts and consultants whose time and expertise can be enlisted in the effort. As the only initiator of rate cases, it has exclusive control over the timing of rate cases. Consequently, when the Postal Service wants to develop a model of postal cost behavior, it can decide for itself what data to access or what new data to collect, how long to spend developing its model, and when to initiate a hearing to present it.

Currently, a mailer or competitor that would like to develop an alternative model of cost behavior has little chance of doing so. Between omnibus rate cases it cannot get access to data that reflect current postal operations. When the Postal Service files an omnibus rate case, an intervenor will have about two months to digest the mammoth filing and determine what to investigate, and perhaps three months to find analysts, request relevant data, develop a model, and defend the results. It might have only a few weeks to do this in more abbreviated hearings, such as hearings on experimental services. It is almost impossible for intervenors in rate cases to plan, complete, and defend models of postal cost behavior in the narrow litigation window allowed. As a result, in over 30 years of Commission hearings, intervenors have almost never succeeded in developing significant alternative cost attribution models. During rate cases, intervenors are confined almost entirely to reacting to and criticizing the models developed by the Postal Service.

Due Process and the Need for Data

The Commission has explained why relevant data are an indispensable tool for researching, analyzing, or modeling postal cost, volume, or revenue behavior. In its comments, the Postal Service takes the position that, under the Act, no entity other than itself may possess this tool between omnibus rate cases. Substantive Comments at 6. It asserts that if the Commission or the intervenors make any use of the datasets or the programs underlying an interim-year CRA, they would be conducting a “de facto rate case” outside the confines of a formal hearing. Id. at 16-17. According to the Postal Service's logic, any activity that others do during a rate case—such as studying postal cost, volume, or revenue behavior—may not be done outside of a rate case; otherwise, the 10-month time limit on rate cases is violated. The Postal Service does not extend this logic to itself, however. It may devote whatever time it wants to studying and preparing for rate cases without violating the 10-month time limit for rate hearings under the Act. Bearing in mind that significant new studies of postal cost behavior almost never can be started, completed, and defended within the allotted portion of a 10-month rate case, the Postal Service's view of the Act means that only it has any realistic chance to develop analytical models for ratemaking, since only it may possess the required data between omnibus rate cases.

The Postal Service insists that if it is to have due process during rate cases, it must be able to withhold basic financial data between rate cases. Otherwise, it argues, it would lose what it believes to be its statutory prerogative to surprise opponents with every element of its rate filing. It insists that its prerogative extends to the generic, background financial data summarized in its standard financial reports. Substantive Comments at 36. Its view that the statute grants it an unlimited right of surprise ignores the due process needs of the affected public that participates in rate cases.

When the Postal Service eventually decides to file a rate case, it may present numerous new models that it has had ample time to prepare. Since alternative models can rarely be developed and defended in the time available, the only effective alternatives to the Postal Service's new models are the Postal Service's old models on which existing rates are based. The Postal Service's models, new and old, are typically the only ones eligible for adoption, because they are typically the only ones that have been presented on the record.

The Postal Service thinks that the due process objectives of the Act are well served under these circumstances. But if intervenors are to ever have a realistic chance to develop alternative cost attribution models for consideration in an omnibus rate case, they will, at a minimum, need access between rate cases to the current Level Four datasets that are used to produce the CRA.

Expertise and the Need To Replicate the CRA

The Commission and the public also need Level Four datasets in order to replicate the various attribution and distribution techniques that the Postal Service uses to produce the CRA. The Postal Service doesn't appear to object to the Commission replicating its Base Year CRA model, and the various cost component analyses used to produce it, in the context of a rate case. As a practical matter, for reasons explained earlier, the Commission must use the Postal Service's “state of the art” attribution engine as the starting point for estimating the subclass attributable costs that will support the Commission's rate recommendations. The Commission must first replicate the Postal Service's CRA model in order to confirm that it understands how the model estimates subclass attributable costs, and that it can accurately reproduce the result that the Postal Service's version of the CRA produced. The Commission must then adapt the Postal Service's CRA model to produce Base Year subclass cost estimates that are consistent with the Commission's recommended attribution methods. Because developing a CRA model for a given year is a mammoth undertaking, even for the Postal Service, errors and inconsistencies are inevitable. Before it bases rate recommendations on the Postal Service's CRA, the Commission must ensure that errors and inconsistencies have been identified and corrected. To do this, the Commission issues Presiding Officer's Information Requests asking the Postal Service to explain or resolve apparent errors.

No thorough and coherent statement of the mechanics of producing the CRA has ever been provided by the Postal Service. Narrative descriptions of something as complex as the CRA, such as the Service provides in rate cases, are unlikely ever to be adequate to enable an analyst to thoroughly understand it. Therefore, replication is the primary tool available to the Commission and the public to diagnose errors in the Postal Service's CRA model, and isolate their sources. Running the model is also the only way that the Commission can test the forecasts on which its recommended rates were based to see if its forecasting assumptions are holding up in interim years, and if not, which ones are failing. The ability to undertake this exercise should significantly improve the Commission's forecasting expertise.

Replication makes diagnostic tests of various kinds possible. For example, to test whether the data that the Postal Service used in its CRA model were properly updated, this year's model could be run with this year's data, and then with last year's data, and the results compared. To test whether a processing program has changed, this year's data could be input into this year's CRA model, and then into last year's CRA model, and the results compared. Replication can also be done in stages, allowing intermediate outputs to be examined, to better isolate errors, or inconsistencies with earlier versions. And replication can be used to do sensitivity analyses, changing only selected inputs, or selected processing steps, to try to find reasons for unexpected results. To replicate the CRA model, or its components, the Commission and the public need the relevant input datasets and processing programs.

The Postal Service is skeptical that the Commission and the public have a need for partial documentation of the CRA. It comments that “[t]he Commission has been carrying out [its] duties for decades without having routine and frequent access to such information.” Substantive Comments at 11. In recounting the history of the Periodic Reporting Rule, the Commission has already described some of the compromises that it has been making during rate cases for decades. For decades, it has been unable to decipher and work with CRA-related databases and models that the Postal Service maintains on its mainframe COBOL computer platform. The Commission ultimately gave up this pursuit and developed its own PC-based CRA model that mimics the Postal Service's inaccessible model. During omnibus rate cases, the Commission most closely analyzes the Postal Service's ever-changing CRA model in areas that are in substantial dispute. Because so much of the available time is spent determining how the Postal Service arrived at its disputed estimates of attributable cost, the Commission's evaluation of undisputed estimation techniques is sometimes less thorough than is desirable. The Commission may not address less significant changes that the Postal Service makes in other costing areas, because there isn't sufficient time in a 10-month hearing to analyze it all.

Intervenors in omnibus rate cases, of course, have, for decades, had similar problems. To quote American Business Media:

With data available on an on-going basis, not only would the Commission be better prepared for a rate filing, but the Postal Service's customers would not bear the burden of having the ten or twelve feet of papers, plus computer materials, dropped on them with the expectation that they can review, understand, question and refute those portions that are relevant in time for the Commission to issue a recommended decision in ten months.

ABM Reply Comments at 2. The OCA adds:

As a participant in rate cases, the OCA has watched the complexity and sophistication of Postal Service presentations rise exponentially. The “lead” time required by the OCA (or any other participant) to match the level of the Service's evidence has also increased exponentially. But the Postal Service seeks to preserve its lead-time advantage of ‘six months' while denying any lead time to participants. At some point (already passed, as far as the OCA is concerned), the advantage to the Postal Service becomes overwhelming, and due process evaporates.

OCA Reply Comments at 4-5 [footnote omitted]. The need for the updated Periodic Reporting Rule seems to be clear to everyone in the postal community except the Postal Service itself.

Replication and Bias

While the Postal Service does not deny that the Commission may use replication of its CRA as a legitimate diagnostic tool in the context of a rate case, it recoils at the thought that this tool might apply to an interim-year CRA. It warns:

as the Commission's staff confirmed, the new requirements are designed to allow the Commission to completely re-run the most recent, updated CRA model based on new or alternative inputs, and thereby give the Commission the capacity to develop anticipatory rate recommendations without any formal request or policy guidance from the Postal Service.

Substantive Comments at 10.

If the very staff that are replicating, validating, and otherwise manipulating the fundamental financial and operating information sought in this rulemaking are inevitably forming impressions and conclusions from their investigations, what is to prevent those impressions and conclusions from influencing the outcome before anyone has had their opportunity to persuade? No safeguards exist which would prevent such contamination of the hearing process, and it is difficult to imagine how such safeguards could be implemented in a practical manner. The Governors are entitled to a recommended decision free from any hint of extra-record determinations, and which gives appropriate recognition to the respective statutory roles of the Governors, the Board of Governors, and the Commission.

