| |
|
|
|
[4001] Under the Postal Reorganization Act, two principal statutory provisions frame the Commission's rate deliberations. First, the Postal Service operates under a break-even constraint. Thus, the Commission's recommended rates and fees are designed to generate sufficient revenues to recover, as nearly as practicable, total estimated test year costs. 39 U.S.C. § 3621. Second, the recommended rates are based on the nine ratemaking criteria specified in section 3622(b).1 The statute also identifies certain public policy considerations, which, within the Commission's discretion, may color its rate recommendations. See, e.g., §§ 101(d) and 403(c).
[4002] The nine ratemaking criteria of section 3622(b) are as follows:
(b) Upon receiving a request, the Commission shall make a recommended decision on a request for changes in rates or fees in each class of mail or type of service in accordance with the policies of this title and the following factors:
(2) the value of the mail service actually provided each class or type of mail service to both the sender and the recipient, including but not limited to the collection, mode of transportation, and priority of delivery;
(3) the requirement that each class of mail or type of mail service bear the direct and indirect postal costs attributable to that class or type plus that portion of all other costs of the Postal Service reasonably assignable to such class or type;
(4) the effect of rate increases upon the general public, business mail users, and enterprises in the private sector of the economy engaged in the delivery of mail matter other than letters;
(5) the available alternative means of sending and receiving letters and other mail matter at reasonable costs;
(6) the degree of preparation of mail for delivery into the postal system performed by the mailer and its effect upon reducing costs to the Postal Service;
(7) simplicity of structure for the entire schedule and simple, identifiable relationships between the rates or fees charged the various classes of mail for postal services;
(8) the educational, cultural, scientific, and informational value to the recipient of mail matter; and
[4003] Of these criteria, only criterion 3 is a requirement. See National Association of Greeting Card Publishers v. United States Postal Service, 462 U.S. 810, 820 (1983). It is the foundation of the Commission's rate recommendations, imposing two obligations on the Commission. First, recommended rates for each class or type of mail must be adequate to recover "the direct and indirect postal costs attributable to that class or type [of mail]." The Commission satisfies this requirement by recommending rates that recover attributable costs, which include volume variable costs and product specific costs, i.e., fixed costs associated with one class. Second, to enable the Postal Service to break even, the recommended rates must also be sufficient to recover "all other costs of the Postal Service," i.e., institutional costs. Recommended rates, therefore, must recover that portion of the institutional costs determined by the Commission to be "reasonably assignable to such class or type." Thus, criterion 3 establishes an attributable cost floor, and the recommended rates must, in total, exceed attributable costs sufficiently to enable the Postal Service to recover its institutional costs.2
[4004] As in prior cases, the issue of attributable cost levels has generated considerable controversy in this proceeding. Several participants, including the Postal Service, OCA, UPS, MPA, and Time Warner have offered testimony. See, e.g., USPS-T-15, 16 and 17; Tr. 27/13144 et seq. (OCA witness Smith), Tr. 27/12770 et seq. (UPS witness Neels), Tr. 24/11211 et seq. (MPA et al. witness Glick), Id. at 11260 et seq. (MPA et al. witness Cohen), and Id. at 11344 et seq. (Time Warner et al. witness Stralberg). The Commission's conclusions regarding costing are contained in Chapter III.
[4005] Costs not classified as attributable are classified as institutional. The Commission applies the remaining (non-cost) criteria of § 3622(b) to assign the institutional cost burden among the various classes and types of mail. These non-cost criteria are quite broad, suggesting both standards of efficiency and equity. Indeed, as the Commission has previously observed, the non-cost (or pricing) criteria serve sometimes-conflicting objectives, e.g., one criterion may suggest lower rates for a particular type of mail, while another may suggest the opposite result. See PRC Op. R94-1, App. F at 17, PRC Op. R90-1, para. 4001, PRC Op. R87-1, para. 4096, and PRC Op. R84-1, para. 4000. The Commission considers each criterion, exercising its informed judgment to balance the competing objectives of the Act in a manner that will result in fair and equitable rate recommendations.
[4006] In prior opinions, the Commission has discussed and reviewed the statutory ratemaking criteria. For example, in Docket No. R87-1, the Commission extensively discussed the ratemaking process, including how the various non-cost criteria are incorporated in its recommended rates. See PRC Op. Docket R87-1, para. 4022 et seq.; see also PRC Op. Docket R90-1, para. 4000 et seq. and PRC Op. Docket R97-1, para. 4001 et seq. The Commission's intent in doing so has been twofold; first to provide sufficient detail so that participants may discern the Commission's interpretation of the criteria, and second, to serve as benchmarks for evaluating whether the new evidence warrants a departure from prior allocations.
