Chapter5F.html

[5933] Introduction. The Postal Service proposes fee and classification changes to most special services. The proposed classification changes are summarized below:

· A proposal to allow customers who return a parcel to the shipper using merchandise return to purchase insurance, to eliminate the current per piece fee for merchandise returns, and to establish an annual advance deposit account fee for merchandise return;

[5934] DMCS Proposals. In the Opinion in Docket R97-1, the Commission accepted a Postal Service proposal to reorganize the DMCS special services sections into seven categories: Addressing (910), Delivery Alternatives (920), Payment Alternatives (930), Accountability and Receipts (940), Parcel Handling (950), Stamped Paper (960), and Postal Money Orders (970). A separate fee schedule (1000) is provided for permits. Subsequently, the Commission recommended the addition of a category for Acceptance Alternatives (980). In the instant docket, the Postal Service reviews each of the DMCS special services sections and offers proposals to rewrite each section for clarity, consistency, and organization. The proposals generally improve the readability of the DMCS special services sections. Therefore, the Commission has incorporated a majority of the Postal Service proposals into its DMCS recommendations. The Commission recommendations for the DMCS text appear in Appendix Two and recommendations for the DMCS Fee Schedules appear in Appendix One.

[5935] Special Services Cost Coverages. The DMCS category structure provides a convenient framework for review of the special services cost coverages. Services that have similar functions, and that theoretically might have similar cost coverages are grouped within categories. However, note that the specifics of each service cause the individual service to diverge from the ideal cost coverage for the category. For instance, several of the proposed rate increases have been tempered by the negative effect of a rate increase upon the general public (criterion 4), resulting in lower than ideal cost coverage.

[5936] Addressing (910). Addressing services generally result in lower Postal Service mail processing costs by providing correct delivery address information to mailers. This directly affects the degree of preparation of mail for delivery into the postal system and reduces the overall costs to the Postal Service (criterion 6). The addressing services make for an efficient mail stream that justifies a low markup. One of the addressing services, address changes for election boards and registration commissions, aids election boards in carrying out their vital function in our democracy. This service justifies the lowest markup possible that still covers costs (criterion 9).

[5937] Delivery Alternatives (920). The post office box and caller service is a high value premium service that can justify a high cost coverage (criterion 2). Post office box service is optional for most mailers. Mailers pay box fee rate "E" when post office box service is the only delivery option. The Postal Service proposes, in this docket, to restructure the post office box service to have the rates more accurately reflect costs (criterion 3). The Postal Service has capped the average increase it proposes due to the fee restructuring to avoid an excessive adverse impact on post office box users. Capping the rate changes considers the effect of the proposed rate increases on the general public (criterion 4). This consideration tends to lower the implicit cost coverages within some of the post office box fee groups to the moderate range. It may take one or more rate proceedings to more fully align rates with costs, and reflect the proper cost coverage in all fee groups.

[5938] Payment Alternatives (930). Payment alternative services generally facilitate the use of other mail services. Except for on-site meter setting service, a permit fee and an accounting fee are integral parts of each service. In determining a cost coverage for these services, the Commission is concerned with adequately covering the cost of the service and providing a moderate contribution to institutional costs (criterion 3).

[5939] Accountability and Receipts (940). Services in this category are generally high value services that enhance the value of the underlying host mail piece to the mailer (criteria 2). This could justify a moderate to high cost coverage for each of the included services. However, other factors tend to lower the cost coverage for several of the services. For example, the impact a rate increase may have on the mailer influences the cost coverage for the collect on delivery service (criterion 4). The Commission considers users of this service may be people of modest means, with few alternatives for business transactions. Therefore, a lower cost coverage is justified. Delivery confirmation is a relatively new service and a modest cost coverage is recommended to allow the service to develop to its fullest potential. The costs for certified mail have risen to such an extent that setting a rate that represents the true value of this service may create consumer sticker-shock. Again, this results in a lower cost coverage for certified mail. The differences and interrelationships among the services within this category also influence the value of service, and tend to lower the cost coverage below the high cost coverage that is otherwise justified.

[5940] Parcel Handling (950). Parcel airlift and special handling require individual consideration when discussing cost coverages. Both services are relatively low volume, and as such, it is difficult to accurately predict their costs. The services also are used only by limited groups. Parcel airlift provides service to certain military post offices on a space available basis. It has a lower value of service because there are alternatives that are more expeditious for sending parcels (criterion 2). The effect of a rate increase on the mailer also justifies a lower cost coverage because this service is used by military personnel that may not be able to afford a higher priced service (criterion 4). In Docket R97-1, an organization of poultry producers successfully argued to not increase the rates for special handling. Their argument was based on the lack of reliable cost information to justify that increase (criterion 3). This reasoning continues to justify current special handling rates.

[5941] Stamped Paper (960). The stamped envelope and stamped card service provides a high value of service because of the convenience it offers postal customers by allowing the purchase of a limited quantity of mailing supplies while at the post office (criterion 2). When purchased in greater quantities the stamped paper service may save postal customers the expense of applying stamps to envelopes and the expense of keeping an inventory of stamps. This could justify a high cost coverage. Balancing the justification for a high cost coverage is the effect of the rate increase on the general public, especially in the case of stamped cards, that tends to lower the final coverage into the moderate range (criterion 6). The Commission's philosophy of maintaining First-Class Cards as a low cost method to send mail also applies to keeping the stamped card rate as low as possible. The integer rounding constraint frequently restricts the Commission in its ability to set a precisely appropriate cost coverage, thereby resulting in coverages that may be above or below the target coverage.

[5942] Postal Money Orders (970). The postal money order service is a high value service because of the quality of the product in both its recognition and negotiability (criteria 2). There apparently are many available alternatives to the Postal Service product of varying qualities (criterion 5). These factors argue in favor of a higher cost coverage. However, the Commission balances the above factors against the effect of a rate increase upon the general public and moderates rates to maintain a somewhat lower cost coverage than the value of service would otherwise indicate (criterion 5). For example, military personnel that purchase APO-FPO money orders are providing a vital service to our country and may have difficult absorbing price increases. The Commission also continues to view purchasers of domestic money orders as people of modest means with a more limited access to alternative financial vehicles. Thus, the Commission's goal is for a moderate cost coverage for the postal money order service.

[5943] Acceptance Alternatives (980). Mailing Online is the only acceptance alternative special service. A comprehensive review of the factors of the act is provided in PRC Op. MC2000-2. The cost coverage for this service is set near the system-wide average. This recommendation was partially based on Mailing Online being a high value experimental service. After operations experience is available, the appropriate level of cost coverage for this will be reexamined.

[5944] Special Services Discussion: A discussion of each special service follows. Separate discussions on permit fees and the DMCS rewrite proposals are included. The recommended fees for special services appear in Appendix One.

[5945] Address Correction Service provides a mailer with an addressee's correct or forwarding address if known, or a reason why the Postal Service is unable to deliver a mailpiece as addressed. Both manual and automated services are available. Manual address correction service provides a mailer with a photocopy of the mailpiece along with the correct address information on Form 3547 (for First-Class Mail, Standard A Mail, or Standard B Mail), or the cover sheet of a periodical with the correct address information attached on Form 3579 (for Periodicals). Automated address correction service provides a mailer with correct address information or a reason for non-delivery electronically-accessible by the mailer using a computer and a modem. Periodicals automatically receive address correction service for 60 days following a change-of-address order. Address correction service is available by itself, or in conjunction with forwarding and return service. USPS-T-39 at 10-14.

[5946] The Postal Service projects test year costs, including contingency, of 13.4 cents for automated address correction and 55.8 cents for manual address correction. USPS-T-29 at 5. The Service proposes retaining the current fee of 20 cents for automated address correction and increasing the fee of 50 cents to 60 cents for manual address correction. USPS-T-39 at 10, 13.

[5947] The Commission recommends the Service's proposed fees of 20 cents for automated address correction and 60 cents for manual address correction. These rates produce a 125 percent cost coverage. This modest cost coverage is appropriate for a service that results in lower Postal Service mail processing costs by providing correct delivery addresses to mailers.

[5948] The Postal Service projects test year costs, including contingency, of $71.15 per one thousand addresses for sorting of mailing list address cards by five-digit ZIP Code. USPS-T-29 at 26. The Service proposes increasing the current fee of $70.00 per one thousand addresses contained in a mailing list to $73.00 per one thousand addresses contained in a mailing list. USPS-T-39 at 162-164. The Commission recommends the $73.00 per one thousand addresses fee proposed by the Postal Service. At these rates, test year costs will equal $70.45, producing a 104 percent cost coverage.

[5949] Eligible mailers may submit a name and address list, or an occupant list (address only list) to the Postal Service for correction. The mailing list may be submitted on cards or on sheets of paper, and must be separated by Post Office or carrier route as required. Name and address list corrections include eliminating names to which mail cannot be delivered or forwarded, providing forwarding information if an order is on file, correcting spelling of names and addresses, correcting ZIP Codes, post office box numbers, and rural box numbers, and providing the name of the head of household when two or more names appear on the list. Occupant list corrections include deleting invalid addresses, correcting the last lines of the address, placing directional signals to indicate carrier route information, and providing the number of units in multiple unit dwellings. Id. at 50-53.

[5950] The Postal Service projects test year costs, including contingency, of 23.2 cents per address for the correction of mailing list service. USPS-T-29 at 25-26. The Service proposes increasing the current fee of 20 cents per address to 25 cents per address contained on the mailing list. USPS-T-39 at 50-53. The Service further proposes to increase the minimum charge per corrected list from $7.00 to $7.50. Under the proposal, the number of addresses submitted per address list equivalent to the minimum charge effectively decreases from 35 to 30 addresses. The Commission recommends the Postal Service's proposed 25-cent fee per address and minimum charge of $7.50. At this rates, test year costs will equal 23 cents, producing a 109 percent cost coverage.

[5951] This service provides election boards and voter registration commissions with change-of-address information. Election boards and voter registration commissions also have the option of using a "Return Service Requested" endorsement or the National Change of Address system to gather similar address information. USPS-T-39 at 6-9.

[5952] The Postal Service uses projected test year costs for the correction of mailing lists as a proxy for the cost of address changes for election boards and registration commissions. USPS-T-29 at 25-26. The Service proposes increasing the 17-cent fee for this service to 24 cents. The Commission recommends a slightly lower fee of 23 cents for this service. At this rate, test year costs will equal 23 cents, producing a 100 percent cost coverage. The recommendation is consistent with prior opinions that require this service to cover costs, but only provide a minimal contribution to institutional costs, because "election boards serve a vital function in our democracy." PRC Op. R97-1, para. 5896.