Id. at 17 [footnote omitted].

It is important to understand what the Postal Service is expressing fear of in these comments. The CRA is the Postal Service's routine financial report that is most relevant to ratemaking because it estimates subclass attribution costs, volumes, and revenues each year. It has been examined by outside auditors, and undergone multiple layers of review by the Postal Service's staff to the point that it is accepted as the most reliable data that it can provide to postal management to guide it in matters of classification and pricing. In the Postal Service view, if the outside world understands little or nothing about how it obtains these estimates, it will not misinterpret them, or be biased, or be misled regarding the relative responsibility of the various subclasses for the Postal Service's financial condition. The Postal Service evidently believes that the more accurate an understanding the outside world gains about the data and estimation methods that produce the CRA results, the more likely that it will be misled, biased, and prejudiced by them.

As the OCA commented, knowledge is a lot less likely than ignorance to produce bias. OCA Comments at 2-3. This is especially true where competence to form an opinion is presumed to require a great deal of industry-specific statistical and economic expertise. GCA points out in its comments that Congress's primary objective in creating the Postal Rate Commission was to ensure that rates would be based on these kinds of expertise. GCA Comments at 4. Congress intended that issues of cost attribution, in particular, should be resolved by application of the Commission's expertise. National Association of Greeting Card Publishers v. USPS, 462 U.S. 810, 823 (1983). The Commission, however, is in a difficult position when it comes to maintaining that expertise.

As explained previously, the Postal Service controls all of the data, has almost exclusive access to field experts, and employs almost all of the analytical resources that are devoted to estimating postal cost, volume, and revenue behavior. For these reasons, when it comes to cost attribution, the Postal Service's cost attribution engine (the CRA) is the starting point for all analysis. Its most current CRA apparently is based on two major new studies of attributable carrier costs and facilities costs. Each study is a “black box” as far as the outside world is concerned, and is likely to remain so for several more years without an updated Periodic Reporting Rule. By imposing a 10-month time limit on the Commission for processing rate requests, the Postal Reorganization Act assumes that the Commission can process such requests with extreme expedition and still base its recommendations on a thorough understanding of all aspects of the record.

Once an omnibus rate request is filed, there isn't sufficient time in the 10-month statutory period to search and resolve issues relating to the mechanics of producing the CRA, and still address the major analytical and policy issues that are raised by a Postal Service omnibus rate request. The mechanics of producing the CRA are generic, background issues, that do not ordinarily depend on a particular time period, a particular revenue requirement, or a particular set of proposed rates. Therefore, documenting this aspect of the CRA should not compromise the right of participants to litigate rate-case-specific issues when a specific rate request is filed. Replication is the primary tool for understanding the technical aspects of estimates found in the CRA.^[13]

The Postal Service suggests that at the technical conference held on March 11, 2003, the Commission's staff somehow signaled its intention to use the documentation required by the Periodic Reporting Rule to develop “anticipatory rate recommendations”' outside of a rate case. Substantive Comments at 10. The Commission does not contemplate going to the considerable trouble to develop rate recommendations that have nothing to do with a particular rate case. This isn't because there would be anything wrong with it. It is because such an exercise wouldn't be very informative if it were not tied to a particular rate request, revenue requirement, and time period.^[14]

With respect to any tendency of replicating the CRA to produce bias in the replicating party, the salient point is that the Periodic Reporting Rule requires only input datasets and the processing steps applied to those datasets—the minimum documentation that will disclose the mechanical process by which CRA estimates were obtained. The data used and the processing steps applied are facts that speak for themselves, devoid of argument, interpretation, spin, or nuance. Making facts available for analysis risks bias only to the extent that the facts made available are selective, or one-sided. If facts disclosed in the documentation of an interim-year CRA were to resemble facts included in the Base Year documentation in a future rate case, the Postal Service's opponents might be able to argue that the Commission was swayed by having an early look at the Postal Service's version of the facts without any alternative version to counterbalance it. If this is a potential source of bias, it works for, not against the Postal Service.

The Postal Service has suggested that it should not have to disclose a set of facts similar to those that it might present in a future rate case without having an opportunity to persuade the outside world of the merits of the procedures that those facts reflect. Id. at 17. If the Postal Service would like to add to the documentation that is required by the rule a discussion of the merits of the procedures that it has used to produce an interim-year CRA, nothing in the rule would prevent it from doing so. Any bias that might result from this opportunity to persuade should benefit the Postal Service.

The Postal Service's main argument, however, is that if the Commission has an opportunity to view facts in an interim year that might resemble those that will be submitted in a future rate case, it will have more time to form opinions about them than it would otherwise have. Ibid. Contrary to conventional wisdom, the Postal Service apparently believes that the less time a staff has to think about a subject, the less likely its thoughts are to be biased. As the Greeting Card Publishers point out, the solution to bias in exercising a judicial function is not to obtain less knowledge, but to exercise the appropriate caution in the use of the knowledge obtained. GCA Comments at 2-3, n.1.

For example, the Postal Service expresses concern that an analyst that obtains an input dataset from the Postal Service could model it differently than the Postal Service modeled it. But this is not a reason to withhold the data. If an intervenor were to model the data differently, it would not affect the Postal Service unless the intervenor subsequently presented it for consideration in a formal hearing. In this way, the right of the Postal Service to debate or oppose it before it had an impact on recommended rates would be preserved. By the same token, if the Commission were to model data differently, it would not affect the Postal Service unless the Commission subsequently asked the participants in a formal hearing to comment on it in a Notice of Inquiry. The Commission could not affirmatively rely on any such model unless it were presented on the record. Here too, the right of the Postal Service to debate it or oppose it before it had an impact on recommended rates would be preserved.^[15]

IV. The Legal Basis for the Periodic Reporting Rule

A. The Periodic Reporting Requirement Is Authorized by the Act

In discussing the legal basis for the Periodic Reporting Rule, the Postal Service argues that the Commission does not need the information required by the rule to perform any statutory function. It also argues that it conflicts with its right to decide when to file a rate case under § 3622, its implied right to decide when to reveal its evidence and argument in support of its proposed rates, and its implied right to have all analytical activity concerning postal rates confined to the 10-month litigation window allowed by the Act for rate cases. Substantive Comments at 15-17, 36.

The Postal Service asserts that the Postal Reorganization Act does not authorize the Commission to adopt periodic reporting requirements. It recognizes that § 3603 of the Act authorizes the Commission to “promulgate rules and regulations and establish procedures * * * and take any other action they deem necessary and proper to carry out their functions and obligations * * *.” In its view, the Commission's only role under the Act is to process a Postal Service request for changes in rates within the 10 months allotted by the Act. Outside of that 10-month litigation window, it reasons, the Postal Rate Commission has no functions or obligations, and therefore § 3603 does not imply any authority to carry them out. Id. at 8-14.

The Scope of § 3603 Is as Broad as Its Language

The Postal Service argues that the scope of § 3603 is much narrower than its broad language suggests. It contends that the following language in § 3624(b) of the Act, which deals with the conduct of formal Commission proceedings, “specifies the type of rules that were contemplated” by § 3603. It quotes:

In order to conduct its proceedings with utmost expedition consistent with procedural fairness to the parties, the Commission may (without limitation) adopt rules which provide for—

(1) The advance submission of written direct testimony;

(2) The conduct of prehearing conferences to define issues, and for other purposes to insure orderly and expeditious proceedings;

(3) Discovery both from the Postal Service and the parties to the proceedings;

(4) Limitation of testimony; and the conduct of entire proceedings off the record with the consent of the parties.

It then comments:

[b]y the very nature of the examples enumerated, the Commission's rulemaking authority is shown to be simply that necessary to implement its limited statutory role: The efficient administration of a hearing after it has been appropriately initiated under sections 3622, and 3623. The rules now contemplated go far beyond this intended role.

Id. at 14.

In drawing this narrowing inference, the Postal Service makes no effort to account for the broad wording of § 3603, which authorizes the Commission not only to “promulgate rules” but to “establish procedures” and to “take any other action” it deems to be necessary and proper to carry out its functions and obligations. If § 3603 were meant to authorize only rules governing formal hearings, one wonders why Congress saw any need to include § 3603 in the Act, since the Act already specifies the kind of rules the Commission may adopt for that purpose in § 3624(b). The Postal Service's interpretation of § 3603 renders the section entirely unnecessary. It is a basic canon of statutory construction that statutory language will not be construed in such a way as to make another section of the same statute “superfluous, void, or insignificant.”^[16]

Section 3603 Extends to Commission Initiated Proceedings Under §§ 3623 and 3622

By authorizing the Commission not just to promulgate rules, but to “establish procedures” and to “take any other action” it is clear that § 3603 authorizes the Commission to do more than promulgate rules of only one narrow type. In characterizing the Commission's authority as “the efficient administration of a hearing after it has been appropriately initiated” the Postal Service glosses over the fact that Chapter 36 of the Act gives the Commission discretion to initiate classification hearings under § 3623, and complaint cases under § 3662. The Act authorizes the Commission to exercise its discretion and judgment as to whether there are good grounds for initiating such hearings. In doing so, the Act contemplates that the Commission will have access to relevant, reasonably current financial data—before hearings are held—that would allow the Commission to make informed decisions as to whether such hearings were warranted.