[4007] The relative institutional cost burden borne by each class or subclass may be measured in various ways. The process of marking up attributable costs to recover institutional costs yields a cost coverage for each subclass. Cost coverage provides a simple measure of the relative institutional cost burden borne by the different subclasses.3 From case to case, cost coverage for one or more classes or subclasses is likely to change, sometimes substantially, due to changed circumstances, e.g., new or modified mail processing operations, sharply increased costs, or classification changes. Thus, to measure relative burdens over time, the Commission employs a markup index, which compares the markup for each subclass to the systemwide average markup. The markup for each subclass is its contribution to institutional costs as a percent of its attributable costs. As discussed below, markups, like cost coverage, may be affected by changed circumstances. Hence, any evaluation of markups over time must account, to the extent practicable, for changed circumstances. Each of these measures of relative burden is presented in Appendix G.
[4008] Postal Service witness Mayes addresses application of the pricing criteria to rate levels proposed by the Postal Service for the various subclasses. See USPS-T-32. Witness Mayo applies the pricing criteria to fee levels proposed for various special services. See USPS-T-39. Several participants advocate that the Commission should give certain non-cost criteria greater weight in recommending (higher or lower) rates. The following are illustrative. GCA and Hallmark argue, inter alia, that equal weight should be given to all the non-cost criteria, and urge "a fuller more effective application of the `ECSI' criterion to First-Class Mail." GCA/Hallmark Brief at 7; see also id. at 8 and 10.4 UPS asserts, based primarily on value of service considerations, that the cost coverage for Priority Mail should be at least equal to that of First-Class Mail. UPS Brief at 43 et seq. On the other hand, and also based on value of service considerations, APMU suggests that Priority Mail's cost coverage be reduced. APMU Brief at 8 et seq. AAP contends that Bound Printed Matter warrants a sharply reduced cost coverage based on ECSI value. AAP Brief at 3-8. MOAA makes a similar contention. MOAA Brief at 23-25. In addition, MOAA advocates a lower cost coverage and, therefore, lower rates for Standard Mail (A) ECR, citing criterion 6, mail preparation, and criterion 2, value of service considerations. As a general matter, proposed markups for classes, subclasses, and services are addressed in Chapter V. However, because of its general applicability, GCA/Hallmark's suggestion that the Commission give "equal weight to all the non-cost factors" merits brief comment. GCA/Hallmark Brief at 10.
[4009] Citing Direct Marketing Association, Inc. v. United States Postal Service, 778 F.2d 96, 104 (D.C. Cir. 1985), GCA and Hallmark contend that "[t]he Act does not give primacy to any single factor but requires that each be given equal weight." GCA/Hallmark Brief at 8. This interpretation misconstrues Direct Marketing, which, while concluding that each non-cost factor ranked equally, held only that "[a]ll factors must be considered, . . . " DMA v. USPS, 778 F.2d 96, 104 (citation omitted). Moreover, the suggestion that the non-cost factors be given equal weight is contrary to the statute since it would effectively strip the Commission of its discretion to apply the non-cost criteria of the Act in a fair and equitable manner. See United Parcel Service, Inc. v. United States Postal Service, 184 F.3d 827, 845 (D.C. Cir. 1999) (citation omitted). ("While the Commission must `take into account all relevant factors and no others,' it need not give each factor equal weight."); see also Association of American Publishers, Inc. v. United States Postal Service, 485 F.2d 768, 774 (D.C. Cir. 1973). ("The [statutory] factors are reminders of relevant considerations, not counters to be placed on scales or weight-watching machines.") Hence, the Commission rejects this suggestion.5
[4010] As noted above, attributable cost serves as a floor which the Commission marks up to determine the reasonable contribution to all other costs. In this proceeding, witness Bradley presents a new method for calculating incremental costs, which he urges the Commission to adopt in lieu of attributable costs. See USPS-T-22. The Postal Service employs incremental costs as a means of testing for cross-subsidy. In Docket No. R97-1, the Commission accepted the incremental cost test described by witness Panzer.6 In that proceeding, however, the Commission rejected witness Takis' calculation of incremental costs. See id., para. 4053. Witness Bradley's work, along with witness Kay's calculations, represents an earnest response to the Commission's concerns with Takis' effort. The task of developing reliable incremental costs for the Postal Service, a multi-product regulated entity with public service obligations, is daunting. This is not to imply that it cannot be achieved. However, for the reasons discussed below, the Commission declines to employ the new method of calculating incremental costs espoused by witness Bradley. Nonetheless, the Commission is satisfied, based on this record, that its recommended rates are subsidy free, consistent with the statute. As the Commission observed in Docket No. R97-1, its calculation of attributable costs by subclass is a reasonable proxy for the incremental costs associated with that subclass or type of mail. Thus, "nonnegative markups are good evidence against the presence of the most elementary cross subsidies." Id., para. 4024.