[5953] This service provides mailers with address cards sorted into delivery sequence. Three levels of service are offered: (1) basic carrier route walk sequencing of cards, including the removal of cards with undeliverable or incorrect addresses, (2) the service described in (1) plus the insertion of blank cards indicating missing addresses, or (3) the service described in (1) plus the insertion of completed cards for the omitted addresses. New address cards are provided free of charge for rural route delivery addresses that have been converted to city delivery route addresses. There is no charge for the delivery sequence sort, or for inserting blank cards. There is a fee applied to every card removed with an incorrect or undeliverable address, and to new completed cards inserted into the delivery sequence. USPS-T-39 at 31-34.

[5954] The Postal Service uses the correction of mailing lists projected test year costs, including contingency, of 23.2 cents as a proxy for the cost of arrangement of address cards in carrier delivery sequence. USPS-T-29 at 25-26. The Service proposes increasing the 20-cent fee for this service to 25 cents. The Commission recommends the Postal Service's proposed 25-cent fee. At these rates, test year costs will equal 23 cents, producing a 109 percent cost coverage.

[5955] The Postal Service proposes a major restructuring of post office box fees. Postal Service witnesses Mayo (USPS-T-39) and Kaneer (USPS-T-40), using location cost data developed by Postal Service witness Yezer (USPS-T-31), would restructure the current five post office box fee groups into seven groups primarily differentiated by location costs.¹ Rate design witness Kaneer proposes the following fee classifications:

[5956] This new grouping is the result of a methodological change that discards the historical approach of basing post office box fees on the type of carrier delivery available at each post office. In Docket No. R90-1 the Postal Service began to align fees with costs more closely by proposing two new fee groups for higher cost locations, recognizing that the market costs of space are not the same in all city carrier offices throughout the country. Rather, these costs vary widely by location. In Docket Nos. R97-1 and MC96-3, OCA proposed to group post office boxes by cost ascertainment group (CAG) as a means to more closely align fees with costs. PRC Op. R97-1, para. 5906. The Commission encouraged these efforts. Id. at para. 5913. In response, the Postal Service has developed a comprehensive source of location-cost information for future use as the basis for optimal fee group design. The Commission commends both the OCA, for providing the impetus towards a more cost based fee structure, and the Postal Service for its efforts to more accurately measure the cost differences between facilities and to design fees which reflect those differences.

[5957] The linchpin of the restructuring is Yezer's development of location-based costs by five-digit ZIP Code, which are then applied by Kaneer. Kaneer contends that the current fee groups do not necessarily correspond with actual space costs. USPS-T-40 at 5. There likely are thousands of ZIP Codes in the low-rate Groups C and D that have location costs as high as ZIP Codes in high-rate Groups A and B. Because space costs are about 45 percent of total post office box service costs, there is a mismatch between these costs and the current fee groups. Undesirable consequences of cost and fee misalignment can include higher fees in low rent urban areas than in high rent "non-city" areas.

[5958] Kaneer explains that cost homogeneous groups can be defined based on Space Provision costs, since Space Support and All Other, the remaining two cost categories, do not vary by location. Space Provision costs are related to the cost per square foot and the space occupied by post office boxes at each facility. Space Support accounts for costs that arise from custodial and maintenance services, fuel and utilities, and custodial/building supplies and services. All Other accounts for costs arising from Postmasters, Supervisors & Technicians, Clerks, Carriers, Motor Vehicle Services, and Other Supplies and Service.

[5959] Costs are allocated in the new methodology just as they have been in the past, except that the costs of space are now distributed based on Yezer's analysis of location space costs. However, if locations were strictly assigned to these seven fee groups based solely on cost, fee shock might result. Therefore, Kaneer sets forth a methodology to temper the shock, which considers current fee levels as well as space costs. Id. at 13.

[5960] In developing the new classification, the Postal Service relied in part on information gathered after implementation of the R97-1 rate changes. The Postal Service reassigned 21 "transition sites" having very high or low location costs and post office box utilization rates to different fee groups (increasing fees where costs and utilization were exceptionally high and decreasing fees where costs and utilization were exceptionally low). Changing fees in 21 sites permitted the Postal Service to observe customer reaction and gain experience in the dynamics of regrouping boxes.

[5961] Kaneer says several lessons were learned. First, the burdens of administering post office box fees by facility have convinced the Service not to define the new fee groups in that way. Rather, the use of five-digit ZIP Codes holds the best potential for customer understanding and administrative convenience. Second, to keep the new classification schedule simple, the Service decided not to propose capacity use as a fee factor at this time.

[5962] In order to gauge the effect of the proposed fee changes on the number of post office boxes in use and the subsequent effect on post office box revenues, the Service also estimated the price elasticity of demand. Id. at 15-17. This was estimated from the 1998 and 1999 Post Office Box (POB) Surveys, one conducted just before the implementation of the R97-1 fees, the second re-surveying the responding locations during July 1999. In both surveys, counts of boxes in use were taken, thus measuring customer behavior in response to the price increase. Boxes were categorized into two groups, Size 1 (used primarily by non-business users) and All Other Sizes, to control for differences in price sensitivity for these two customer groups. The study obtained elasticity estimates of -0.229 for size one boxes and -0.306 for size two-five boxes.

[5963] Kaneer calculates test year before rates post office box costs of $585 million. He apportions these costs to the nearly 18 million post office boxes estimated to be in use during the test year to derive their unit costs (i.e., average cost per box for each fee group and box size). These unit costs serve as the basis for setting box fees.

[5964] Kaneer states that the general goal of achieving true cost and fee alignment while carefully mitigating fee shock for current box customers is consistent with fairness and equity, and the other statutory classification criteria. A fundamental principle of fairness, pricing in accord with cost causation, is better accommodated by the new rate design system, according to Kaneer. The proposed classification groups increase the desirability of post office box service by apportioning post office box costs to each group and box size in relation to the cost of the resources employed. Many box customers in low cost locations will see fee decreases from the new classification, while people who prefer box service in costly locations can better be accommodated because fees that better reflect costs encourage the addition of new boxes in those areas. Kaneer says the new methodology is desirable for the Postal Service and users, as it will provide accurate price signals to service providers and consumers, allowing the forces of supply and demand to operate. In terms of societal benefit, cost-based pricing encourages entry by alternative service providers only where warranted and thus conserves society's resources.

[5965] Yezer estimates the rent per square foot at Postal Service facilities, particularly those that provide post office box service. He uses data on leased properties such as annual rent paid, square feet of space rented, date of endorsement and term of the lease, provisions for payments of utility and maintenance costs, physical characteristics of the space, and the physical location of the property. According to Yezer, rents are based on all these factors, and vary with market conditions at the time that the lease is signed. USPS-T-31 at 3. He also projects rents per square foot forward based on recent market trends.

[5966] The variables used in his statistical analysis are all taken from Postal Service data. In brief, the dependent variable (R/SQFTj) in the statistical analysis is the quotient of total annual rent for the entire facility divided by size of the rented space in square feet, for facility j. A vector of seven dummy variables reflects responsibility for maintenance and utilities. A vector of eight dummy variables indicates the facility setting, e.g., business park, office building, etc. He notes that setting should have an effect on rent per square foot, e.g., facilities within malls are generally higher cost. Variables also indicate the time at which the current lease was endorsed and lease length. Dummy variables indicate particular physical features of the facility such as branch offices or presence of a loading dock. The natural logarithm of the square feet of interior space is also included. A vector of variables measures the physical location of the facility. For example, for facilities located within the 65 largest MSAs, location is based on the distance between the facility and the center of the central business district (CBD) as well as distance north-south or east-west from the CBD. The expectation is that rents should decline with distance from the CBD, although this effect may not be significant in cities with multiple centers. The general form of the equation used to estimate rent per square foot is: R/SQFTj = a0 + Sa=1 aiMij + Sb=1 bmSmj + QTTj + QT95T95j + QDTDTj + QTLTLj + QDTLDTLj + QBBj + QDDj + QIDIDj + QLILIj + gNoPNoPj + gPPj + Sk=1lkLkj + ej. USPS-T-31 at 3-9.

[5967] Yezer's data set included information extracted from 27,407 leases and 11,608 owned Postal Service properties. Yezer also explained how he handled incomplete documents (e.g., 434 documents were so incomplete that their results were not used). Id. at 10-11.

[5968] Mayo proposes new post office box fees that represent both increases and decreases when compared to the equivalent current fees. Under Postal Service methodology, the cost coverage for post office boxes (including caller service and reserve number) would be 138 percent. The range of the post office box fee changes in the individual fee cells is -25 percent to 73 percent. The total proposed percentage increase is about nine percent. See USPST39 at 102.

[5969] Mayo discusses the pricing criteria for post office boxes. She states that provision of boxes offers a high value of service, citing elements such as privacy, convenience, protection, and, in some cases, a prestigious address. The proposed fees cover the cost of the service and contribute beneficially to other costs. As to effects on customers, she states that the cases having the highest percentage increases represent a little over one-half of one percent of all boxes. Although 33 percent of all boxes are proposed to increase 31 to 43 percent, 35 percent of all boxes (including Group E boxes at no fee) are proposed to decrease or have no change. She argues that while some increases are not small, they do not represent a substantial outlay for most users and should not have a substantial impact on them. Further, the alternative of free carrier delivery is an option, as is using a commercial mail receiving agency.

[5970] Services Related to Post Office Boxes. In addition, Mayo adds three classification proposals for services related to post office boxes: (1) a new classification for a fee to provide more than two keys for a box, or to replace a key due to loss, damage or breakage, (2) a new classification for a fee for a customer initiated post office box lock change, and (3) a classification change to eliminate the DMCS section concerning transfer of street-addressed mail to a post office box.

[5971] The proposed classifications for post office box keys and customer initiated post office box lock changes would have fees of $4.00 and $10.00, respectively. The implicit cost coverage for additional or replacement post office box keys under Postal Service methodology is 142 percent, and for a customer initiated post office box lock change it is 143 percent. Mayo states key replacement and lock change fees will apply to highly valued services. The two fees cover the cost of the services and make a reasonable contribution to other costs. The proposed cost coverages are reasonable when one considers these services used to be provided free-of-charge. The effects of the proposed fees should be negligible because most customers will probably never be in the situation where they would have to pay the fees. The fees are simple and fair (as the costs incurred are recovered from those who caused them).