The Commission's duty to determine if hearings are warranted potentially extends to hearings to reclassify rate categories as subclasses, or vice versa, depending, in part, whether they have enough common cost-driving characteristics. The Commission's duty to determine if hearings are warranted also potentially extends to complaints that certain rates or discounts are unfair because there have been major shifts in relative cost savings since rates were last approved. Determining whether such hearings are warranted requires the Commission to make informed judgements about subclass and rate category attributable costs. To exercise the discretion that the Act calls for, the Commission needs access to attributable cost estimates about which it can make some judgments, not just unreviewable, bottom-line estimates based on unknown data sources and estimation methods. The partial CRA documentation required by the Periodic Reporting Rule can fulfill this need.

Section 3603 Authorizes Measures That Make Processing Postal Service Requests More Efficient and More Fair

The more important need that partial documentation of the CRA fulfills relates to the omnibus rate cases that the Postal Service files, and the myriad minor rate and classification cases that it files under abbreviated hearing schedules. The “efficient administration” of those hearings, to quote the Postal Service, is not the Commission's only function. Its function is not just to conduct those hearings “efficiently” in a severely compressed time frame, but to conduct them fairly.

In order to conduct omnibus rate hearings initiated by the Postal Service efficiently, the Commission has to be able to read, comprehend, and in some respects, repair, the Postal Service's “cost attribution engine” before it can address the analytical and policy issues raised by a rate request. For this process not to swallow up the majority of the available hearing time, the Commission needs to begin the process with an understanding of the current CRA. This requires reasonable familiarity with the mechanics of the Postal Service's current cost attribution methods. If it has to devote a major portion of the litigation window to acquiring this familiarity, it may have to give the merits of the Postal Service's cost attribution methods, as well as basic volume and revenue estimation issues, short shrift. All of these considerations apply equally to intervenors in Commission proceedings. For them to have meaningful due process, they not only need a reasonable chance to understand and respond to the Postal Service's entire case in the time allotted, which they cannot do without the documentation required by the Periodic Reporting Rule, they need a reasonable chance to develop, propose, and defend alternative cost estimation techniques in the time allotted. This they cannot do without the datasets required by the Periodic Reporting Rule.

Section 3603 Authorizes Measures Designed To Expedite Minor Cases

Between omnibus rate cases, the Postal Service often files requests for changes in rates for individual mail categories. When it does, it usually seeks to expedite the case by seeking a waiver of the Commission's normal documentation requirements for rate and classification cases found in rules 54 and 64. These require the Postal Service to provide full documentation of its base year attributable cost, volume, and revenue estimates. The Postal Service usually files these requests under Commission rules that drastically shorten the 10-month period that the statute makes available to intervenors to litigate a rate case. These range from “experimental” cases, which have a 150-day litigation schedule, to Negotiated Service Agreements, for which a 60-day litigation schedule has been proposed.

These rules allow expedition when intervenors raise only issues of limited scope and complexity. The ability of the Commission and the intervenors to process such cases within severely compressed schedules also depends on their ability to do without a fully documented request. If a partially-documented CRA has been filed under the Periodic Reporting Rule for an interim year, the intervenors are much more likely to be able to do without a fully documented base year CRA, and the Commission is much more likely to grant the waivers that expedition requires, and allow cases to proceed under its expedited rules.

The abbreviated hearing schedules provided for by the Commission's expedited rules are Commission attempts to implement the ratemaking provisions of the Act in a manner consistent with modern needs of the Postal Service. The concept behind them is that there is not a need in every case to litigate every issue—including the mechanical structure of the CRA—from scratch. The Periodic Reporting Rule is based on that concept as well.^[17]

Section 3603 Is Not Restricted to Measures Whose Need Is “Compelling”

The Postal Service comments that certain information required by the Periodic Reporting Rule is not “strictly needed to conduct rate and classification proceedings” and “not required by a compelling and legitimate function.” Substantive Comments at 11, 13. It thereby implies that the statutory threshold for invoking the authority of § 3603 is that a regulation be an indispensable means of achieving a statutory purpose. This is a gloss on § 3603 that cannot be found in the legislative history or inferred from the structure of the Act. The provision authorizes actions that the Commission deems to be “necessary and proper to carry out their functions and obligations * * *.”

As the Commission has explained, it has an obligation to competently address the full range of issues presented by a rate request in the short time allotted by statute or the even shorter time allotted by special rule. To help do this it must maintain a certain level of expertise in current methods of attributable cost, volume, and revenue estimation. The partial documentation of the CRA required by the Periodic Reporting Rule will help it to maintain this expertise. Maintaining the requisite expertise to do the Commission's job effectively is not a minor consideration in making sure that the Postal Reorganization Act functions as Congress intended. As the Supreme Court, in National Greeting Card Publishers v. USPS, observed:

Congress recognized that the increasing economic, accounting, and engineering complexity of ratemaking issues had caused Members of Congress, “lacking the time, training and staff support for thorough analysis,” to place too much reliance on lobbyists. House Report, at 18. Consequently, it attempted to remove undue price discrimination and political influence by placing ratesetting in the hands of a Rate Commission, composed of “professional economists, trained rate analysts, and the like,” id., at 5, independent of Postal service management, id., at 13, and subject only to Congress' “broad policy guidelines,” id., at 12.

462 U.S. 810, at 822. [Emphasis added.] In its comment, the Greeting Card Association points out that—

Since [the Commission's] duties centrally include acting as an expert decisionmaker on matters arising under ch. 36 of title 39, it seems clear that obtaining the information covered by this docket regularly and systematically, in usable form, and in a timeframe allowing it to be given mature consideration is a legitimate need.

GCA Comments at 4. UPS agrees. See UPS Comments at 2,4.

Similarly, the Commission is obligated to afford intervenors meaningful due process in its rate hearings. Making the information required by the rule available to prospective intervenors will help them comprehend the prodigious amount of technical information presented in an omnibus rate request, and to develop and present alternative estimation techniques, within the short time allotted by statute. As the OCA points out, “providing due process to all participants within a ten-month time period is ‘a compelling and legitimate Commission function.’ ” OCA Reply Comments at 3-4 [footnote omitted].

The information required by the rule might not “make or break” the achievement of the legitimate statutory goals of maintaining the Commission's ratemaking expertise, and affording prospective intervenors due process, but it will greatly improve the odds. That is all that is required to come within the authority of § 3603.

Illegitimate Purposes Hypothesized for the Periodic Reporting Rule

Having identified the legitimate statutory functions that the Periodic Reporting Rule facilitates, it is helpful to identify what functions the rule is not designed to serve. The Postal Service suggests that the purpose of the rule is to allow the Commission to conduct “day-to-day monitoring of [the Postal Service's] detailed operations and finances” [Substantive Comments at 7], “auditing of the Postal Service's books on a regular basis” [Id. at 11] and obtain “oversight,” “data collection,” and “investigatory powers.” Id. at 19.

As UPS points out, the updated Periodic Reporting Rule does none of these things. UPS Comments at 1. The rule does not require the Postal Service to provide any backup data with which to audit its books of account, any documentation of its data collection activities, or any data that would make possible daily monitoring of operations or finances. No oversight is involved, and no investigation is involved. Nor does it involve any hearings, or any comments from the public. The rule simply requires the Postal Service to file some of the routinely prepared documentation that support its periodic financial reports that bear on ratemaking.^[18]

In discussing the legal basis for the Periodic Reporting Rule, the Postal Service argues that the rule conflicts with its right to decide when to file a rate case under § 3622, and what it considers to be two corollary rights—the right to decide when to reveal its evidence and argument in support of its proposed rates, and the right to have all analytical activity concerning postal rates confined to the 10-month litigation window allowed by the Act. Substantive Comments at 15-17, 36.