[4011] In sum, the Commission's recommended rates for each class or type of mail must recover its attributable costs, plus a reasonably assignable portion of all other costs. Under the Commission's costing methodology, approximately 37 percent of total costs are classified as institutional, to be reasonably assigned among the various classes and types of mail. In fulfilling that statutory obligation, the Commission balances the competing criteria of the Act to recommend rates that are fair and equitable. This process follows the Commission's long-standing rate setting practices.
[4012] This proceeding presents three overarching pricing challenges. Each has significant ratemaking implications. The most broad reaching issue is the first ounce rate for First-Class Mail. Lesser, but nonetheless significant issues include large cost increases affecting certain classes and subclasses of mail and newly enacted legislation affecting the costs and rates for several subclasses of mail. Each is addressed below.
[4013] Ratemaking is an iterative process which, as the Commission has explained in prior opinions, involves developing target coverages expressed, initially, in general terms, e.g., near or slightly below average. That analysis begins with reference to the existing, presumptively reasonable rate structure. In addition, the Commission is cognizant of its prior recommendations. In other words, the Commission evaluates the Postal Service's Request and the intervenors' proposals in light of their affect on existing rates and also with an eye on historic relationships. Both play a role, the former to assess current developments and impact, the latter as a relative benchmark.
[4014] However, with respect to the latter, changes in postal operations, mail classifications, rate relationships, and markets may, over time, cause changes in the Commission's coverage determinations which any comparison of relative class burdens would need to consider.
[4015] The basic First-Class rate, i.e., for mail weighing one ounce or less, has always been designed in whole integers. The practice is premised on simplicity and administrative convenience. The rate is used by the general public and small businesses. Postage is often purchased in small increments. It is simple to administer. No participant proposes that the Commission recommend a fractional first ounce rate.7
[4016] The whole integer convention directly affects the ratemaking process. First-Class letter mail is the most commonly used rate, affecting nearly 100 billion pieces of First-Class Mail, or approximately 48 percent of the Postal Service's total volume. A one-cent change in the First-Class rate will generate approximately one billion dollars, 0an amount that far exceeds the proposed adjustments to other subclasses and services of mail. Compounding the issue, in this proceeding, unlike some prior ones, the Postal Service has proposed to increase the first ounce First-Class rate by one cent. Thus, at the outset, in considering its pricing options, the Commission's First-Class target rates are largely limited to retaining the current $.33 rate or recommending the one cent change proposed by the Postal Service.
[4017] The Commission closely considered, but ultimately rejected, maintaining the current rate for two principal reasons. First, spreading an additional billion dollars to the other classes and services of mail would have required rate level changes that could, in the Commission's view, cause severe economic dislocation. Second, it would have required unacceptably large increases in the other classes. Some participants, for example, suggest that the markups for Standard A Mail be increased. However, the Commission's above systemwide average increase for Standard A Mail Regular largely reflects the increasing attributable costs for that mail. A further increase necessitated by retaining the current First-Class rate would lead to an excessively high increase. Moreover, the Commission harbors some concerns that the increasing relative elasticities of demand for Standard A could render an increase of that size problematic. Nor could the remaining classes absorb the difference. Aside from any equity concerns, their volume is simply insufficient to sustain an increase of the magnitude that would otherwise be required. 8
[4018] The Commission recognizes its recommended $.34 first ounce rate will produce a greater institutional cost contribution than under the current rate. To moderate this, as discussed more fully in Chapter V below, the Commission has reduced the additional ounce rate, and, based on its analysis of the cost savings, adjusted certain workshare discounts. Moreover, the increase in the first ounce rate, 3 percent, is modest and substantially below the system average. Under the circumstances presented in this proceeding, the Commission's recommended rates for First-Class letter mail best satisfy the competing policies of the Act.
[4019] As noted above, changed circumstances, e.g., classification changes, may affect coverage levels, which, in turn, may cause markup relationships to change over time. There is some evidence this has occurred in First-Class Mail as the volume of workshared mail has increased. The following table shows the relative and absolute change in the mix of single-piece and workshared First-Class Mail for several years beginning with 1988. The volume data are from the Postal Service's RPW reports.
of Total[4020] The trend is evident. For the entire period measured, workshared volumes have increased both in absolute terms and relative to First-Class single piece.9 That trend continues into the test year as well. Test year single-piece volume is estimated to equal 52.9 billion, while workshare is estimated at 47 billion, or in percentage terms 47 percent of total First-Class letter mail.