[5972] Mayo also proposes a fee increase for caller service and a fee decrease for reserve numbers. Caller service is a premium service that allows business customers to pick up their box mail at a post office call window or loading dock during the time the office is open. Reserve number is a service that allows a company to reserve a box number for future caller service use. Businesses could find this useful if they are planning a promotion, campaign or advertisement and would like to use a number that would correspond. The caller service fee is proposed to increase by 36 percent to $375 for a six-month period. The implicit cost coverage is 123 percent under Postal Service methodology. Witness Campbell provides the test year cost estimates for caller service. USPS-T-29 at 22 et seq. The estimates are derived from a 1999 Caller Service Study. The study included a review of the total storage space required and calculation of annual window accounting cost. Mayo testifies that caller service has a high value to users. Caller service customers are able to pick up their mail early in the day to process orders and financial transactions and it provides them a means to receive post office box type service when their volumes are too large or post office boxes are not available. Although 123 percent is not a high cost coverage for this type of service, the proposed fee increase was limited to 36 percent to reduce the adverse impact on caller service customers. Even so, caller service customers have private sector alternatives. The proposed fee is simple-it is uniform nationwide.

[5973] The reserve number fee would decrease by 17 percent to $30 per year. The proposed fee for reserve number was designed by applying a markup over the per-piece cost of $16.98, with a ten-dollar rounding constraint. Although the proposal represents a fee decrease, the fee revenue will cover the cost of the service and contribute substantially to other costs.

[5974] Finally, Mayo proposes to eliminate DMCS section 921.222, which provides a limited right for box customers to redirect delivery of mail from some other address to a box. Redirection of mail to a customer's post office box often depends upon the memory of individuals, or recognizing the significance of a handwritten note. Misdeliveries thus may occur. Section 921.222 can also conflict with current policy which calls for delivery of mail containing both street and box addresses to the address that appears directly above the city/state line. However, customers will be able to have their mail forwarded from one address to another, including a post office box, based on current forwarding procedures.

[5975] Intervenor Popkin on brief opposes the Postal Service's proposal to alter the method for determining post office box fees.² He complains that a post office location in New Jersey has been assigned to the highest cost category, stating that the said locality is located in a 40-year old strip mall. He states the ZIP Code assignment of the post office is related to mail processing needs and not to the cost of the facility. He argues that the library reference associated with Yezer's study was not filed until March 24, 2000, after the period for discovery expired, limiting his time for examination. He questions the validity of data, such as facilities showing a negative Erent³ value and facilities that use a generic rent value. Erents for a California location are characterized as low while his knowledge of the area tells him real estate values are high. He is concerned that the average cost per square foot for all facilities in a certain three-digit ZIP Code area is higher than for Manhattan facilities. Popkin questions how 160 facilities could have no boxes in use. He states that offices listed in the database show an apparently incorrect number of boxes installed, and that data is not shown for all existing facilities. He concludes that the data are so unreliable that using the data would be inappropriate. Popkin Brief at 4.

[5976] The Service replied to these issues in interrogatory responses (also filed under seal). The Service asserts it is not surprising that there are some locations without post office boxes being in use, given a large number of facilities that have only a few boxes being rented. Further, the 160 locations represent less than one-sixth of one percent of all ZIP Codes. It states that as to errors that may appear in grouping certain post office locales within specified groups, it will be updating the data (which it would have had to do anyway, given that post office changes occur continually). It argues that the proposed reclassification should be seen as an ongoing process and acknowledges that there may be errors in the data, which is to be expected given the large data set.

[5977] As to some negative Erent results, Yezer states that the Erents reported are uncensored estimates from the estimating equations reported in his supplementary testimony, but that the actual rents per square foot as reported are not negative. He states that uncensored estimates may produce negative Erents, reflecting the fact that expected rent per square foot is close to zero, perhaps less than $5.00. Yezer notes that the actual lease rental figures per square foot in Manhattan are low also. As to "generic" costs per square foot being used, he acknowledges that he did not have detailed characteristics for each facility within a specified area and therefore Erents were computed based on location alone. Regarding the complaint that a specific New Jersey location has a high Erent, Yezer states that small facilities tend to have higher rents per square foot, and that the computed Erent for this location is below the non-standardized rent per square foot actually being paid. As to Popkin's assertion that Erents for a California location are low in comparison to purported property values, Yezer states that the data he used were correct.

[5978] Commission Analysis. In R97-1, the Commission urged the Postal Service to reexamine this issue and offer a more cost-based approach. The proposal in this docket is a large step in the right direction. The Commission adopts the Service's analysis of the statutory pricing and classification criteria for post offices boxes and the related services.

[5979] A fee structure that reflects a closer alignment with costs is fairer to the public as a whole. Although some rate cells (for certain size boxes in certain fee groups) experience large percentage increases, the resulting cost coverages for those rate cells continue to be much lower than cost coverages for other rate cells. Conversely, cost coverages for rate cells receiving fee reductions are in some cases still relatively high. For example, while Group D6 experiences increases that range from 14 percent to 73 percent under PRC methodology, cost coverages will still be in the 66 percent to 87 percent ranges. Conversely, Group C5 fees will be reduced from 12 percent to 25 percent under PRC methodology. As discussed, the Service is proposing that the post office box fees in this docket for individual cells represent a transition to a time when all cells more directly align costs and fees. The gradual adjustment of box fees to avoid rate shock is appropriate.

[5980] The Service's responses to Popkin's arguments are reasonable. It can be expected that in a large data set some anomalies will be present. The overall methodology used by Yezer appears appropriate. As to questions about the validity of a small portion of the underlying data, it is sufficient that the Service relied on the latest internal censuses. Further, the Service indicates that it will be updating data over time as part of its ongoing reappraisal in this area. As to the location costs for a specified New Jersey location, official government published records show this to be an area with an extremely high per capita income.⁴ Consequently, it is not surprising that property values are high also. Popkin's assertions regarding California locations are unsupported by factual analysis and cannot be given much weight.

[5981] Popkin's complaint on brief about having had little time to analyze Yezer's library reference, filed March 24, 2000, is tardy. Popkin had until May 22, 2000 to file a case-in-chief in rebuttal to the Service, yet he failed to avail himself of this opportunity, or tp ask for additional time to prepare evidence. Initial briefs were not due until September 13, 2000, which means that he had about 5 months to analyze the library reference for his brief. Further, he conducted a substantial amount of discovery on the library reference (filed under seal), e.g., DBP/USPS-144, 145, 146, 148, 155, 156, 206, 207, 208, 209 (with 24 sub-parts), and 216, and responses thereto. The Commission finds that no adverse due process consequences occurred.

[5982] The Commission concurs with the Postal Service's analyses of the statutory pricing and classification criteria as to post office boxes and related services, and recommends the rates as proposed by the Postal Service. The cost coverage for this category is a moderate 139 percent overall, under Commission cost methodology. Two points merit emphasis. First, the Service's post office box fee proposal, in better aligning fees with costs, is consistent with the statutory mandates to establish and maintain fair and equitable schedules and classifications. Second, the mitigation of rate changes for individual cells reflects the statutory command that the Commission consider the effect of rate increases on the general public. The Commission also agrees with the Postal Service's analysis regarding its proposal to eliminate DMCS section 921.222. This change seems desirable from the point of view of the public and the Service itself.

[5983] In closing, the record discloses there are post offices having no or low utilization of existing boxes. The Postal Service should explore whether discounts from box fees could be developed for post offices having very low box utilization rates.

[5984] Business Reply Mail (BRM) is a special service for First-Class Mail and Priority Mail. BRM allows the distribution of reply mail envelopes or cards indicating that no postage is necessary for mailing back to the distributor. On receipt of reply mail, the addressee must pay postage plus the applicable BRM fee.

[5985] There are currently three fee categories associated with BRM. USPS-T-29 at 67. Qualified Business Reply Mail (QBRM) pieces are those cards and one- and twoounce envelopes that are automation compatible, have both a Facing Identification Mark (FIM) C and a unique ZIP+4 barcode, and have qualified for Business Reply Mail Accounting System (BRMAS) processing. QBRM mailers pay a per-piece fee of five cents in addition to postage. They also must maintain an advance deposit account, with a balance sufficient to cover the projected postage due and per-piece fees for a specified future period, and pay an annual advance deposit account fee. A rate discount from the First-Class mail postage is given to QBRM pieces.

[5986] The second category encompasses non-QBRM advance deposit BRM pieces. They are not required to qualify for BRMAS processing, although these pieces are often prebarcoded. Like QBRM, per-piece fees and postage due are deducted from an advance deposit account. Non-QBRM advance deposit BRM mailers pay a per-piece fee of eight cents in addition to postage.

[5987] Advance deposit weight-averaged nonletter-size BRM qualifies for a one-cent per-piece fee. PRC Op. MC99-2. Mailers of such pieces also pay a $600 monthly fee and an annual permit fee. BRM flats and parcels, as well as BRM letters weighing more than two ounces, are eligible for the classification. Weight averaging entails weighing and rating eligible BRM in bulk at postal facilities, and eliminates the need for manual counting and rating. In MC99-2, the Commission found that "weight averaging, relative to manual accounting, substantially reduces postal costs . . .." Id. at 2.

[5988] The third category is non-advance deposit BRM pieces, which may or may not be automation compatible or barcoded. Non-advance deposit BRM mailers do not pay the postage due and per-piece fees through an advance deposit account. Instead, these pieces are delivered to the BRM originating mailer upon payment of postage and fees due. Mailers receiving low volumes of BRM generally use non-advance deposit BRM. Non-advance deposit BRM mailers currently pay a 30-cent per-piece fee in addition to postage.

[5989] In addition to the applicable postage and BRM per piece fees, BRM mailers pay an annual permit fee, and those mailers with advanced deposit accounts also pay an annual accounting fee. The advanced deposit account allows mailers to have postage and fees automatically deducted from their accounts as mail pieces are delivered.

[5990] Postal Service witness Campbell describes BRM mail flows that affect BRM costs. USPS-T-29 at 7 et seq. To determine the per-piece counting, rating, and billing costs associated with QBRM and BRM, one must focus on destinating facility operations. BRM letters and cards are generally separated out in the Incoming Primary operation and sent to either the BRMAS operation or to a manual sortation operation (often the Postage Due Unit or Box Section). This differs from other non-presort First-Class Mail letters and cards which, after sortation in the Incoming Primary operation, are processed in an Incoming Secondary operation (either automated or manual), and are then sorted to address either in a Delivery Point Sequence (DPS) operation or in a manual operation (i.e., cased by the carrier).