The Rule Does Not Affect the Timing of Rate Filings

When the Postal Service files a rate request under § 3622 of the Act, a complex set of legal consequences attach. These include the right to receive a recommended decision on proposed rates from the Commission, the right to receive it within ten months, and a complex set of options that the Governors have to respond to the Commission's recommended decision, including acceptance, rejection, modification, and the right to appeal that recommended decision. The public's right to intervene, to present evidence, and to appeal the result, also attach when the Postal Service files a rate request.

No legal consequences that affect the Postal Service or the rates that it may charge attach to the filing of information under the Periodic Reporting Rule. When it complies with the rule, the Postal Service has no further legal obligation to do anything. Its complaints, therefore, can only be based on the effects that the rule might have, if any, on the way that its request is handled during a rate case.^[19]

Section 410(c)(4) Is Not Relevant to the Rule

Section 410(c)(4) of the Act exempts “[i]nformation prepared for use in connection with proceedings under Chapter 36 of this title” from mandatory disclosure under the Freedom of Information Act. The Postal Service argues § 410(c)(4) protects information required by the Periodic Reporting Rule from public disclosure not just through Freedom of Information Act requests, but through any other mechanism, including Commission rules adopted under § 3603 of the Act. Id. at 26.

Whether § 410(c)(4) provides a general shield of protection for materials prepared for Chapter 36 litigation is not a question that needs to be decided as part of this rulemaking. There is a crucial distinction between standard financial reports that are routinely prepared for the benefit of management, and financial reports that have been adapted to support a specific proposed revenue requirement and a specific set of proposed new rates, to be implemented in a specific test period. The former class of reports are normal business records not prepared primarily for litigation. The Periodic Reporting Rule requires that some standard business records be provided. The principal effect of providing them is to allow the interested public to learn enough about the way that the Postal Service routinely estimates costs and revenues to comprehend an enormously complex omnibus rate filing in the narrow litigation window provided. Only the latter class of reports are prepared primarily for litigation purposes. The Periodic Reporting Rule does not apply to them. See GCA Comments at 5, n. 6. Therefore, the rule does not infringe on the Postal Service's right under § 410(c)(4) not to disclose attorney work product intended for Chapter 36 litigation, even if that provision were to apply outside of the context of Freedom of Information Act requests.

There Is No Statutory Ban on Evaluating Rate-Related Matters Between § 3622 Proceedings

Section 3624(c)(1) of the Act requires the Commission to transmit a recommended decision on the Postal Service's request for new rates within 10 months after receiving the request. The Postal Service argues that this provision entitles it to a respite from litigating rate-related matters outside of this 10-month period.^[20] But the Postal Service goes further, and argues that this respite includes a right not to have to think about rate-related matters, and a respite from having others think about rate-related matters. The Postal Service argues that the Periodic Reporting Rule robs it of the respite to which it is entitled, because it provides others with information that would enable them, in the period between rate cases, to study how postal costs and revenues behave. Substantive Comments at 17.

Apparently, in the Postal Service's view, a “rate case” happens whenever, and wherever a person's thoughts turn to postal cost or revenue behavior, and if the Postal Service didn't ask them to do it, § 3624(c)(1) is violated. It is hard to take this proposition seriously. One obvious flaw in this logic is the fact that the Act contains a complaint procedure whereby the public may ask the Commission, at any time, to hold a hearing on whether current rates violate the policies of the act. See 39 U.S.C. § 3662. This cannot be reconciled with the Postal Service's “respite” theory. It should also be noted that there is nothing in the legislative history of § 3624(c) to suggest that it was motivated by a desire to give the Postal Service a respite from other Chapter 36 hearings, let alone give it a respite from others' rate-related thoughts.^[21]

B. Disclosure Policy Under the Periodic Reporting Rule

The Postal Service expresses a deep-seated insecurity about the Commission's willingness and ability to afford confidential treatment to materials that it might provide under the Periodic Reporting Rule. The cause of this insecurity, it asserts, is its experience with Freedom of Information Act (FOIA) requests for information that it has provided to help the Commission prepare its report to Congress on international mail under 39 U.S.C. § 3663. Substantive Comments at 29-30.

The Postal Service emphasizes that it is exempt from a duty to disclose commercial information that “under good business practice would not be publicly disclosed” by § 410(c)(2) of the Postal Reorganization Act. It argues as though this section of the Act is a general-purpose exemption from the duty to disclose commercial information. Section 410(c)(2), however, is expressly limited to the Postal Service's duty to respond to FOIA requests. The Commission's authority to require the Postal Service to provide information on a periodic basis is not derived from FOIA, but from its authority under § 3603 of the Act to adopt procedures that are “necessary and proper” to carry out its ratemaking and classification functions.

More to the point, the stated concern about the potential need to afford confidential treatment to data provided to comply with the proposed rule changes appears to be largely a red herring. As discussed at some length above, all of the information called for by the proposed rules have been provided and made public in recent years, either as part of a rate case filing or as a courtesy to the Commission. Except for facility-specific data, the Postal Service has never seen fit to request confidential treatment for any of this information, and it does not now identify any competitive disadvantage it is likely to suffer as a result of public access to these historical operating results. Except for facility-specific data, it seems highly unlikely that the Postal Service will have justification to seek confidential treatment of the materials it will provide to comply with revised rule 102.

Furthermore, the Commission always has conscientiously dealt with requests for confidential treatment of data both during and outside of docketed cases. Existing procedures assure that both the Postal Service, and all others providing information to the Commission, have ample safeguards to assure that their concerns will be fully heard and evaluated in timely fashion, and their rights fully protected.

Protecting Commercially Sensitive Information Required by the Rule

In a rate case, a litigant occasionally will seek to discover information that another litigant considers to be commercially sensitive. The Commission resolves such issues by balancing one litigant's need for the information to support its case against the potential commercial harm that disclosure might cause to the other litigant. In such cases there is a general presumption that discovery should be granted in order to afford the discovering party its due process right to prove its case. The tension between the discovering party's need to prove its case and the opponents need to protect commercially valuable information is sometimes resolved by granting discovery subject to various protective conditions. For example, only temporary access may be granted, and only to specified persons or groups.

The CRA documentation that the rule requires is among the documentation that the Postal Service has consistently disclosed in rate cases without asserting that the documentation has commercial value and without seeking protective conditions. The minor exception to this Postal Service policy has been its consistent request that facility-specific data, and mailer-specific data, be coded so as to mask the identity of the facility or the mailer. The Postal Service has not explained why the same approach could not be satisfactorily applied to disclosure under this rule.

In its comments, the Postal Service strenuously objects to following the same disclosure policy with respect to the same documentation in the context of the Periodic Reporting Rule. It asserts that “the majority of the information designated by the Commission's proposed rules consists of commercial information that would not be disclosed under good business practices.” As part of its justification for this position, it notes that the material covered by the rule includes information specific to particular facilities. Substantive Comments at 31-32. The position that the Postal Service has taken appears to be based in part on the assumption that facility-specific data would somehow be at risk if it were disclosed under the rule. The Commission has no intention of affording less protection to the information obtained under the Periodic Reporting Rule than it has consistently afforded in the context of a rate case. The position that the Postal Service has taken also appears to be influenced by its overly-broad and inaccurate characterizations of the documentation that the rule requires.

Express Authority for the Rule Is Not Required

The Postal Service emphasizes that the Postal Reorganization Act does not expressly authorize the Commission to require the Postal Service to provide access to information outside the context of Chapter 36 rate hearings. Id. at 24. The Postal Service further emphasizes that § 410(c) of the Act exempts it from the obligation to disclose the following information in response to Freedom of Information Act requests:

Information of a commercial nature, including trade secrets, whether or not obtained from a person outside the Postal Service, which under good business practice would not be publicly disclosed.

The Postal Service assumes that § 410(c) not only exempts such information from mandatory disclosure in response to requests filed under the Freedom of Information Act, but, by implication, exempts it from mandatory disclosure under any circumstance, other than pending Chapter 36 hearings. Id. at 26.

Although the Act does not expressly authorize the Commission to require the Postal Service to provide information outside the context of Chapter 36 hearings, express authority is not required, given the availability of § 3603. Because of its broad language, any exercise of § 3603 authority is necessarily an exercise of implied authority. The issue is whether it is plausibly and reasonably implied.

The Commission has already explained why it needs the documentation required by the rule if it is to effectively evaluate all of the issues presented in § 3624 hearings within severely compressed litigation windows. It has already explained why that information is needed if the Commission is to ensure that intervenors in future rate cases have a realistic opportunity to understand the immensely complex documentation supporting Postal Service rate requests, and to develop alternatives, in the severely compressed litigation window available. Having shown the need for the rule to carry out its functions under §§ 3622, 3623, and 3662 of the Act, the Commission has demonstrated that the rule is the kind of procedure that Congress meant to authorize by the general language of § 3603.