[4021] As workshared letters have become a greater proportion of total First-Class Mail volumes, cost coverage for the class has generally increased over time. See, e.g., Tr. 22/10195 and Tr. 26/12459, 12646. In turn, this has caused its markup to increase as well. Id. at 10196 and 12459. For example, from Docket No. R87-1 through Docket No. R97-1 coverage for First-Class letters has increased from 158 percent to 172 percent. Its markup index has also increased over that time, from 1.200 to 1.308.10 These results stem, in large measure, from setting the discounts consistent with efficient component pricing. Other factors may have caused the markup index of other subclasses to vary over time as well.
[4022] The Commission recognizes the beneficial effects of the Postal Service's automation program on reducing processing costs of First-Class Mail. Conversions to workshare have also contributed to reduced unit costs. These lower costs have benefited First-Class mailers directly in the form of below average rate increases in Docket No. R97-1 and this proceeding. In both dockets, the Postal Service proposed only a one-cent increase in the first ounce rate. In this case, that represents only a 3 percent increase. This is not to suggest that the Commission has abandoned its goal of reducing the relative burden on the monopoly class. Indeed, as indicated, the Commission has taken steps to moderate the contribution by First-Class Mail.
[4023] Two participants, in particular, argue that First-Class Mail bears an excessive institutional cost burden. Neither adequately supports its claims. OCA advocates retention of the current First-Class rate. OCA Brief at 142 et seq. Citing witness Callow's testimony as support, OCA argues that the institutional cost burden on First-Class Mail has increased. While the testimony is laudable in concept, it is flawed in execution. In particular, it fails to account for intervening changes, such as those suggested above, that may cause markups to change. See also DMA Brief at 6-8. Nor does it account for circumstances that uniquely influenced the Commission's recommended rate levels. See, e.g., PRC Op. R94-1, para. 4107.
[4024] ABA & NAPM witness Clifton argues that the cost coverage for workshared First-Class Mail should be reduced because it is discriminatory relative to commercial Standard A Mail and to First-Class single-piece mail. Tr. 26/12458. To that end, he proposes to increase discounts for First-Class Mail with the reduction in revenues made up by increasing the cost coverage for commercial Standard A Mail. Id. at 12463. The comparison is unavailing. First, cost coverage is applied at the subclass level, not by rate category. Second, that rate differences exist is not enough. The Commission is concerned with undue discrimination. The simple comparison with Standard A Mail is insufficient to prove the point. Third, the increase in First-Class cost coverage over time, absolutely and relatively, may manifest changed circumstances, e.g., in postal operations or mail mix. Indeed, notwithstanding the increase, it would appear that First-Class Mail's relative contribution to institutional costs has remained relatively stable since 1990. Id. at 12747. These are among the factors that would need to be explored in greater depth to give any credence to witness Clifton's claim.
[4025] Finally, OCA's novel, rate stability proposal merits brief comment, and, more importantly, further study. The proposal, under which the single-piece First-Class (SPFC) rate would be held constant through two rate cases, is designed to provide household mailers with greater rate stability, while providing business mailers with smaller, but more frequent rate changes. OCA Brief at 182-83. According to OCA, it is "not intended to shift costs between classes of mail or otherwise adversely affect larger mailers." Id. at 183. The proposal is contingent on establishing an "SPFC Reserve Account," under which excess revenues in the first rate period would offset the need to increase the SPFC rate in the second period.
[4026] The Postal Service opposes the proposal, contending that its adoption would impinge on management prerogatives. Postal Service Brief at VII-86-89. The Postal Service, however, states that the "reserve account idea is not uninteresting." Id. at VII-88. DMA also opposes the proposal, asserting, inter alia, that it is "unworkable and probably unlawful." DMA Brief at 9-12. See also Postcom/MASA Brief at 8, fn. 1.
[4027] The record is not sufficiently developed to enable the Commission to fully address the merits of this proposal, including its policy and legal implications. Consequently, the Commission declines to recommend its adoption. The proposal is, however, intriguing and merits close attention by the Postal Service. To that end, the Postal Service is encouraged to take the initiative, as it did, for example, in organizing the Periodicals Operations Review Team, to further consider this and related rate design issues affecting First-Class Mail.
[4028] Throughout the proceeding, the Commission evaluates evidence submitted by the Postal Service and intervenors. This process enables the Commission to focus on issues that may require special attention. As it sees fit, the Commission will issue orders requesting additional testimony to explore specific issues. In this proceeding, the Commission's concern with increasing cost trends caused it to request the Postal Service to submit additional testimony concerning Periodicals and Media Mail (formerly Special Standard B).
[4029] In its initial Request, the Postal Service proposes a 12.6 percent increase, on average, in Periodicals rates. USPS-T-38 at 6, revised February 18, 2000. Several intervenors claim that the proposed increase is substantially greater. See Periodicals Mailers Brief at 1. In addition, they contend that the Postal Service's proposed rates will increase postage costs for users of Periodicals by approximately $300 million. Ibid.