[5991] This discussion will focus on QBRM processing, for it is that rate category whose fees are in dispute. Campbell testifies that QBRM goes through the Incoming Primary operation, and then can be sorted to permit number (corresponding to a unique ZIP + 4 Code) in a BRMAS or Barcode Sorter (BCS) operation. Id. at 8. Because the ZIP + 4 Code is unique to a BRM customer, this sort is equivalent to that obtained in a DPS operation. Thus, these pieces avoid the Incoming Secondary distribution that other First-Class Mail pieces get. BRMAS operations vary according to facility, e.g., employing either Delivery Barcode Sorters or Mail Processing Barcode Sorters. See id. at 9, for a fuller description. For those pieces finalized by BRMAS, the BRMAS program also performs counting and rating functions, and can provide a report (i.e., a bill) for the BRM originating mailers of postage due. BRMAS does not deduct the postage due from the advance deposit account.

[5992] Campbell testifies that even at facilities that have the BRMAS operation, "not all QBRM gets finalized to permit number in the BRMAS operation." Ibid. For example, there may be operational limitations such as the number of bins available for sortation, or diversion to other mail streams (e.g., mixing with other First-Class Mail that got distributed in a DPS operation). These residual pieces are usually sorted, counted and rated manually in the Postage Due Unit. Even when all QBRM pieces for a mailer are finalized in BRMAS, verification and accounting activities are performed in the Postage Due Unit.

[5993] According to Campbell, at facilities without BRMAS operations, QBRM is counted, rated and billed using a variety of methods, both manual and automated. Id. at 10. Manual counting of mail pieces, end-of-run (EOR) report counts and weight averaging techniques are all employed. Rating and billing functions are typically performed manually or through the PERMIT system or other software. Ibid. As shall be seen, the parties dispute the frequency with which particular methods for processing QBRM are used, e.g., BRMAS versus manual counting. They also dispute the productivity of each method.

[5994] Postal Service witness Mayo, USPS-T-39, proposes two interrelated classification changes and several fee changes for BRM. The two classification changes would split QBRM into two fee categories. The first category, useful for high-volume users, would have a low per-piece fee and a quarterly fixed fee. The second QBRM classification, useful for lower-volume users, would only have a higher per-piece fee.

[5995] Currently, QBRM per-piece fees cover the cost of what may be described broadly as performing QBRM piece-counting, as well as the rating and billing functions. As Campbell notes, though, some costs incurred in the process are fixed in nature. "[T]he costs of rating, preparing meter readings, and completing postage due forms are incurred each time a QBRM account requires a transaction, regardless of the QBRM volume or the method used (manual or automated). For example, if a QBRM account receives 1,000 QBRM pieces, the time required to generate a bill is the same as if the account receives 10,000 pieces. Similarly, rating 1,000 QBRM pieces (i.e., calculating postage due given a piece-count) requires the same amount of time as rating 10,000 QBRM pieces." USPS-T-29 at 14. As a result, productivities for rating and billing, previously incorporated into the per-piece costs, have been isolated and incorporated into a monthly fixed cost for potential high-volume users. Arguably, smaller volume users will find the per-piece fee simple and financially advantageous, while larger volume users will find their total QBRM fees to be lower if they choose the new two-part structure. According to Mayo, at the proposed fees, the volume at which it will pay to switch to the two-part fee structure is approximately 113,000 pieces per year.

[5996] Mayo thus proposes that the per-piece fee for the existing QBRM category, currently 5 cents, increase by 20 percent to 6 cents, producing a 122 percent implicit cost coverage. The proposed QBRM per-piece fee for those mailers paying a quarterly fee would be 3 cents, yielding a 146 percent implicit cost coverage. The quarterly billing fee for the new category would be $850 (resulting in annual payments of $3,400) resulting in a 119 percent implicit cost coverage.

[5997] Both categories would qualify for the proposed QRBM postage discount of 3 cents. Postal Service witness Campbell finds the test year mail processing cost avoidance of a QBRM piece to be 3.38 cents. Id. at 41. He says that improvements in RBCS character recognition have lowered the cost associated with handwritten single-piece processing, and thus have reduced QBRM's cost avoidance. Postal Service witness Fronk (USPS-T-33) uses this cost avoidance to propose retaining the current 3.0 cents QBRM discount off the single-piece First-Class letter rate.

[5998] The current eight cents per-piece fee for regular BRM with an advance deposit account would increase by 25 percent to 10 cents, providing a 132 percent implicit cost coverage. The current 30 cents per-piece fee for non-advance deposit account BRM would increase by 17 percent to 35 cents, providing a 128 percent implicit cost coverage. The weight-averaging nonletter-size BRM monthly fee of $600 and the per-piece one-cent fee would remain unchanged, resulting in 117 and 173 percent implicit cost coverages. The annual advance deposit accounting fee for BRM would rise from $300 to $375, a 25 percent increase, with an implicit cost coverage of 116 percent. The annual permit fee for BRM would increase from $100 to $125, a 25 percent increase, with an implicit cost coverage is 117 percent. The overall cost coverage for BRM would be 123 percent.

[5999] Mayo argues that BRM (including QBRM) is a high-value special service. Alternatives to BRM include toll-free phone numbers and company-supplied envelopes with pre-affixed postage. USPS-T-39 at 27. With BRM, the distributor only incurs the cost of postage for returned mailpieces, which aids organizations unsure of a mailing's potential response rate. BRM also makes a good impression on customers because it demonstrates a company is willing to pay the postage. Mayo says the proposed BRM fees individually and as a whole cover their costs and moderately contribute to institutional costs. She asserts that at the highest increases of 25 percent, there would be no adverse impact on users, especially QBRM users that also receive a postage discount. However, Mayo notes that over the past 5 years BRM volume has decreased 39 percent and BRM revenue has decreased 11 percent. From 1997 to 1998, BRM revenue decreased 9 percent. Mayo states that the proposed fees were designed to attain an overall moderate cost coverage.

[6000] Intervenors' Arguments. Intervenors KeySpan Energy and the Long Island Power Authority (KeySpan) agree that QBRM fees should be deaveraged by establishing separate fee structures for high and low volume recipients. However, KeySpan believes the Service's plan is flawed because it effectively assumes that volume has no effect on the counting methods used or the costs of counting.

[6001] KeySpan recommends per-piece fees of 0.5 cents for high-volume users, based on its calculation of a unit cost of 0.17 cents, and 4.5 cents for low-volume users, based on a unit cost of 3.43 cents. It also recommends that the First-Class rate for QBRM reply mail be raised by only 0.5 cents (to 30.5 cents based on a 34.0 cent single-piece rate), stating that the cost avoidance is 5.242 cents per piece. Tr. 44/19107.

[6002] KeySpan proposes a higher accounting fee ($12,000 versus $3,400 annually) to insure the Service can employ efficient counting methods. The higher annual fee will mean a higher breakeven point, so only mailers with very high volumes that clearly justify using bulk counting procedures (like BRMAS) will use the new rate category. It calculates that the breakeven volume for high volume recipients would be 300,000 pieces per year; there are 288 recipients who have received more than this amount of mail in the past 12 months or in FY1999. However, KeySpan witness Bentley states that the monthly fee "is also much greater than any markup that might be reasonably justified from application of the statutory criteria of the Act." Tr. 29/13992. KeySpan also says that if the Commission is concerned about access to the high volume plan, "there is plenty of room to lower that fee." KeySpan Brief at 23.

[6003] KeySpan's arguments concerning the proper per-piece fees focus on the allocation of projected volumes among the different methods used to process pieces (BRMAS, end-of-run or EOR, special counting machines or SCM, weight averaging, and manual) and how efficient each method is. Bentley determined the percentages that would be counted by each counting technique using volume data from the Corporate Business Customer Information System (CBCIS) coupled with his analysis of a survey of high volume offices done by Campbell (Campbell Survey), and various estimation techniques. Tr. 29/14053-54.

[6004] The Postal Service utilizes costs developed from an allocation of volumes based on an update of the BRM Practices Study. The BRM Practices Study was originally prepared for use in R97-1.⁵ Campbell claimed to update results of the BRM Practices Study, based on discussions with on-site personnel. Tr. 14/6026, 6029-30. He makes no distinction between high-volume and low-volume users. The comparison of Campbell's and Bentley's findings is summarized in Table 5-21 (see Tr. 29/13998):

Table 5-21
QBRM Counting Methods
		Percent of QBRM Counted By:
QBRM Category	Data Source	BRMAS	ECR	SCM	Weight	Manual	Total

High Volume QBRM	BRM Practices Study	14	19	10	9	47	100
	CBCIS Data System	52	28	1	8	11	100
Low Volume QBRM	BRM Practices Study	14	19	10	9	47	100
	CBCIS Data System	21	23	1	7	48	100
All QBRM	BRM Practices Study	14	19	10	9	47	100
	CBCIS Data System	44	27	1	8	20	100

[6005] For the cost of counting low-volume QBRM, Bentley uses the same productivities for hand counting and weight conversion techniques that he developed for high-volume QBRM. Tr. 29/14027, KeySpan Reply Brief at 14. He assumes that above 400 pieces per day, it would not make economic sense to hand count. He also calculates that there is a close match among the post offices that receive substantial volumes of both high- and low-volume material and that it is logical to assume that such facilities will use the same counting methods for relatively high-volume accounts. He also makes the conservative assumption that for the 46 million QBRM pieces received by low-volume accounts in volumes less than 100,000 per year, all such pieces would be hand counted, although he contends such mail may be counted using more efficient methods.

[6006] KeySpan notes by way of comparison that the recently approved rate for nonletter-size BRM, primarily consisting of small bulky packages that contain film canisters, has a one-cent per piece counting fee, based on a unit cost of 0.57 cents. KeySpan Brief at 6. "[I]mplicit in this comparison is the absurd notion that it costs three and one-half times as much to process uniform, compact QBRM letters and cards as it does to process non-uniform, bulky parcels." Ibid. It notes Campbell's agreement that the 7,272 pieces per hour productivity factor for weighing bulky non-letter size parcels derived in Docket No. MC99-2, might be a reasonable, even "conservative" productivity factor for counting uniform QBRM pieces by weight averaging. Tr. 14/617475.

[6007] In support of the contention that the Service's overall QBRM costing approach is flawed, KeySpan states that Campbell admits he wanted to study whether it costs less to count QBRM received in high volumes than it costs to count QBRM received in low volumes, and admits "that data obtained from such a study could improve the cost estimates presented in this rate case filing," (Tr. 14/6015) but claims that "time constraints" precluded him from conducting such a study. Tr. 14/6014-17. According to KeySpan, Campbell also acknowledged that, although he traveled to three facilities to observe QBRM processing, he "[did] not have specific recollection of discussions with Postal Service personnel regarding whether the QBRM reply letters they were counting were addressed to high volume recipients or addressed to low volume recipients" and "[did] not recall specific volumes or percentages of the `high volume' pieces observed in relation to the QBRM recipient's total pieces received on that day." Tr. 14/5978, 598081, 5982.