The Rule Does Not Conflict With the Policy Underlying § 410(c)

The Commission sees no conflict between the Periodic Reporting Rule, as authorized by § 3603, and the disclosure policy reflected in § 410(c) of the Act. Section 410(c) does not expressly apply in contexts other than Freedom of Information Act (FOIA) requests. There are special considerations that are likely to have led Congress to exempt the Postal Service from mandatory disclosure of commercially sensitive information in response to FOIA requests. The most significant of those considerations is that in deciding whether to comply with a FOIA request, an agency may not take into account the need of the requesting party for the information that it is requesting, NLRB v. Sears, Roebuck & Co., 421 U.S. 132 (1975) at 143 n. 10, and it generally may not take into account its own burden in complying with such a request. See Ruotolo v. Dept. of Justice, 53 F.3d 4, 10 (2d Cir. 1995). Disclosure is mandatory unless the information falls within one of the narrow exemptions that the FOIA makes available. These simplifying procedures were thought necessary to make the FOIA effective, but they introduce a procedural arbitrariness that is not necessarily appropriate in all circumstances involving disclosure of sensitive materials. Aware that the FOIA does not allow a balancing of the public's need for information against the potential harm to the agency of providing it, Congress exempted the Postal Service from disclosing commercially sensitive materials in the FOIA context.

The Postal Service does not have express authority to withhold information that the Commission needs to effectively carry out its functions under the Act. Outside of the FOIA context, Congress did not expressly exempt the Postal Service from disclosure of commercially sensitive materials, or expressly make the Postal Service the arbiter of what materials should be considered commercially sensitive. What Congress intended where another Federal agency, such as the Commission, has demonstrated a substantial need for information from the Postal Service, and is willing and able to balance its need for that information against the burden and potential commercial harm of providing it, can only be surmised from the other provisions of the Postal Reorganization Act.

The Postal Service frequently points out that Congress, in adopting the Postal Reorganization Act, intended that the Postal Service function more like a private business than it had been functioning. Exemption from responding to some kinds of FOIA requests is one way in which the Postal Service resembles a private business.

At the same time, however, Congress gave the Postal Service special monopoly privileges and made it clear that these privileges carried with them special duties toward the public that a private business does not have. The first section of the Act, 39 U.S.C. § 101, is replete with these special duties. Chief among these are its obligation to provide the public with universal service [§ 101(b)] that is capable of binding the nation together [§ 101(a)].^[22] The Postal Service is also required to charge rates that are based on principles of equity and other public policies, not just profitability. See § 101(d). The Postal Rate Commission has a key role to play in ensuring that rates comply with these policies.

Given the remarkable dissimilarities between the Postal Service and a private business with respect to its obligations to the public, it is implausible that Congress would have intended the Postal Service to have the power to decide for itself what the outside world may know about it, including what the outside world may know about its operations and finances that bear on ratemaking, except during rate litigation. The Commission is not aware of any government monopoly that has been granted absolute power to decide for itself what its disclosure policy will be. It is much more plausible to surmise that, apart from litigation, the Postal Service's power to decide what its disclosure policy will be is not absolute, but qualified. One of the respects in which it is qualified is where the Commission (which has primary ratemaking responsibility under the Act) ^[23] has demonstrated a need for a limited amount of documentation of routine financial reports to make the ratesetting process function properly.

The Commission's Periodic Reporting Rule is a restrained exercise of its authority to establish procedures that ensure that its hearings afford meaningful due process both to the Postal Service and to the affected public, and ensure that the Commission maintains the expertise that is required to make informed rate recommendations. As such, it is authorized under § 3603 of the Act.

In deciding whether the updated Periodic Reporting Rule is an appropriate exercise of the Commission's § 3603 authority, the questions to be answered are whether the need for the information identified by the Commission is real and substantial, and whether providing it would significantly impair the functioning of the Postal Service.

The Need of Litigants for Meaningful Due Process

If there is a common theme among the comments received in this docket from potential intervenors in rate cases, it is that, under current circumstances, the litigation “playing field” in omnibus rate cases is tilted so steeply in favor of the Postal Service that their basic right to due process is jeopardized. The causes are severe asymmetry in the time and resources available to prepare for a rate filing, severe asymmetry in access to relevant information, the immense scope and detail of the filing, and the short statutory deadline for digesting and reacting to it. As the American Bankers Association points out

[m]uch of the complexity associated with omnibus rate cases for the Commission and intervenors arises from the fact that they cannot even begin to prepare for such a case until it is filed. The changes to the Commission's rules as proposed in Commission Order No. 1358, which would require the USPS to periodically file much of the basic information upon which requests for new rates are based, would greatly facilitate effective decision making by the Commission and effective participation by intervenors in omnibus rate cases.

ABA Comments at 1-2. American Business Media concurs:

[w]ith data available on an ongoing basis, not only would the Commission be better prepared for a rate filing, but the Postal Service's customers would not bear the burden of having the ten or twelve feet of papers, plus computer material, dropped on them with the expectation that they can review, understand, question, and refute those portions that are relevant in time for the Commission to issue a recommended decision in ten months.

ABM Reply Comments at 2. The OCA summarizes the dilemma of intervenors:

[a]s a participant in rate cases, the OCA has watched the complexity and sophistication of Postal Service presentations rise exponentially. The “lead” time required by the OCA (or any other participant) to match the level of the Service's evidence has also increased exponentially. But the Postal Service seeks to preserve its lead-time advantage of “six months” while denying any lead time to participants. At some point (already passed, as far as the OCA is concerned) the advantage to the Postal Service becomes overwhelming, and due process evaporates. Whatever may have motivated the Commission to propose the new periodic reporting rules, the effect of their implementation will be to level the litigation field for participants other than the Postal Service.

OCA Reply Comments at 4-5. [Footnote omitted.]

That the playing field is tilted is not merely the self-serving perception of intervenors in rate cases. The Federal court of appeals in Association of American Publishers, Inc. v. Governors of the United State Postal Service, 485 F.2d 768, 779 (D.C. Cir. 1973) has commented on the problem as well:

[The Postal Service] alone takes in the full scope of Postal Service operations when presenting its proposals. And it alone is in a position to influence the Postal Service's day-to-day accounting procedures and record keeping. Outsider challenges to the fundamental approach the Postal Service takes to ratemaking are unlikely to meet with stunning success under these circumstances [footnote omitted].

The problem of securing meaningful due process for intervenors in rate cases is real and substantial, and the information required by the updated Periodic Reporting Rule is reasonably designed to partially solve that problem. The remaining question is whether providing that information would significantly impair the functioning of the Postal Service. The Commission has explained earlier that the burden of complying with the rule is a tiny fraction of the burden of documenting a full-blown rate case, primarily because the Postal Service produces almost all of this information routinely anyway, for reasons apart from the rule.

The Postal Service's Assertions of Commercial Sensitivity Are Unexpectedly Broad

The Postal Service, unexpectedly, asserts that the majority of the information required by the rule is commercially sensitive. It is very difficult to evaluate this assertion because it is made at such a general level. The Postal Service makes almost no effort to identify which of the diverse information required by the rule it now believes falls in this category.

There are a number of reasons that this broad assertion of commercial sensitivity is unexpected. One is because the subject matter of the information required by the rule is information that the rule has required for decades, and the Postal Service has provided for decades, without suggesting that it is commercially sensitive. Another reason that this assertion is unexpected is because the information required by the rule is information that the Postal Service has routinely provided in omnibus rate cases without asserting that it is sensitive and without seeking to file it under protective conditions. Furthermore, on several occasions, the Postal Service has voluntarily disclosed much of the CRA documentation that it now seeks the most strenuously to withhold. In 1998, for example, after the Docket No. R97-1 omnibus rate case was concluded, the Postal Service voluntarily provided the Commission with an extensively documented interim-year CRA that was calculated according to Commission attribution methods. The documentation included the B workpaper spreadsheets and some of the input datasets that the updated Periodic Reporting Rule now requires. There was no assertion that this information was commercially sensitive, and no request that it be maintained under protective conditions. Similarly, on May 30 of this year, the Postal Service voluntarily included with its CRA submitted under the prior version of the Periodic Reporting Rule, the B workpaper spreadsheets for the major costs segments (Segments 3, 6, 7, and 14).^[24]

Taken together, this history adds up to a shift in the attitude of the Postal Service toward the commercial sensitivity of the CRA documentation that the updated rule requires. This new attitude is not satisfactorily justified by the Postal Service. Once facility identifiers are removed from the required datasets, there is no apparent commercial use to which any of this documentation could be put, and the Postal Service has suggested none.