[4030] In response to P.O. Information Request No. 4, the Postal Service provided the processing costs for various classes and subclasses of mail by shape, including Periodicals, for the period 1989 through 1999. Tr. 46-D/21807 et seq. After analyzing the data submitted by the Postal Service, the Commission issued Order No. 1289 requesting the Postal Service to provide detailed evidence explaining the causes of the increase in inflation-adjusted costs of processing Periodicals since 1993. PRC Order No. 1289 (March 28, 2000) at 1. In response, the Postal Service submitted the testimony of witnesses O'Tormey (USPS-ST-42) and Unger (USPS-ST-43).
[4031] To its credit, the Postal Service organized, along with industry representatives, the Postal Service Periodicals Operations Review Team (Review Team) following the Commission's opinion in Docket No. R97-1. See Tr. 24/11166 et seq. The Review Team identified more than $200 million in test year cost savings and reductions affecting Periodicals. See Tr. 38/17329. Notwithstanding these reduced cost levels, Periodicals attributable costs continue to increase. For example, from the test year in Docket No. R97-1 to the test year in this proceeding, Outside County unit costs are estimated to increase by 10.2 percent. Under these circumstances, a rate increase above the system average is unavoidable.
[4032] To the extent practicable, however, the Commission has minimized the increase by moderating Periodicals coverage. As In Docket No. R97-1, this result reflects the Commission's concern about the reported costs. In that regard, it represents a continuation of the reduced markup for Periodicals from Docket No. R97-1, and, in the same vein, is viewed as a temporary solution.11 The Commission is hopeful that, longer term, the Postal Service's efforts to reduce flat processing costs will bear fruit.
[4033] Media Mail also reported sharply higher costs. As discussed in greater detail in Chapter V, the Commission sought further explanation for the increase upon finding that the Postal Service's institutional response inadequately explained its causes. PRC Order No. 1300 (August 18, 2000). In response to this order, the Postal Service submitted the testimony of witness Degen who proposes to reduce FY 1999 Media Mail processing costs by 12.6 percent based on his finding that certain IOCS tallies had been misidentified. Tr. 45/20051-60. The Commission's recommended rates reflect this adjustment to the reported costs. In Docket No. R97-1, the costs warranted a 9.6 percent reduction, on average, in the Commission's recommended Media rates. In this proceeding, the reported costs require an increase. The Commission, however, has moderated its cost coverage based on consideration of the non-cost criteria, including, in particular, concern over the cost increases.
[4034] Large cost increases can play havoc with mailers expectations; they also impact the Commission's coverage deliberations under criterion 4, the effect of rate increases on the general public, business mailers, and private carriers. Plainly, cost increases outside the norm, e.g., in excess of inflation, wage rates, or costs for other postal products, not only limit the Commission's flexibility, but also raise concerns. Dramatic changes in costs from case to case appear more likely to affect smaller volume subclasses. This is not meant to imply that the reported costs are not valid, but simply that confidence in the data can be undermined without reasonable assurance that the data are reliable. Bound Printed Matter is a case in point.12
[4035] As more fully addressed in Chapter V, during its coverage deliberation, the Commission considers each of the pricing criterion of the Act. Cost increases, however, have frequently been the overriding consideration in this case. Measured from the base year in Docket No. R97-1 to the base year in this proceeding, BPM unit costs have increased by more than 40 percent. BPM costs have been contested and closely examined. They have not been shown to be inappropriate or otherwise unreliable for ratemaking purposes. Given the magnitude of this increase, however, the Commission is substantially reducing BPM's cost coverage from its historic levels, i.e., from approximately 136 percent in Docket No. R97-1 to approximately 114 percent in this proceeding. Under the circumstances, the coverage represents a reasonable assignment of institutional costs to BPM.
[4036] In sum, while the Commission's cost coverages are based on consideration of all the non-cost criteria of the Act, unique circumstances may compel the Commission to emphasize a particular criterion. The Commission has long practiced this policy when faced with facts that limit its ability to recommend higher (or lower) rates. See, e.g., PRC Op. Docket No. R87-1, para. 4027; PRC Op. Docket No. R90-1, para. 4017. The foregoing examples, in which the Commission's coverage was tempered, in particular by criterion 4, are further manifestation of this policy.
[4037] Newly enacted legislation, PL 106-384, alters the rate relationships between certain nonprofit and commercial subclasses of mail. The nonprofit (or preferred) subclasses include: Standard A Nonprofit and Nonprofit ECR, Classroom and Nonprofit Periodicals, and Library Mail. The rate for these subclasses is to be derived by reference to its corresponding commercial subclass.