[6008] KeySpan argues that the Service's statements to the Commission in Docket No. MC99-2 are particularly important. KeySpan Brief at 15. In responding to an information request concerning the status of any work or planning related to cost effective methods for counting, rating and billing letter or card size BRM, the Service said:⁶

In response to the Decisions of the Governors in Docket No. R97-1 (June 29, 1998), management has established two objectives. The first is to focus on improved utilization of machine- or automation-based QBRM accounting alternatives to the manual accounting method. . . .

Given the relatively high degree of automation-compatibility of BRM letters and cards, the Postal Service is committed to more fully utilizing its capacity to perform automated or machine-based accounting, where appropriate. Particularly with higher-volume QBRM letter and card recipients, as each separate recipient's mail is isolated, the opportunity exists - either during mail processing or in the accounting function - to obtain a machine count of such mail, to a greater extent than is currently being done.

[6009] As to efficiency of the counting methods, instead of relying on Campbell's 951 piece-per-hour (PPH) manual productivity figure, Bentley performed two demonstration studies that led him to conclude the productivity factor for counting QBRM manually was 2,746 PPH and by weight averaging was 68,901 PPH. Tr. 29/14032-35. Bentley says these are conservative estimates, assuming a low level of clerical productivity, i.e., he multiplied the initial productivity results by 0.6. Id. at 14035.

[6010] Bentley states that the 951 PPH figure, which is based on a 10-year old study, is flawed. Tr. 29/14050-51. Furthermore, he contends that increased automation likely means less sorting needs to be done in the postage due unit. Id. at 14050. Bentley argues that because Campbell had no data on the productivities used for special counting machines or weighing techniques, he erroneously assumed the figure for manual productivity applied to those techniques. Tr. 14/5957.

[6011] Postal Service Counter Arguments. In rebuttal to Bentley, Campbell states that Bentley had no basis for assuming that the counting methods used for accounts receiving between one and ten million pieces annually would apply to accounts receiving 250,000 to one million pieces annually. Tr. 39/17506. As to counting methods for low-volume accounts (less than 300,000 pieces annually), Bentley's analysis is said to be arbitrary, especially the assumption that the allocations applied to higher volumes would apply for accounts receiving 100,000 pieces annually (or about 400 pieces per day). Tr. 39/17506-07. Campbell also contends that in contrast to Bentley's analysis, the Practices Study underlying his analysis represents a comprehensive sampling effort. Id. at 17509. Campbell states: "Although it might be `logical' to assume that more efficient counting methods are used to a higher degree with larger accounts, the only information which definitively shows what methods are applied to particular accounts is reflected in response to KE/USPS-T29-49 (Tr. 14/6025, 6026, 6030)."⁷ Id. at 17509-10.

[6012] On brief, the Service maintains there are defects in Bentley's analysis of per-piece counting costs as a result of inflated manual counting productivities and manipulation of data. Postal Service Brief at VIII-23. It opines that Bentley does not understand the sorting relevant to QBRM processing, and that the 1989 study by Postal Service witness Pham on the issue is more reliable than Bentley's derived productivity. It contends that Bentley's counting method percentages are skewed in favor of low-cost, automated accounting methods for both high and low-volume accounts. Id. at VIII-24-25. He is said to have erroneously included 56 million QBRM pieces in his high-volume analysis, which results in his underestimating the unit cost to count QBRM received in high volumes. Tr. 39/17503-04.⁸ The Service cautions against use of extra-record material concerning the inclusion of data from a customer with multiple accounts. Postal Service Reply Brief at VII-3-4.

[6013] It takes issue with KeySpan's comparison to the 0.57-cent average unit cost estimate for nonletter-size BRM, saying this figure solely reflects weight averaging alone. Id. at VII-4. It argues that there are many reasons why the highly efficient BRMAS processing method is not used more. "Foremost among these is the fact that the BRMAS accounting window and the Delivery Point Sequencing window are open concurrently and DPS has preferential access to automated equipment, often shifting BRMAS-eligible QBRM to less efficient counting methods." Id. at VII-6. Also, "The Postal Service is utilizing its automated equipment to do more automated mail processing, cutting down on opportunities to run BRMAS." Id. at VII-6-7. It notes that CBCIS contains no actual counting methods data, just BRM customer volumes. Id. at VII-9. It continues to urge that Bentley's analysis of counting methods employed is an unreliable estimate. Id. at VII-10; Tr. 39/17502 et seq.

[6014] It concedes, however, that Campbell's "2.00-cent average unit counting cost estimate for QBRM is too high." Id. at VII-5. The Service concludes that another comprehensive BRM Practices Study is needed before it can take de-averaging to the next level. Postal Service Brief at VIII-25. If the Commission feels it cannot rely on Campbell's analysis, the Service encourages the Commission to "consider a reasonable adjustment to Campbell's methodology to establish an appropriate `middle ground' estimate upon which to base its QBRM high-volume per-piece fee recommendations." Id. at 26.

[6015] The Service contends that Mayo's quarterly fee is cost-based and opens up lower fees to more mailers than Bentley's proposal, estimating 1,300 mailers meet the breakeven test for its plan as opposed to 300 under Bentley's plan. Ibid. Mayo indicates that under her proposed quarterly fee, mailers will be allowed to opt in or out by quarter depending on their mail volumes. Tr. 39/17652-53.

[6016] Finally, it also criticizes the weight-averaging and manual-counting productivities derived from Bentley's demonstration studies. Postal Service Brief at VIII-23-24. It states that the studies underlying its productivity figures reflect real world operations, and incorporate such factors as set-up time and clerk fatigue. Id. at 24, n. 20.

[6017] Commission Analysis. The Commission recommends the proposed BRM fees and classifications advanced by Mayo, except for the fees and rates for QBRM discussed below. Mayo's discussion of how the pricing and classification criteria apply to this service is sound, and is adopted by the Commission.

[6018] As to the QBRM issues, the Commission finds that the cost of the counting function is closely related to the number of pieces counted, while the cost of the remaining functions is driven by the number of bills that must be prepared. Thus, the splitting of QBRM into two fee structures is appropriate. However, the Commission cannot recommend the Service's fee proposals for QBRM.

[6019] Counting Methods. In determining the proper per-piece fees for high- and low-volume mailers, one must evaluate the frequency with which certain methods (e.g., BRMAS, manual, etc.) are used to count the mail pieces. This data can then be projected onto expected QBRM volume patterns under the new rate structure. The extent to which BRMAS is used is important because BRMAS counts as well as sorts in one operation. Tr. 39/17543. For high-volume QBRM, using BRMAS means that most such pieces come to the postage due unit in full trays and require no added sorting. Id. at 17545.

[6020] The Service states "it is apparent" that the BRM Practices Study relied on by Campbell understates the percentage of high-volume QBRM that is counted by BRMAS. Postal Service Brief at VIII-22. It says that "[s]hould the Commission conclude that it cannot rely completely on Campbell's analysis, the Postal Service encourages the Commission to . . . consider a reasonable adjustment to Campbell's methodology to establish an appropriate `middle ground' estimate . . .." Id. at VIII-25-26. It recognizes that "a limitation of the BRM Practices Study is that it was not a census which permits one to determine which accounting methods are employed at every site for every account, large and small." Id. at VIII-21. Campbell in fact agreed that just the volume associated with the highest 77 accounts was two times greater than his predicted BRMAS usage for QBRM overall. Tr. 39/17616.

[6021] A problem with the Service's concession is that it does not explain why, if Campbell's survey disclosed such glaring defects in the Practices Study for analyzing BRMAS usage, it should be relied upon at all. Indeed, at one point Campbell is of the opinion that "there is really no data available currently to deaverage by counting method." Id. at 17548. Further, once one adjusts the BRMAS usage figures shown in the Campbell Survey data, which is necessary given the state of data in the Practices Study, one has to adjust the figures for the four remaining methods. Id. at 17627-28. Campbell gives no guidance on how this should be done. He states: "We've got data on the top 75. We don't know any further down the line, how the - what volumes, what accounts, are going to be counted in what methods." Id. at 17616. [emphasis added] The Service suggests adopting a middle ground, without explaining how or why to get there. In contrast, KeySpan has offered an approach that seems plausible and reasoned.

[6022] The Commission finds that KeySpan's high-volume analysis presents the best available evidence, incomplete as it is. KeySpan uses a combination of volume and customer number data from the Service's CBCIS system, supplemented with the Campbell Survey and other estimating procedures set forth above. Tr. 29/14057 et seq. Bentley's assumptions appear reasonable overall. For example, he removes data from the two largest accounts when analyzing remaining high-volume accounts. He accepts data that show hand counting is taking place in some post offices despite there being high volumes at those offices.⁹

[6023] The updated processing information supplied by Campbell shows results that implausibly seem to favor manual counting, the most inefficient counting method. (Tr. 14/6030) It is easy to believe that high volume offices would use the more efficient counting methods; it strains credulity to think that offices receiving large volumes would hand count most or all of the pieces, as Appendix 2, Tr. 14/6030, indicates to some extent. Indeed, Campbell admits that low volume customer accounts are sometimes processed on automation. Tr. 39/17548.

[6024] Campbell argues that Bentley erroneously included in his cost analysis data concerning a recipient with 2,500 separate accounts. Id. at 17523-24, 17586. Campbell says he does not know "how many of those accounts would be considered high volume and which ones would be considered low volume." Tr. 39/17523, see also Tr. 39/17591. He states that many of these accounts could comprise the universe of recipients qualifying for the Service's high-volume fee. Id. at 17524.

[6025] The Commission finds that Bentley's handling of this data is reasonable.¹⁰ In arriving at this conclusion the Commission did not give any consideration to material not admitted into evidence regarding account consolidation. Id. at 17600 et seq. However, it infers that account realignment of substantial magnitude would take place under the revised fee structure the Commission recommends; the inference that no realignment would take place is patently unreasonable. The Service itself has used assumptions concerning how potential users might switch mailing methods upon introduction of a new classification; see, e.g., the estimates of Postal Service witness Fronk for Prepaid Reply Mail and QBRM in PRC Op. R97-1, paras. 5128-29.