The commenters argue that the Postal Service's main motive in opposing the update to the Periodic Reporting Rule is not its new-found concern for the commercial sensitivity of this data, but the loss of the tactical advantage that it has entering rate cases fully prepared while the intervenors play a desperate game of catch up in the short litigation window available. See ABA Comments at 3-4, OCA Reply Comments at 1, UPS Reply Comments at 6. American Business Media's comments are representative

American Business Media submits that what the Postal Service fears is actually the loss of the enormous advantage it obtains by springing mountains of data, new costing methodologies, and hundreds of proposed rates upon the Commission and other parties a mere ten months before a heavily litigated case with dozens of active parties must be resolved.

ABM Reply Comments at 5. [Footnote omitted.]

The comments in this docket from those who have participated in past rate hearings agree that the Postal Service's assertions of the commercial sensitivity of the information required by the Periodic Reporting Rule are indiscriminately broad. The American Bankers Association, for example, argues that

* * * virtually no other enterprise, including those that compete with the Postal Service in the small area where there is direct competition, has a cost structure that even remotely resembles the cost structure of the USPS. Thus, the sort of data the proposed rule changes would require the Postal Service to produce does not seem to be the sort of data that would give competitors in the small area where there is competition information of value.

ABA Comments at 4.

These commenters point out that only a few of the Postal Service's products are provided in competitive markets (Express Mail, Priority Mail, and parcel post), and that their commercial significance is minor. (They accounted for only 7 percent of net postal revenue in FY 2002.) They argue that this competitive “tail” should not wag the dog in matters of information disclosure. They note that if there is sensitive information about these services in the materials required by the rule, the Postal Service could identify it with specificity and seek appropriate protective conditions to prevent any perceived harm. ABA Comments at 4, GCA Comments at 6-7, UPS Reply Comments at 6.

Commercial Sensitivity Objections Involve Only the Frequency of Disclosure Required by the Rule

As already noted, the Postal Service has consistently disclosed the CRA documentation required by the updated Periodic Reporting Rule to the public in its omnibus rate requests and has voluntarily disclosed most of this documentation in some interim years as well. It has not asked that it be treated as commercially sensitive in either context. The Postal Service's objections to its disclosure under the Periodic Reporting Rule, therefore, have to be based almost entirely on the prospect that under the rule, these same materials would be disclosed more frequently than they otherwise would be. The logic of the Postal Service's position seems to be that disclosing these materials is not a significant commercial risk when their disclosure occurs at the frequency that it has in the past, but would become a significant risk if disclosed annually.

The Potential Harm of Annual Disclosure Has Already Been Tested

If annual disclosure were the true test of the dangers of disclosing these materials, the Postal Service has already conducted this test. Every year for a six-year period starting with FY 1995 (provided in Docket No. MC96-3), and ending in FY 2000 (provided in Docket No. R2001-1), the Postal Service has publicly disclosed the CRA documentation that the Periodic Reporting Rule now requires. Rather than suffer financially, this was by far the most prosperous period that the Postal Service has had since the adoption of the Postal Reorganization Act in 1970. The Postal Service's competitive services participated fully in this unprecedented prosperity. This should dispel the Postal Service's fears that annual disclosure of the information required by the Periodic Reporting Rule will adversely affect its financial prospects or its competitiveness. By the same token, it should also dispel the notion that the institutional calamities that the Postal Service warned would ensue from annual disclosure will, in fact, occur. Over this period there was no attempt by the Commission to monitor day-to-day management of the Postal Service, to audit its books of account, to supervise its data collection activities, or develop rate recommendations outside of a pending rate case. Nor did the Commission compromise any Governors' decision through interim-year research, or take any other action that “fundamentally altered” the institutional relationship between the Commission and the Postal Service.

What did happen over this six-year period of annual disclosure was that the Commission had a better opportunity to keep current on the “state of the art” of cost attribution as practiced by the Postal Service. It had a better opportunity to evaluate its forecasting models to see where their assumptions held up and where they didn't. The Commission was better able to waive some documentation requirements in minor cases, due to the availability of recent Base Year CRA documentation. And, finally, some of the pressure was taken off of the Commission and the intervenors to quickly digest the enormous amount of supporting material filed with omnibus rate requests.

Recent history with annual disclosure, therefore, confirms that it significantly improves the ability of the Commission to process rate hearings without causing any of the various forms of institutional harm that the Postal Service posits. The purpose of the updates to the Periodic Reporting Rule is to continue the successful pattern of the FY 1995-FY 2000 period, rather than to restructure the Postal Reorganization Act, as the Postal Service asserts.

V. Provisions of the Rule Not Related to Documentation of the CRA

In addition to updating the portions of the Periodic Reporting Rule that deal with documenting the CRA, the updated rule reduces the lag allowed for reporting billing determinants for Express Mail, Priority Mail, and parcel post to 12 months after the close of the fiscal year, [§ 102(a)(10)]; requires the Postal Service to provide its Integrated Financial Plan [§ 102(a)(11)]; requires it to provide the input data and calculations used to produce annual Total Factor Productivity estimates [§ 102(a)(12)]; requires it to provide a finer level of detail in its quarterly RPW reports [§ 102(b)(1)]; and requires it to provide On-roll and Paid Employee Statistics (OPRES) [§ 102(c)(4)]. The Postal Service indicates that it does not object to these changes in the updated Periodic Reporting Rule. Substantive Comments at 36-37.

The Postal Service does object to § 102(c)(5), which requires it to provide the “HAT” report, relating to the Postal Service's Active Employee Statistical Summary. It argues that the HAT report includes miscellaneous information about postal employees, most of which is not related to ratemaking. Id. at 37, n.22. While it includes miscellaneous information about employees, the HAT report is very relevant to ratemaking because it lists the number of employees within each pay grade and step within the many different pay scales used by the Postal Service. This information, which can not be found in the On-Rolls and Paid Employee Statistics, is used to develop several estimates that are important in determining the revenue requirement in rate cases. For example, it is used to develop the average step increase for bargaining unit employees, as well as their attrition rate. The information in the HAT report cannot be associated with individual employees, and it is not reported by facility. The Postal Service does not contend that the availability of the data will have any specific detrimental effect on it or its employees. For these reasons, it is included in the updated rule.

VI. Suggestions of the Commenters

In the past, the Periodic Reporting Rule required billing determinants to be reported within two weeks of their presentation to postal management. Since FY 1995, billing determinants have been received from 6 to 16 months after the close of the fiscal year. In the past the Periodic Reporting Rule allowed billing determinants for Express Mail, Priority Mail, and parcel post to be delayed an additional year, causing them to be from 18 to 28 months old when received. In its NPR, the Commission proposed that the rationale for the distinction between billing determinants for competitive mail classes and other mail classes be reexamined. In order to focus the reexamination, the Commission proposed that billing determinants for competitive classes be provided within a year after the close of the fiscal year to which they apply. NPR at 7.

UPS is the only party to offer substantive comments on this issue. It argues that there has never been a reasoned justification for treating billing determinants for competitive classes differently than for other classes in the context of the Periodic Reporting Rule. It points out that the Postal Service provides all billing determinants at the same time in support of its omnibus rate requests, without suggesting that billing determinants for competitive classes are commercially sensitive. It argues that the Postal Service has never pointed to any instance in which providing current billing determinants for competitive services during a rate case has caused it competitive harm, nor identified any way in which a competitor could use current billing determinants to put the Postal Service at a competitive disadvantage. UPS Comments at 2-4. It also argues that all billing determinants should be provided on a date certain, shortly after the close of the fiscal year, rather than on a floating time table as they now are. It cites an example of the Postal Service voluntarily publicizing current volumes for parcel post rate categories at a recent National Postal Forum as evidence that the Postal Service itself does not consider them to be commercially sensitive. Id. at 5.

The Commission agrees that it would be desirable to receive billing determinant information at a consistent and shorter interval after the close of each fiscal year. The Commission has, however, tried to adhere to a policy of requiring reports under the Periodic Reporting Rule that do not add significantly to the burden that the Postal Service already bears when it prepares these reports for its internal purposes. For that reason, the Commission declines to require that most billing determinants be provided on a date certain.

The final Periodic Reporting Rule adopts the billing determinant provision as it was proposed in the NPR. It requires the Postal Service to provide billing determinants for competitive categories within one year of the close of the fiscal year to which they apply. UPS correctly notes that the Postal Service occasionally voluntarily discloses current volumes for competitive services by rate category. However, to the Commission's knowledge, the Postal Service does not voluntarily disclose other current billing determinant detail, such as weight and zone, for competitive categories. The updated rule should cut the delay in reporting billing determinant information for competitive categories from roughly two years to one year. The Commission is reluctant to go further in this regard, without a more thorough discussion of the ramifications than has been provided in this docket.