[4038] For ratemaking purposes, PL 106-384 directs that the attributable costs of the commercial (or regular rate) subclass and corresponding preferred subclass be combined. The ratemaking criteria of § 3622(b) are to be applied to the combined costs to determine the regular rate. The preferred rates fall out of this process as follows:
· Standard A Nonprofit and Nonprofit ECR rates, overall, are designed so that the estimated average revenue per piece by subclass equals, as nearly as practicable, 60 percent of the estimated average revenue per piece of Standard Regular and ECR, respectively.
· Nonprofit and Classroom rates are designed so that the postage on each mailing of such mail is, as nearly as practicable, 5 percent lower than the postage for the corresponding regular (Outside County) rate.13
· Library rates are set so that the postage on each mailing of such mail is, as nearly as practicable, 5 percent lower than the postage for the corresponding Media rate.
[4039] In addition, PL 106-384 includes a transitional provision under which the estimated reduction in revenues from Nonprofit Standard (A) is, for purposes of this proceeding, treated as a "reasonably assignable" cost under criterion 3.
[4040] PL 106-384 is designed to address rate anomalies which were deemed to preclude application of the appropriate markup to the preferred subclasses. See S. Rep. 106-468, 106th Cong., 2nd Sess. 2-4 (2000). PL 106-384 attempts to preserve the preferred rate status of these subclasses through a different formula. The Commission's recommended rates reflect the recent amendments to the Act.
[4041] As part of its direct case, the Postal Service submitted the testimony of witness Bernstein regarding Ramsey pricing. USPS-T-41. The testimony provides, inter alia, "a guideline for postal pricing based on the principle of economic efficiency." Id. at 3. Several parties representing Standard Mail (A) interests endorse the use of Ramsey pricing for setting postal rates. These include MOAA (MOAA Brief at 10-12), DMA (DMA Reply Brief at 10), and SMC (SMC Reply Brief at 3, fn. 1). MOAA, for example, urges the Commission to "give explicit consideration to Ramsey rate levels." MOAA Brief at 11. GCA and Hallmark oppose the use of Ramsey pricing, contending that it is inconsistent with the Act and that witness Bernstein's testimony is flawed. GCA/Hallmark Brief at 16 et seq. NAA also opposes the use of Ramsey pricing, arguing that Bernstein's presentation should be given no weight. NAA Reply Brief at 15. See also UPS Brief at 45, fn. 32. Witness Mayes, the Postal Service's pricing witness, made no formal use of the Ramsey prices developed by Bernstein. USPS-T-32 at 19.
[4042] Economic efficiency is neither the exclusive nor even the paramount ratemaking objective under the Act. Hence, the premise for using Ramsey pricing is dubious at best. Economic efficiency, the Commission has observed, "is not a justification for pricing in a way that might impair basic and fundamental postal services." PRC Op. R87-1, para. 4057. See also Direct Marketing Association, Inc. v. United States Postal Service, 778 F.2d 96, 103-04 (D.C. Cir. 1985). ("We disagree with the argument that Congress intended relative demand to be the benchmark for the assignment of institutional costs. Rather, it is clear that no single factor was intended by Congress to be the `primary' factor in making the assignments.") Under the Act, the Commission, exercising its informed judgment, must balance the competing ratemaking criteria of the Act. Application of a Ramsey pricing formula to a multi-product firm that includes captive, monopoly products, would be contrary to the policies of the Act as it would elevate one factor to the exclusion of all others. Therefore, consistent with its prior opinions, the Commission places no reliance on Ramsey pricing for its recommended rates. This is not to imply that the Commission pays no heed to own-price elasticity estimates in assessing value of service issues. The Commission's recommended rates reflect consideration of all relevant statutory criteria.
[4043] In Docket No. R97-1, the Postal Service developed incremental costs for all mail subclasses and for six pairs of subclasses. Witness Panzar presented a methodology for calculating the incremental costs and witness Takis carried out the calculations. As noted above, the incremental cost test is used to identify cross subsidies and, in Docket No. R97-1, the Commission accepted witness Panzer's description of that test, i.e., that "[t]he revenues collected from any service (or group of services) must be at least as large as the additional (or incremental) cost of adding that service (or group of services) to the enterprise's other offerings." PRC Op. R97-1, para. 4022. In Docket No. R97-1, the Commission made no use of witness Takis' incremental cost estimates because, inter alia, those estimates were based on cost models that the Commission rejected. Id., para. 4053. Nonetheless, the Commission commended the Postal Service's effort, highlighting certain advantages of its approach. Id., para. 4055.
[4044] In addition, the Commission identified several deficiencies in that approach, including, for example, that the six product combinations were inadequate, that the Postal Service failed to apply the incremental cost test, and the difficulty in converting from base year to test year using a simple ratio approach. Id., para. 4056.