[6026] Productivity in Handling Pieces. KeySpan's unit cost of 0.17 cents for high-volume QBRM depends in part on productivity factors Bentley develops for pieces counted manually, and for weight conversion techniques or special counting machines. Instead of Campbell's 951 piece-per-hour (PPH) manual productivity, derived from work Postal Service witness Pham did for Docket No. R90-1, Bentley performed studies that led him to conclude the productivity factor for counting QBRM manually was 2,746 PPH, and by the weight averaging method 68,901 PPH.

[6027] As to that part of QBRM processed through weight averaging, the Service at first asserts the PPH is 6,390, based on the 1987 Reply Mail Study. But Campbell recognizes non-letter BRM processing (film cans in bags) has been found to be 7,272 piece per hour. He agrees that it would be less costly to apply weight techniques to letter size QBRM than it would be to apply weight techniques to nonletter sized bulky packages, and therefore 7,272 PPH probably is too low.¹¹

[6028] The Commission agrees it is logical to assume that weighing letters is much more efficient than weighing bags containing film, based on the simple physics that uniform, slender shapes are easier to handle than bulkier, non-uniform shapes. Under one processing scenario discussed at the hearings it might take nine times longer to weigh non-letter BRM than letter-sized QBRM. Tr. 39/17584. The Service, having discredited its own initial findings, then in essence suggests a higher figure (7,272 PPH) also may be too low. Upon viewing the video of Bentley's demonstration, the physical actions of the demonstration worker are moderately paced. Furthermore, Bentley's results are formulated by multiplying his original results by 0.6, which should make up for defects in the design of his demonstration. This mathematical adjustment is intended to account for times when clerks are not being productive or are not performing optimally. Campbell agrees that the 0.6 figure is more conservative than the 22 to 23 percent figure representing mail processing overhead costs that is traditionally assumed by the Service. Tr. 39/17569. The Commission therefore adopts Bentley's recommended hand-weighing productivity for use in QBRM costing analysis.

[6029] Similar reasoning leads to adoption of Bentley's manual productivity figure. The Postal Service argues that "[u]nlike witness Bentley's video demonstration, the productivity derived by the Postal Service encompasses numerous tasks and incorporates such factors as set-up time, clerk fatigue, and travel time." Postal Service Brief at VIII-24 fn. 20, citing Tr. 39/17499. However, Bentley's study does not appear unreasonable; the pace of the workers appears moderate. Further, Bentley has substantially pared down his study results (multiplying his result by 0.6) in order to be conservative. In addition, the idea that it would take one hour to count 995 pieces (or, less than 17 pieces per minute) of uniformly-sized letter mail by hand, as the Service suggests, seems unreasonably slow on its face, even with fatigue and set-up times considered.

[6030] Concluding that KeySpan has the better of both the method allocation and productivity arguments, the Commission finds that for high-volume users, it is reasonable to accept Bentley's unit cost calculation. On the basis of these findings, the Commission recommends a one-cent piece rate. This piece rate is the same as that proposed and recommended for weight-averaged non-letter BRM. Though there are differences in how the pieces are handled (all weight averaging versus a variety of methods) the fact remains that processing letter material is apparently at least as efficient as handling bulky film containers overall. The Commission does not recommend the lower 0.5 cent per-piece fee KeySpan suggests in this case due to the imprecision in the costing data and its commitment to ensuring that costs are covered. Hopefully, this precision will improve when actual experience with high-volume QBRM is reviewed.

[6031] For low-volume users, the Commission recommends retaining the current per-piece rate. The Service has failed to justify any per-piece rate increase, given the inadequate state of the method allocation and productivity evidence it relies on. The low PPH figures initially advanced by the Service are flawed, as discussed, and the Campbell Survey brings the entirety of the BRM Practices Study into question. The Commission declines, however, to cut the low-volume rate, as KeySpan suggests, though there is some evidence pointing in that direction. KeySpan's assumptions about how low-volume QBRM is handled seem reasonable. However, they do require additional steps in the estimating process. Following the system wide implementation of best practices methods for handling QBRM, it should be possible to justify a lower, cost-based per-piece fee.

[6032] The Commission declines to follow the KeySpan suggestion, offered on brief, that it recommend an alternative combined low-and-high volume piece rate in case the Governors decide to reject these recommendations. There is ample evidence upon which to recommend a two-part QBRM rate structure, and continuance of the current rate structure would be inconsistent with the statutory guide to recommend rates that are fair and equitable.

[6033] Accounting fee. KeySpan proposes a higher accounting fee for high-volume QBRM to ensure the Service can employ efficient counting methods, which is a commendable goal in view of how processing method choice affects overall costs. USPS claims the breakeven figure for its $3,400 per year fee is 113,000 pieces per year, which it says would cover 1,358 recipients, though Bentley argues the Postal Service misestimated and the real figure is 522 recipients. Tr. 29/14001 fn. 18. KeySpan admits its $12,000 fee is higher than statutory criteria warrant and says there is room to change the fee. The Commission adopts a middle ground approach, and will recommend setting the fee at $1800 quarterly (the equivalent to $600 monthly). This produces a break even figure of 180,000 annually (or 15,000 monthly), and at current usage patterns (see KE-LR-1) one might expect about 431 high-volume recipients to take advantage of the new QBRM fee structure. In addition to encouraging more efficient processing, as suggested by KeySpan, this figure will be consistent with the non-letter size fixed fee (of $600 a month). At the same time, the fee is not so high as to unfairly limit usage to the highest volume accounts. The Service's opt-in opt-out procedure will tend further to encourage the most efficient usage of QBRM as recipients focus their use on high-volume seasons.

[6034] The QBRM Discount. Campbell finds the cost avoidance of a QBRM piece to be 3.38 cents, and rounds the discount to 3.0 cents off the First-Class rate. The Commission calculates this cost avoidance to be 3.36 cents, an insignificant difference. KeySpan suggests a 3.5 cents discount off the First-Class rate, but Bentley's calculation includes, inter alia, avoided window costs. KeySpan Brief at 42. The Commission will not follow this cost avoidance approach for the reasons stated in the discussion of the Pitney Bowes proposal for a metered mail discount.

[6035] Classification and Pricing Criterion. The recommended new fee categories for QBRM will allow fees that are aligned more closely with costs than existing fee, and hterefore more fair and equitable to QBRM mailers. The Commission notes that BRM volume has decreased substantially over time. This may have occurred in part because firms that would otherwise use BRM have been deterred by a non-cost based fee structure for QBRM that made its use uneconomical. Adopting a more cost-based QBRM fee structure may spur volume. Because such mail pieces are often related to bill paying, greater usage may aid the Service in staving off electronic diversion. Improvements in QBRM pricing and costing are therefore quite important, and the new structure is desirable from the point of view of both users and the Postal Service.

[6036] The recommended rates for high volume QBRM should provide a generous contribution to institutional costs. The Postal Service estimates high volume accounting costs of less than $250 per month. USPS-T-29 at 15, 21. The recommended fee of $600 per month will easily recover these costs. Similarly the $0.01 per piece fee substantially exceeds projected costs of $0.0017. This should ameliorate any concerns that Bentley's cost analysis was optimistic in assessing the efficiency of QBRM processing. These fees should encourage more efficient, high volume rating and billing, fully recover QBRM costs, and provide a contribution to institutional costs that reflect the high value of BRM to both senders and recipients.

[6037] When the Governors rejected the Service's own Prepaid Reply Mail category in Docket No. R97-1, they directed the Service to "explore further such matters as the extent to which reply mail volume should influence fees charged to different recipients."¹² Work remains to be done in the QBRM costing area. Further, there are no guidelines to tell field employees how best to process varying QBRM volumes. Processing method decisions appear to be conducted on an ad hoc basis. In the Service response to an information request in Docket No. MC99-2, the Service said it was committed to more fully using its capacity to perform automated or machine based counting, saying that this was particularly true as to QBRM. Docket No. MC99-2, Postal Service response to POIR Request at 3-4, filed June 18, 1999. The Service should expeditiously introduce a plan to guide local facilities' processing decisions.

[6038] Introduced in MC79-4, Merchandise Return allows a permit holder, for a piece fee plus postage, to receive parcels containing merchandise being returned to the permit holder by a mailer without the mailer having to pay the postage. Similar to business reply mail, it can be utilized for parcels mailed at First-Class, Priority and Standard B rates. The permit holder can purchase additional services such as registry, insurance or special handling. USPS-T-39 at 67. The Postal Service has proposed classification changes to allow senders using merchandise return to purchase insurance, to eliminate the per-piece fee, and to establish an annual advance deposit account fee of $375.00. Id. at 66.

[6039] Commission Analysis. The Postal Service bases its proposals on cost data provided by witness Eggleston. Originally, costs for Merchandise Return included weighing and rating as part of the acceptance cost function, and was thought to include an additional sortation. According to the Postal Service, the Merchandise Return cost study update in 1986 changed this view. USPS-T-26 at 42. That cost study had three components: distribution and separation, weighing and rating, and billing and trust fund accounting. It was determined that, for the most part, weighing and rating takes place in postage due units. Thus, weighing and rating should not be reflected in both the per piece fee and the underlying postage fee. Id. at 41. Eggleston isolates the proposed cost reduction by using as a benchmark a parcel which an individual "sends to a business by taking it to the window for weighing and rating." She determines that Merchandise Return parcels incur no additional costs. Id. at 42. To reflect absence of additional costs the Postal Service proposes to eliminate the per-piece fee. Id. at 44. The Postal Service's proposal was unopposed by the parties.

[6040] Recommendation. The Commission finds that the evidence supports the Postal Service proposals. The Commission recommends the elimination of the per-piece fee, the establishment of an annual deposit account fee in the amount of $375, and the authorization for all mailers, not just shippers, to purchase insurance on Merchandise Return parcels.

[6041] On-site meter service is a service whereby a Postal Service employee travels to a customer site or to a meter manufacturer site, and sets a meter, examines a meter, or checks a meter in or out of service. The nature of postage meter service has changed over the past few years due to the recent de-certification of all mechanical meters and proliferation of electronic meters. Over 90 percent of postage meters in use today are remote-set electronic meters that are set by the manufacturer, not the Postal Service. USPS-T-39 at 80-87. The Postal Service proposes to change the name of this service from on-site meter setting to on-site meter service. The new name will reflect the changes in postage meter technology and that the service provides more than meter setting. The Commission recommends the name change to "on-site meter service."

[6042] The Postal Service proposes to restructure the fee classifications for on-site meter service. The current fee structure differentiates between a scheduled visit, $27.50, and an unscheduled or emergency visit, $31.00. The charge for the visit includes resetting or examining one meter. The fee for resetting or examining additional meters is $4.00 each. A separate fee of $8.50 is charged for each meter checked in or out of service.