At the beginning of each fiscal year, the Postal Service prepares an operating budget that includes detailed operating expense and revenue projections for the coming fiscal year, broken out by accounting period. Under the Periodic Reporting Rule, the Commission receives the Postal Service's Financial and Operating Statements several weeks after the close of each accounting period. See rule 102(c)(1). These statements compare the detailed operating revenues and expenses projected in the Postal Service's operating budget with actual results. Under the updated Periodic Reporting Rule, the Postal Service would provide an annual Operating Plan as part of its Integrated Financial Plan. See rule 102(a)(11). This annual Operating Plan is less detailed than the operating plans contained in its accounting period reports.

The OCA proposes that the updated Periodic Reporting Rule require the Postal Service to provide its operating budget projections for all 12 accounting periods at the beginning of the fiscal year, rather than provide them shortly after each accounting period closes. The OCA's rationale for adding this requirement to the updated rule is that the Commission and the public should not have to wait until several weeks after each accounting period to find out the Postal Service's operating plan for that accounting period. OCA Comments at 4-5.

It is not clear that the Board of Governors approves the operating plan as an annual summary document or as a document that is as detailed as the OCA describes. It is, therefore, not clear that requiring accounting period operating budget projections would conform to the criteria that the Commission applies to the Periodic Reporting Rule that it be confined to reports that have already been presented for use by postal management. In view of this ambiguity, and the tenuous demonstration of need for altering the time at which the rule would obtain this information, the Commission declines to include this change in the updated rule.

The Postal Service is required to file a number of reports with Congress to meet the requirements of chapters 24 and 28 of the Postal Reorganization Act. One is the Comprehensive Statement of Postal Operations, which includes an Annual Performance Plan, and annual Program Performance Reports. Another is the five-year Strategic Plan. The OCA proposes that these and all other reports that the Postal Service is required by the Act to provide to Congress be provided under the Periodic Reporting Rule as well. Id. at 3-4. The Commission declines to add reports to Congress to its Periodic Reporting Rules. The Commission prefers to restrict the rule to reports prepared for postal management. The Commission notes that these reports are all readily accessible on the Postal Service's Web site.

VII. Conclusion

For the reasons discussed above, the Commission hereby amends 39 CFR part 3001 (the Periodic Reporting Rules) as set forth below in this order. Any suggestion or request to modify the Commissions rules raised by any participant not specifically addressed herein is denied.

It is ordered:

(1) The Commission adopts the provisions set forth below amending 39 CFR § 3001.102.

(2) The Secretary shall cause this notice of adoption of a final rule to be published in the Federal Register.

Issued November 3, 2003.

By the Commission.

Steven W. Williams,

Secretary.

List of Subjects in 39 CFR Part 3001

• Administrative practice and procedure
• Postal Service

For the reasons stated in the accompanying Order, the Commission adopts the following amendments to 39 CFR part 3001.

PART 3001—RULES OF PRACTICE AND PROCEDURE

1. The authority citation for part 3001 continues to read as follows:

Authority: 39 U.S.C. 404(b); 3603; 3622-24; 3661; 3662; 3663.

Subpart G—Rules Applicable to the Filing of Reports by the U.S. Postal Service

2. Revise § 3001.102 to read as follows:

§ 3001.102

Filing of reports.

Each report listed in this section shall be filed with the Secretary of the Commission within two weeks of its presentation for use by postal management, unless otherwise noted. Each report should be provided in a form that can be read by publicly available PC software. A processing program that was developed specifically to produce an accompanying workpaper must be provided in a form that can be executed by publicly available PC software. COBOL processing programs in use prior to FY 2003 are exempt from this requirement. The reports and information required to be provided by this subpart need not include matters that are exempt from disclosure by law. Whenever a specific source is cited in this section, that citation includes any successor or substituted source.

(a) Annual reports. The following information will be filed by the Postal Service annually. If it is presented for use by postal management at more frequent intervals, it shall be filed at those intervals:

(1) All input data and all processing programs that have changed since the most recently completed general rate proceeding, if they are used to produce the Cost and Revenue Analysis Report (CRA). Each change in attribution principles or methods from the previous report will be identified. The Postal Service shall submit a CRA-USPS Version, followed within two weeks by a CRA-PRC Version. Documentation of both versions of the CRA shall include, but not be limited to, the following:

(i) Spreadsheet workpapers underlying the development of segment costs by cost component. These workpapers should include the updated factors, and data from the supporting data systems used, including the In-Office Cost System (IOCS), Management Operating Data System (MODS), City Carrier Cost System (CCCS), Rural Carrier Cost System (RCCS), and National Mail Count.

(ii) Documentation of any special study that has a substantial effect on estimated costs in any cost segment and was not documented in the most recently completed general rate proceeding.

(A) Documentation shall consist of all input data and all processing programs used to obtain the results of the special study.

(B) The Postal Service may elect to provide a written or oral presentation describing the data and the estimating techniques used, as well as the results of the special study, and to apply for a waiver of the requirement in paragraph (a) of this section.

(2) Cost Segments and Components Report. Documentation shall include, but not be limited to, the following:

(i) Cost segments and components reconciliation to financial statements and account reallocations.

(ii) The Manual Input Requirement, the “A” report, and the “B” report;

(iii) The control string commands for the “A” report, the “B” report (including the PESSA factor calculations), and the “C” report;

(iv) The master list of cost segment components, including the components used as distribution keys in the development of the “B” report and the “C” report.

(3) City delivery information, including the number of routes by type, the number of possible deliveries by type, the number of collection boxes, and the number of businesses served (120 days from the close of the fiscal year).

(4) Rural carrier information, including the number of routes by type and miles, stops, boxes served, and mail pieces by route type (120 days from the close of the fiscal year).

(5) Civil Service Retirement Fund Deficit Report (two weeks after release of the Annual Report of the Postmaster General).

(6) Worker's Compensation Report, including summary workpapers (two weeks after release of the Annual Report of the Postmaster General).

(7) Annual Report of the Postmaster General.

(8) Congressional Budget Submission, including workpapers. The Postal Service will also file concurrently Summary Tables SE 1, 2, and 6 (coinciding with their submission to Congress).

(9) Audit Adjustment Vouchers, if any.

(10) Billing Determinants, at the level of detail employed in the most recent formal request for a change in rates or fees. The provision of billing determinants for Express Mail, Priority Mail, and parcel post may be delayed up to 12 months from the close of the fiscal year to which they apply.

(11) USPS Integrated Financial Plan.

(12) Input data and calculations used to produce annual Total Factor Productivity estimates.

(b) Quarterly reports. The following information will be filed by the Postal Service quarterly:

(1) Revenue, Pieces, and Weight, by rate category and special service.

(2) Origin/Destination Information Report National Service Index.

(c) Accounting period reports. The following information will be filed by the Postal Service each accounting period:

(1) Summary Financial and Operating Report.

(2) National Consolidated Trial Balances and the Revenue and Expense Summary.

(3) National Payroll Hours Summary.

(4) On-Roll and Paid Employee Statistics (OPRES).

(5) Postal Service Active Employee Statistical Summary (HAT report).

(d) Miscellaneous reports. The following information will be filed by the Postal Service:

(1) Before/After Pay Increase Reports.

(2) Before/After COLA Cost Reports.

(3) A master list of publications and handbooks including those related to internal information procedures, when changed.

(4) Data collection forms and corresponding training handbooks, when changed.

(5) Notice of changes in data reporting systems, 90 days before those changes are implemented.

Footnotes

1.  Throughout this order references to “financial information” include cost, volume, and revenue information in aggregate, and for individual products, unless a narrower meaning is indicated.

2.  The Postal Service did not analyze specific language of the proposed rule in its substantive comments. It complained generally, however, that the proposed rule is “broadly worded,” and cited this as one ground for concluding that it would require it to provide documentation that is comparable to an omnibus rate case in its “scale and scope.” Substantive Comments of the United States Postal Service (July 2, 2003) at 4, 22. Presumably it was referring to proposed paragraph 102(a)(1), since the remaining language in the rule is quite specific.

3.  See PRC Order No. 141 (October 21, 1976) at 6-8.

4.  In 1997, the Postal Service discarded the LIOCATT-based method of distributing mail processing costs in favor of the much more complex MODS-based method. In its recent submissions under the Periodic Reporting Rule it provides much less documentation of its new, complex method than it had been providing for the older, simpler method.

5.  For example, in 1998, the Postal Service voluntarily provided a fully documented CRA reflecting Commission-approved attribution methods for FY 1997.