[4045] Witness Bradley endeavors to address the Commission's concerns. Thus, for example, the number of product combinations examined is expanded to 32; witness Mayes applies the incremental cost test; Exhibit USPS-32E; and, in lieu of a simple ratio method, the new method separates volume variable and fixed costs and applies a roll-forward factor to each. See USPS-T-22 at 42-45.
[4046] In general, the Postal Service calculates base year incremental costs as the costs that are avoided when one or more products are eliminated while the remaining postal products are still provided without changes in the operating plan. In Docket No. R97-1, the Commission questioned the validity of the assumption that the operating plan would remain unchanged if a large subclass or combinations involving substantial volumes of mail were eliminated. PRC Op. R97-1, para. 4056. Witness Bradley notes the Commission's observation concerning this no reconfiguration assumption and suggests that "the problem may not be as general as it first seems." USPS-T-22 at 48. In support, he points to his Table 3 as illustrating the "relatively few instances in which a very large proportion of the driver is caused by a single subclass." Ibid.
[4047] Reference to Table 3 does not dispel the view that elimination of a large subclass or combinations involving substantial volumes would render the operating plan irrelevant. Certainly, the assumption is doubtful concerning any combination including the elimination of First Class or Standard A. Moreover, it is questionable whether the plan would remain unchanged if other subclasses or combinations, e.g., Parcel Post or Priority involving a significant portion of some cost component, were eliminated. The suggestion that the problem may not be as great as it seems is not sufficiently supported. Table 3 indicates the share of the largest product in 26 components ranges from 16.1 percent to 58.3 percent, most or all of which appear to be large enough to impact operating plans if eliminated.
[4048] The major concern with the Docket No. R97-1 proposal was a dependence on the accuracy of assumed or fitted cost functions over considerable volume ranges. The cost functions were used to sum the marginal costs avoided from eliminating product(s). Since the Commission rejected the cost functions proposed for mail processing, it was also necessary to reject the incremental costs calculated with them. In this proceeding, the Postal Service attempts to remedy the dependence on cost functions by adopting the Commission's procedures for rolling forward base year costs to the test year.
[4049] In the rollforward process, the volume effect calculations essentially assume variabilities to be constant over the range of anticipated volume changes. Witness Bradley endorses, at least implicitly, the validity of this assumption. See Id. at 43. ("The new method of calculating incremental cost is entirely consistent with the established methodology for calculating test-year attributable costs.") Witness Bradley further buttresses the use of a constant variability, by noting that volume reductions from eliminating even the large postal products are within the range of data in the data sets used to calculate component cost variabilities: "In other words, data sets like HCSS, CCS and MODS all have variations in their `volume' variables that exceed 50 percent of the mean value." Id. at 46.
[4050] The assertion that the constant elasticity assumption frees the calculation process from the underlying cost function or method that is used to develop the variability is overly optimistic. Indeed, in another forum, witness Bradley indicates that the assumption of a constant elasticity is equivalent to assuming a cost function with an exponential form.14 Witness Bradley recognizes that the use of a constant variability provides an approximation of the true incremental costs. USPS-T-22 at 45. Thus, the validity of the Postal Service's assumption depends on how well the exponential cost function approximates costs.
[4051] While acknowledging that his method assumes constant elasticity, witness Bradley contends, based on previous research, that incremental costs are not sensitive to that assumption. Id. at 45-46.15 In the cited research, a simulation is used comparing the costs with the translog and constant elasticity functions to demonstrate a close approximation for selected cases. The results show a close approximation for the cases examined when eliminating volumes representing less than 15 percent of the total. The results are limited to the cases examined. However, the research indicates that when more than one cost driver is involved, biases can develop that may overestimate the incremental costs. See Bradley Research at 13-14. This may indicate a potential problem in the calculation of city carrier elemental load costs due to the dependence on both pieces delivered and shape.
[4052] For the smaller subclasses, the approximation of true cost function to an exponential form may be acceptable. However, what constitutes small and what level of approximation is acceptable is not yet apparent. Additional research is warranted on the sensitivity of incremental costs to the constant variability assumption, with simulations using a range of possible cost functions.
[4053] Lastly, the range of volume changes in the rollforward process is on the average 1.75 percent and on a subclass basis ranges from 0.5 to 8.5 percent. These changes are considerably less than the volumes that would be eliminated from dropping most postal products in calculating incremental costs. Therefore, it is questionable if the analogy to the rollforward process is applicable.
[4054] The Commission remains interested in continuing the development of the incremental cost test to the point that it can be applied to reliably identify cross subsidies in proposed rates. Witness Bradley's testimony improves upon witness Takis' Docket No. R97-1 application of the test, but still leaves the Commission uncertain about the impact of the assumptions of constant variability and the stability of operating plans when major postal services or major groups of services are eliminated. On the other hand, the test in the form proposed by witness Bradley ought to be a reliable test for cross subsidies among the small subclasses where the assumptions of constant variability and stability of the operating plan are less problematic.