[6043] Under the proposed fee structure, the Service will charge $31.00 per visit to a customer or manufacturer site. This fee includes employee and travel related expenses in getting to the site, but no longer includes resetting or examining the first meter. A separate fee of $4.00 will be charged for each meter that is reset, examined, or checked in or out of service. For a site with one meter that has an appointment for meter service, the effective fee increase is from $27.50 to $35.00 ($31.00 Meter Service plus $4.00 Meter Reset and/or Examination fee). The Service proposes to reduce the fee for check in or out of service from $8.50 to $4.00. This reflects the lower cost of the service due to the introduction of modern postage meters and the de-certification of the older mechanical type meters. The fee structure changes are summarized in Table 5-22.

Table 5-22
On-Site Meter Setting
	Current	Proposed	Recommended

First Meter By Appointment	$27.50	Eliminated	Eliminated
First Meter Unscheduled Request	$31.00	Eliminated	Eliminated
Additional Meter	$4.00	Eliminated	Eliminated
Meter Service (per Employee)	-	$31.00	$31.00
Meter Reset and/or Examined (per Meter)	-	$4.00	$4.00
Check In/Out of Service (per Meter)	$8.50	$4.00	$4.00
Source: Adapted from USPS-T-39 at 81.

[6044] The Postal Service proposes that the check in/out of service fee not be applied to "Secured Postage" meters. Secured Postage meters are remotely set, contain a postal security device and print information based indicia. Postal Service witness Davis states that these meters do not have significant check in/out costs. USPS-T-30 at 18. The Commission recommends that the check in/out of service fee not be applied to "Secured Postage" meters.

[6045] The proposed fees will result in a cost coverage of 123 for the on-site meter service as a whole. The Postal Service proposals simplify the fee schedule and the proposed fees include an adequate contribution to recovering institutional costs. The Commission recommends the new rates and classification proposed by the Postal Service for the on-site meter service.

[6046] BPRS, introduced in Docket No. MC97-4, is designed to allow a qualifying bulk mailer of machinable Standard A parcels to have its undeliverable parcels returned to either specific postal facilities for pick-up by the shipper or for delivery to the shipper by the Postal Service after a payment of a fee for each returned parcel. Prior to its establishment, shippers relied upon Standard A single piece mail (which was discontinued after R97-1) as the only mailing option for the return of undeliverable merchandise. PRC Op. MC97-4/C97-1 at 3.

[6047] In Docket No. MC 97-4, the Commission accepted a stipulation and agreement offered by participants that set the piece fee at $1.75 and the annual authorization permit fee at $85.00. The $1.75 fee represented a cost coverage of 156 percent, the systemwide average at the time, as determined in Docket No. R97-1. It differed with the original Postal Service request by lowering the annual qualifying minimum volume for BPRS recipients from 50,000 to 10,000 pieces, while maintaining the requirement that qualifying parcels must weigh less than 16 ounces. Id. at 6. The Agreement also directed the Postal Service to complete a study that would develop an estimate of unit volume variable costs for BPRS. The Bulk Parcel Return Service Study Plan was completed and provided to the Commission on October 30, 1998, and included a total estimated FY 1998 volume variable unit cost for BPRS of $0.93. Bulk Parcel Return Service Cost Study (Oct. 29, 1998) (Postal Service 1998 BPRS cost study) at 7.

[6048] BPRS originally was restricted to parcels that were determined under postal regulations to be undeliverable as addressed (UAA) and thus unopened. PRC Op. MC99-4 at 1. In Docket No. MC99-4, the Postal Service proposed to change the definition of BPRS to include those parcels which had been opened, resealed and redeposited in the mail by the shipper's customer. It also provided for the shipper to include in each parcel a return label enabling the customer to return the parcel, under BPRS, at the expense of the shipper. It did not address fees. A settlement agreement to this effect was recommended by the Commission and approved by the Governors.

[6049] During the course of its consideration of Docket No. MC99-4, the Commission received a complaint initiated by the Continuity Shippers Association (CSA) objecting to the BPRS fee. The Commission will note, that as was pointed out both in briefs and in testimony, that even though CSA filed this complaint, it was a signatory to the agreement that led to the Commission's decision in MC99-4. CSA's complaint was received as Docket No. C99-4. CSA complained that the $1.75 per piece fee, in effect since Docket No. MC97-4, was excessive when compared to the costs identified in the Postal Service's Bulk Parcel Return Service Study.

[6050] For the purposes of Docket No. C99-4, and as a result of using the Commission's mail processing cost methodology and identifying additional cost data, the Postal Service revised its earlier BPRS unit attributable cost figure from $0.93 to $1.037. While there was some discussion regarding the appropriateness of certain cost adjustments to reflect FY 1998 to FY 2000 cost changes, the Commission adopted the $1.037 cost figure. PRC Op. C99-4 at 12.

[6051] The Postal Service filed its request in this case while Docket No. C99-4 was pending before the Commission. Contained in the Request is a proposal, based on still more updated cost and volume data, to decrease the per piece fee from $1.75 to $1.65. The Postal Service also requests that the matters initiated within Docket No. C99-4 be suspended and consolidated with Docket No. R2000-1. The Commission decided the arguments presented in the complaint case indicated a need to address the complaint promptly rather than consolidating it with Docket No. R2000-1. On April 14, 2000, the Commission issued its Decision in C99-4. The Commission recommended the Postal Service lower the per piece fee from $1.75 to $1.62 to reflect the previously agreed to 156 percent cost coverage. On June 5, 2000, the Governors rejected the proposed fee change in C99-4¹³.

[6052] Participant Positions. Postal Service witness Mayo proposes one classification and one fee change for BPRS. USPS-T-39 at 15. She proposes to lower, by six percent, the existing $1.75 per piece fee to $1.65 and to establish an annual advance deposit account fee "similar to the accounting fee for Business Reply Mail (BRM)" of $375.00. Mayo estimates test year unit costs for BPRS of $1.135 per piece and a cost coverage of 146 percent. Ibid.

[6053] Costs are provided by Postal Service witness Eggleston, who submits cost data based largely on the cost study performed as part of Docket No. MC 97-4, with corrections. USPS-T-26 at 30. Witness Buc, in testimony on behalf of the Continuity Shippers Association et al., provides a different cost analysis and suggests that the appropriate unit cost should be $98.2 cents without a contingency or 99.9 cents including a one percent contingency. He proposes a cost coverage of 132.9 percent, "the same as for Standard A Regular." He concludes that the fee for BPRS should be $1.33. Tr. 23/10643.

[6054] OCA dismisses the arguments for fee reductions from both the Postal Service and CSA and argues strenuously for maintaining a cost coverage of at least the systemwide average. OCA Brief at 215 et seq. OCA contends BPRS has a relatively high level of value of service and low price elasticity of demand. OCA states that certain service characteristics of BPRS provide high levels of intrinsic value of service. To support its argument it cites Buc's testimony that Cosmetique prefers to receive its returns via BPRS and that it now receives those returns directly from customers without having to go to Postal Service Mail Recovery Centers. OCA concludes that the existing rate of $1.75 is far more fitting, based on its high level of service, than either of those proposed by the Postal Service or CSA.

[6055] Throughout this Docket there has been a good deal of discussion as to the true nature of BPRS and what is a suitable proxy for identifying an appropriate contribution to institutional costs from this service. In her direct testimony, Eggleston estimates the volume variable costs of BPRS in five different components: collection, mail processing, transportation, delivery and postage due. Buc accepts the Postal Service's unit costs for the delivery and postage due components, but he argues that the Postal Service overestimates the unit costs for collection by 1.2 cents, mail processing by 6.6 cents and transportation by 3.8 cents. Tr. 23/10643.

[6056] Buc states that the Postal Service's collection cost component includes 1.16 cents per piece for window acceptance. The Service uses historical Standard A single-piece collection costs as a proxy for BPRS. Standard A window acceptance activities such as weighing, rating, and collecting postage are not performed for BPRS. Buc proposes that the collection costs for Merchandise Return are a more appropriate proxy for BPRS. Buc argues that Eggleston found that Merchandise Return parcels do not incur weighing and rating during window acceptance because they are incurred in the postage due unit, not at the window. BPRS parcels, according to Buc, are handled the same way. Tr. 23/10644. Thus, to avoid double counting, Buc concludes collection costs for BPRS should be reduced by 1.16 cents, from 3.22 cents to 2.06 cents. Ibid.

[6057] In response, the Postal Service complains that Buc does not understand the definition of "acceptance" as applied in the Eggleston cost study and that his comparison is incorrect because Merchandise Return "is a service for customers paying postage for an underlying class of mail. That postage would cover window costs for MRS parcels. Unlike MRS, the BPRS fee is a hybrid that must cover costs, such as acceptance costs, normally covered by postage." Postal Service Brief at VIII-7.

[6058] Buc also takes exception to the manner in which the Postal Service applies the CRA adjustment factor in its calculation of mail processing costs. In this instance he proposes that the Standard Mail Special CRA adjustment be reduced by 70 percent to reflect the cube and weight difference between BPRS and Standard Mail Special, which was used as proxy. The Postal Service's mail processing unit costs, as calculated by Buc, are overestimated by 6.6 cents. Buc focuses on the use of the Special Standard CRA adjustment, and the assumed inter- and intra-BMC weights. The Service uses both proportional and fixed CRA adjustment factors. The fixed cost pools that CRA adjustment factors are applied to contain both "expected" and "unexpected" costs. Buc argues that both expected and unexpected fixed costs are affected by cube and weight, similar to the proportional cost pools. Proportional costs are modeled at about 70 percent of the modeled costs of Special Standard B. Therefore, the fixed costs should have a fixed CRA adjustment factor that is 70 percent of the Standard B Special fixed CRA adjustment. By applying this 70 percent factor to the fixed costs, Buc concludes that the Postal Service has overestimated mail processing costs by 6.34 cents. Tr. 23/10645-10647.

[6059] The Postal Service argues that the cube and weight differences are already reflected in the modeled weighted average cost of BPRS, which had been reduced by 70 percent before the adjustment factor was applied. It further cited several reasons why BPRS parcels could be more expensive than Standard B Special Mail to process.