6.  For example, the Postal Service spent a major portion of the most recent fully-litigated omnibus rate case (Docket No. R2000-1) producing and defending Level Five and Level Six documentation for the Engineered Standards data on which the Postal Service based its attribution of carrier street time labor costs.

7.  Historically, the Postal Service has rarely incorporated major new attribution models or distribution techniques into its interim-year CRAs, because they have not been scrutinized in an omnibus rate case. In its FY 2002 CRA, the Postal Service apparently has departed from this traditional practice by incorporating major new attribution models in the areas of carrier street time labor and in facilities costs in the FY 2002 CRA before they have been presented in an omnibus rate case. This coincides with significant shifts in subclass attributable cost shares of the effected cost components. There is no way for the outside world to interpret these shifts, however, because the undocumented FY 2002 CRA provides no way of distinguishing between shifts in attributable costs, and shifts in the techniques that the Postal Service uses to measure attributable costs. If the outside world had the ability to replicate the Postal Service's cost attribution model, it could run the model with FY 2002 data using the established method, and again using the new method. This would provide a reasonable basis for separating changes in underlying economic activity from changes in the methods by which they are measured.

8.  Sometimes this is a two-step process where component costs are modeled to see what portion varies with an intermediate cost driver, then that portion is modeled to see how much of it varies with volume.

9.  This may be described as “piggybacking” the indirect costs on the direct costs. For example, the CRA model spreads the costs of supervising city carriers to subclasses in the same proportion as the B workpapers distribute the cost of city carrier direct labor to subclasses. It does this separately for each of the numerous in-office and street time components of city carrier direct labor costs.

10.  The Van-Ty-Smith SAS programs that construct mail processing labor cost distribution keys from current-year IOCS tally data, stop coverage in the CAT/FAT study of coverage variability, and pieces-per-delivery in the Load Time Variability study are examples of studies whose accuracy depends upon using updated input data.

11.  One way to diagnose a failure to update a special study is to compare the Postal Service's results to those obtained by rerunning the program with old data. This diagnostic tool is available to an outside analyst only if he or she is able to replicate the program.

12.  It should be borne in mind that the burden of documenting new cost studies is not increased by the Periodic Reporting Rule, since it is part of the burden of preparing an omnibus rate request.

13.  The Commission has restricted the Periodic Reporting Rule to documentation of the mechanics by which the CRA is produced (datasets, processing programs, spreadsheets, etc.) rather than justifications of theories or policies that are likely to be contested in a rate case. In this respect, the documentation performs a function similar to an informal technical conference held off the record during a rate case. The purpose of such conferences is to gain an understanding of what was done mechanically to implement a particular analysis, and avoids questions touching on the merits of the analysis. Restricted in this way, the Periodic Reporting Rule strikes a reasonable balance among the Postal Service's right to “surprise” intervenors with every aspect of its support of proposed rates, the intervenors rights to comprehend the Postal Service's rate request and respond to it with alternatives in the brief time allotted, and the Commission's need to enter a rate case already understanding how the Postal Service prepares its routine financial reports. If the Postal Service believes that its reasons for making changes to the CRA should be explained in order to eliminate misconceptions, it is free to provide them. The Commission does not require such explanations, in order to minimize the burden of complying with the rule.

14.  The Commission disagrees that the rule seeks enough documentation to make it feasible to develop anticipatory rate recommendations. For example, two categories of inputs that would be needed to develop a realistic alternative rate schedule (if a test year and revenue requirement were known) are the appropriate DRI inflation factors, and details about the magnitude and timing of the Postal Service's cost reduction programs. Neither is required by the Periodic Reporting Rule.

15.  In its Substantive Comments, at 12-13, the Postal Service mentions two instances in the Commission's 30 year history in which a reviewing court remanded a Commission Recommended Decision on the ground that it had employed an analytical technique without observing the full range of procedural safeguards required in formal hearings. This frequency of remand is probably among the lowest of any Federal regulatory body, and does not offer legitimate grounds for presuming that the Commission will ignore procedural safeguards that accompany formal hearings.

16.  Public Citizen Health Research Group v. FDA, 704 F.2d 1280, 1285 (D.C. Cir. 1983) quoting 2A C. Sands, Statutes and Statutory Construction § 46.06 (4th ed. 1973).

17.  The Commission's expedited rules of practice incorporating abbreviated hearing schedules were adopted at the request of the Postal Service. They strain the due process protections that the Act provides to intervenors to the limit. So far, the intervenors have cooperated in this attempt to accommodate the Postal Service's need for speed and flexibility in ratemaking and classification procedures. The Commission's expedited rules ask the intervenors to assert all of the formal rights that they have under highly abbreviated record hearing procedures in order to meet the Postal Service's need for flexible ratesetting. It is ironic that after receiving the voluntary cooperation of the intervenors in foregoing some of rigidities of the statutory hearing process that benefit them, the Postal Service so adamantly seeks to retain all of the rigidities of the formal hearing process that benefit the Postal Service. These include its perceived right to surprise intervenors with all aspects of its rate requests, thereby maximizing the time pressure under which they must litigate, and minimizing their access to meaningful due process. See Substantive Comments at 36.

18.  It is worth noting that when outside auditors review the Postal Service's CRA, they perform a “process audit” that is designed only to confirm that the processing steps that are intended to be performed are in fact performed. It does not involve a conceptual audit that addresses the suitability of the estimation methods used or the reasonableness of the results obtained. Review of the CRA by the Office of the Inspector General, or by GAO, are generally not conceptual audits either. There does not appear to be any provision in the current regulatory regime for regular conceptual audits of the Postal Service's cost, volume, and revenue estimates like the ones that the Commission performs intermittently during omnibus rate cases.

19.  The existing Periodic Reporting Rule contains a long list of routinely gathered financial and operating information that is prepared for the benefit of postal management, but also gives the Commission and the interested public useful background information that will help them process a rate case when it is eventually filed. Almost any of the categories of information covered by the rule are the kind that could be expected to eventually be discussed and analyzed in a rate case. For example, for decades, the rule has included accounting period financial reports, reports on revenue, pieces, and weight for groups of mail (RPW), various management plans, and the CRA and the Cost Segments and Components report, with partial documentation. For decades the rationale for including these reports in the rule was that they help the Commission and the affected public understand the kind of information, but not the specific information, that will be used to support rates when a case is eventually filed. This helps the Commission process a rate case more efficiently and more fairly. This is the principal effect that the rule has on the way that a Postal Service rate request is handled.

20.  This argument, of course, ignores the various kinds of hearings that the Act authorizes the Commission to initiate between rate cases.

21.  Section 3624(c)(1) was adopted as part of the 1976 amendments to the Act. The legislative history of this provision indicates only two motives for adopting the 10-month time limit for completing rate cases. The overriding motive was the desire to shorten rate cases so that revenues could be increased more quickly, and the financial crisis that prompted the 1976 amendments would not reoccur. See Report of Committee on Post Office and Civil Service, House of Representatives, on H.R. 8603, Postal Reorganization Amendments of 1976, Pub. L. 94-421, 94th Cong., 2d Sess., October 1976, at page 334 (Remarks of Senator McGee). A secondary motive was to reduce the Postal Service's control over the ratesetting process by lengthening the period that the Postal Service must wait before it puts temporary rates into effect, from 90 days to 10 months after it files a rate request. Id. at 51. The OCA asserts that under the original statute, temporary rates, as a practical matter, became permanent rates, shortening the hearing time before de facto permanent rates were implemented to 90 days. It contends that the effect of the 1976 amendments was to greatly expand the opportunity of intervenors in rate cases to influence the selection of permanent rates. OCA Reply Comments at 4, n.9.

22.  Section 101(a), states:

The United States Postal Service shall be operated as a basic and fundamental service provided to the people by the Government of the United States, authorized by the Constitution, created by Act of Congress, and supported by the people. The Postal Service shall have as its basic function, the obligation to provide postal services to bind the Nation together through the personal, educational, literary, and business correspondence of the people. It shall provide prompt, reliable, and efficient services to patrons in all areas and shall render postal services to all communities. The costs of establishing and maintaining the Postal Service shall not be apportioned to impair the overall value of such service to the people.

23.  See National Association of Greeting Card Publishers v. USPS, 462 U.S. 810, 821 (1983) quoting S. Rep. No. 91-912 at 4 (1970).

24.  See Letter of May 30, 2003, from Daniel J. Foucheaux, Chief Counsel, Ratemaking, to the Hon. Steven W. Williams, Secretary, Postal Rate Commission. This was apparently done on the assumption that the B workpapers for these cost segments are the current counterpart to the CRA documentation that it had provided in past years under the Periodic Reporting Rule.