[4055] On the whole, it appears to the Commission that the test proposed by witness Bradley would be most reliable where it is least needed. The Commission's attributable cost floor serves as an effective screen for small subclasses because incremental cost for individual subclasses equals the Commission's definition of attributable cost when marginal costs are constant. Marginal costs are approximately fixed for the small changes in cost drivers that are involved in applications of the incremental cost test for small subclasses, but probably not for the substantial changes that are involved for the large subclasses. The Commission has not employed witness Bradley's incremental cost test in this proceeding because it suspects that the results of the test may still be unreliable where deleting a subclass or combination of subclasses causes a large reduction in an important cost driver.
1Under § 3622, the Commission's authority extends to rates and fees. For purposes of this discussion, the term "rates" encompasses fees as well.
2Postal Service witness Mayes employs volume-variable and incremental costs. USPS-T-32 at 16-19. She marks up volume-variable costs and uses the ratio of revenue to volume variable cost for purposes of assessing revenue requirement burdens. Exhibit USPS-32B. Incremental costs are used to test for cross subsidy. Exhibit USPS-32E. Witness Bradley presents the Postal Service's method of calculating incremental costs. USPS-T-22. It is his goal "to encourage the Commission to adopt incremental costs in place of attributable costs in its costing analysis." Id. at iv. Witness Bradley's testimony is addressed below.
3Cost coverage reflects the contribution to institutional costs provided by a subclass, reflected as the ratio of revenue to attributable cost.
4ECSI value refers to the phrase "educational, cultural, scientific, and informational value" in criterion 8.
5Similarly, the Commission rejects FGFSA's suggestion that the Commission's analysis of the non-cost factors should begin with a uniform per piece contribution. FGFSA contends that Parcel Post's low value of service justifies a per piece contribution near the average. FGFSA Brief at 14. FGFSA's proposal is flawed since the average contribution per piece is heavily influenced by lightweight, letter-sized First-Class and Standard (A) Mail. Thus, a simple comparison of unit contributions fails to reflect different handling or piece characteristics, e.g., shape, weight, and distance transported, which would justify a greater or lesser unit contribution.
6"`The revenues collected from any service (or group of services) must be at least as large as the additional (or incremental) cost of adding that service (or group of services) to the enterprise's other offerings.'" PRC Op. R97-1, para. 4022.
7The Commission recognizes that OCA's rate stability proposal contains elements of a fractional rate.
8Witness Bentley, representing MMA, a coalition of large first-class mailers, nonetheless recognizes the inherent difficulties involved in retaining the current rate. The remaining mailers would be required to bear the burden of the entire increase, a result that "would be very difficult for all other mailers to do." Tr. 26/12281.
9Based upon the Commission's analysis of the RPW data, for the period 1978 through 1999, the average annual growth rate is 0.4 percent for single-piece and 15.3 percent for workshared First-Class letter mail. For the period 1990 through 1999, the average annual growth rates are (0.4) percent and 5.2 percent, respectively.
10See PRC Op. Docket No. R97-1, Appendix G, Schedule 3. The results from Docket No. R97-1 are slightly below those from Docket No. R94-1, a result that may stem from the intervening classification proceedings, including Docket No. MC95-1.
11Periodicals Mailers urge adoption of the Postal Service's variability analysis. Periodicals Mailers Brief at 36 et seq. As discussed in Chapter III. A., the Commission adheres to its long-standing conclusion regarding mail processing variability. While the resulting cost coverage reflects the Commission's somewhat higher attributable cost levels, that low coverage is critically dependent on the circumstances of this proceeding. Without the uncertainty surrounding Periodicals costs, a markup closer to historic levels may have been warranted regardless of the underlying variability analysis.
12In Appendix H the Commission comments on and suggests refinements to the Postal Service's data estimation systems. In a system with more than 200 billion pieces, accurately sampling and reporting data are monumental tasks. The problem is perhaps most acute with respect to relatively smaller subclasses and types of mail since the effects of sampling or non-sampling errors may be magnified. An examination of the Postal Service's current practices in this area would appear to be useful in various ways, including the ratemaking process.
14See Michael D. Bradley, Jeff Colvin, and John Panzar, "Issues in Measuring Incremental Costs in a Multi-function Enterprise, Managing Change in the Postal and Delivery Industries, Kluwer Academic Publishers, 1997 at 9 (Bradley Research).
 Postal Rate Commission http://www.prc.gov Voice: (202) 789-6800 Fax: (202) 789-6886 prc-admin@prc.gov |
| |
|
|
|