[6060] Additionally, Buc takes exception to the Postal Service's attribution of mail processing costs as they relate to inter-BMC versus intra-BMC. The Postal Service assumes that 95.2 percent of BPRS parcels are inter-BMC parcels and 4.8 percent are intra-BMC parcels. Buc alleges that this assumption is clearly incorrect since one of the eight BMC recipients surveyed (accounting for 3.5 percent of the BPRS volume) does not receive returns on a national basis. Buc assumes that all of this recipient's returns are intra-BMC. By incorporating this assumption, only 91.9 percent of BPRS parcels should be considered inter-BMC. Given a mail processing cost difference of 8.7 cents between inter-BMC and intra-BMC parcels, and the assumed lower percentage of inter-BMC parcels, Buc asserts the Postal Service has overestimated this portion of mail processing costs by 0.3 cents. Id.

[6061] Buc alleges the Postal Service made two erroneous assumptions that result in transportation costs being overestimated by 3.8 cents. First, zone distribution of inter-BMC parcels is the same as that for inter-BMC Parcel Post parcels, and second, only one out of every 21 BPRS parcels is intra-BMC.

[6062] When calculating inter-BMC transportation costs, Parcel Post inter-BMC volumes are distributed between eight zones. Buc notes that four of the eight BPRS recipients, accounting for 61 percent of BPRS volume, rarely receive packages from zones six through eight. Assuming that no recipient receives parcels from zones six through eight, Buc calculates inter-BMC transportation unit costs. He claims his transportation cost estimate provides a lower bound while the Postal Service's calculations provide an upper bound estimate, therefore, he averages the two cost estimates. He then subtracts his averaged estimate from the Postal Service's estimate and multiplies the result by 91.9 percent (his percentage estimate of total inter-BMC parcels) in concluding that the Postal Service has overestimated inter-BMC transportation costs by 2.9 cents. He reduces this by another 0.9 cents by applying his assumptions regarding the intra-BMC versus inter-BMC split, 91.9 percent of which are inter-BMC and claims a 0.9 cent difference in inter-BMC versus intra-BMC distributions. Buc concludes that total transportation costs are overestimated by 3.8 cents, assuming the 2.9-cent difference for inter-BMC zone costs and the 0.9-cent difference for intra- versus inter-BMC distributions. This argument is made more difficult in that as both CSA and the Postal Service agree, neither have any origin-destination data. Postal Service Brief at VIII-9.

[6063] During the process of updating for a 1999 base year, in response to Commission Order No. 1294, the costs increased over those originally submitted and discussed above. CSA on brief filed cost data that demonstrated costs by function for collection did not change, but mail processing, delivery and postage due increased by 0.9, 1.0 and 0.1 cents respectively. Transportation decreased by 1.7 cents. CSA argues that the new unit costs for BPRS are now 99.2 cents as opposed to their earlier 98.9 cents. CSA Reply Brief at 4-5. The Postal Service's updated information reported that the costs had increased to $1.315 from $1.105. USPS-T-26 Attachment Q at 1.

[6064] Commission Analysis. The Commission has found that the systemwide average cost coverage is often appropriate for new products and services, which often do not have actual cost data to support the rates requested. However, it never intended the application of the systemwide average to BPRS to preclude its ability to take into consideration other factors that would become available as the service reached its audience.

[6065] In this case the Postal Service has proposed a cost coverage of 146 percent and CSA has argued instead that the appropriate cost coverage should be the same as for Standard A Regular, or 133 percent. Both of these are significantly lower than the 156 percent applied to the existing rate. OCA argues for a coverage close to the systemwide average in the Postal Service's proposed case of 168 percent. OCA Brief at 215. The Commission believes BPRS should be evaluated as a special service that was ardently sought after by shippers of merchandise, including CSA. Thus, its high level of service and demand are self-evident and borne out by its history. Under these circumstances, a mark-up near systemwide average remains appropriate.

[6066] As to the individual costing arguments, those made by CSA as to collection costs have merit. The Commission agrees with CSA that BPRS parcels should receive little, if any, handling at the post office window. These parcels are small, less than 16 ounces, and the Commission agrees with CSA that the majority of these parcels will be returned by being placed in collection boxes or left for carriers, thereby avoiding window collection costs. The Commission therefore reduces the attributable collection costs to the 2.1 cents proposed by CSA on brief.

[6067] Since there were no disagreements among the parties as to the attributable costs for the Delivery or Postage Due components, the Commission accepts the amounts proposed by the Postal Service of 4.3 and 4.7 cents, respectively. Similarly, the Commission accepts the Postal Service's claims regarding transportation costs. While CSA made a valiant attempt to shed light on the transportation costs of BPRS, the fact is it is a new service with little data for light to be shed upon. Therefore, the Commission accepts the transportation costs of the Postal Service of 40.6 cents.

[6068] The arguments regarding mail processing costs made by CSA also have merit. Intuitively, if one reduces the modeled cost by 70 percent it makes sense to reduce its components by a similar number. Therefore, the Commission accepts CSA's proposed costs of 51.4 cents. Adding a appropriate amount for a contingency provides a unit cost for BPRS of $1.046, which is only slightly higher than the $1.037 calculated in C99-4.

[6069] Recommendation. The proposal to add an advance deposit account fee in the amount of $375 was unopposed. The fee is necessary, according to witness Mayo to recoup the costs of providing the accounting service and to make a small contribution to institution costs. USPS-T-39 at 17. The Commission agrees with the Postal Service's request for an advance deposit account fee and makes that recommendation.

[6070] The Commission recommends a per piece fee of $1.62 based on the Commission's cost analysis for BPRS. This rate is slightly lower than the rate proposed by the Postal Service. It provides a cost coverage of 155 percent, which is close to the systemwide average, and reflects the high value of BPRS posited by OCA.

[6071] Shipper paid forwarding provides forwarding service for a period of one year from the date a recipient files a change of address form. The service is available to eligible shippers mailing machinable standard parcels. Standard A mailers have the option to pay either the single-piece First-Class rate or the Priority rate for the return. Standard B mailers pay the applicable single piece rate for the return. Address change service is a prerequisite for shipper paid forwarding service.

[6072] The Postal Service proposes to add the availability of Parcel Post rates to shipper paid forwarding service. This will reflect the concurrent proposal to make Parcel Post rates available to parcels weighing one pound or less. The Postal Service also proposes to establish an annual accounting fee similar to the advance deposit account accounting fee for Business Reply Mail. The Business Reply Mail cost of $323.06 is used as a proxy for the shipper paid forwarding accounting fee cost in arriving at a $375.00 fee for this service. USPS-T-39 at 138-40. The Commission recommends the addition of Parcel Post rates to shipper paid forwarding and the annual accounting fee as proposed by the Postal Service. The Service proposes that this fee appear in a new Fee Schedule 936. The Commission recommends grouping all advance deposit account accounting fees in Fee Schedule 1000, and reserving Fee Schedule 936 for future use.

[6073] Certified mail is a special service for use with First-Class and Priority mail. The mailer receives a mailing receipt if the mail is deposited at a post office window or given to a rural carrier. The Postal Service obtains a signature upon delivery and retains a delivery record of the item mailed. Certified mail may be used in conjunction with return receipt or restricted delivery service. The fee for certified mail is in addition to postage. USPS-T-39 at 40-43.

[6074] The Postal Service proposes a 50 percent increase in the fee for certified mail from $1.40 to $2.10. The proposed fee divided by the estimated volume variable cost of $1.68 results in a cost coverage of 125 percent. However, certified mail exhibits a large difference between volume variable costs and incremental costs. The fee is not designed around a cost coverage, but is designed to cover the incremental cost of $2.00. Witness Kay asserts that 77 percent of the difference between incremental costs and volume variable costs can be accounted for in four city carrier (C/S 7) cost pools, Letter Route Load SDR, Letter Route Load MDR, Letter Route Load BAM, and Street Support Load. USPS-T-23 at 18. Witness Mayo states that certified mail is a high value service and acknowledges that a fee increase of this magnitude will have an adverse impact on users. Eight-two percent of certified mail customers also use return receipt service. Under the Postal Service rate proposal, the fee for mail sent using both certified mail and return receipt increases from $2.65 to $3.60, plus postage.

[6075] The Postal Service also proposes a DMCS classification change to reflect the use of electronic signature capture for certified mail. The Postal Service now scans signatures for accountable mail for inclusion in a centralized database, rather than storing hard copy signatures at each office of delivery. Thus, DMCS references to keeping delivery records at the "office of delivery" are proposed to change to "retention of delivery records by the Postal Service."

[6076] Intervenor Comments. Intervenor Popkin argues that a 50 percent rate increase should not be granted because provision of certified mail service is not in conformance with requirements for the service. Furthermore, he contends there is no alternative for certified mail other than an expensive expedited service, so mailers have no choice but to pay the increase.

[6077] Intervenor Carlson argues that delivery problems with certified mail lower the value of service. Therefore the Commission should reject an increase, or lower the rate for this service. He also compares return receipt for merchandise with certified mail plus return receipt and questions why the rate for certified mail plus return receipt is higher than for return receipt for merchandise given the similarities in the services. He states that both services provide a mailing receipt, a record of delivery, and a return receipt. However, the combined fee for certified mail plus return receipt return receipt is $3.60, and the fee for return receipt for merchandise is $2.35. Finally, he requests the Commission to recommend that the Postal Service cease advertising certified mail and return receipt, in the context of mailing income tax returns, until alleged delivery problems are fixed.

[6078] Commission Analysis. Certified mail should be a high value service that would justify a cost coverage above the system-wide average. In R97-1, the Commission recommended a smaller increase than proposed by the Postal Service to lessen the impact on the mailing public and to bring the increase closer to the system-wide rate increase. The high cost of this service still does not allow setting a fee providing a sufficient cost coverage without an adverse impact to the public.

[6079] The Commission acknowledges the intervenors' concerns with the allegations of certified mail delivery problems. There exists no national measures of the speed or reliability of certified mail, but anecdotal individual and press reports leave a persistent perception that service is inconsistent at best. The problems cited are in execution of the service, not with the concept of the service.

[6080] The Postal Service, if it has not already done so, must take appropriate corrective action to fix known problems so that certified delivery operates as advertised. It should also consider collecting data to evaluate the speed and reliability of this service so that problems can be identified and remedied. Persistent problems may indicate that something is wrong with the service itself and require a closer evaluation of the value of service in the future.

[6081] Comparisons to return receipt for merchandise are misplaced. Certified mail plus return receipt function in a different market than return receipt for merchandise, which results in significantly different costs. Certified mail plus return receipt is generally used with letters, whereas return receipt for merchandise is generally used for parcel shaped mail. The Postal Service provides several reasons that may explain the cost differences. Acceptance costs may be lower for return receipt for merchandise because more return receipt for merchandise parcels may be entered in bulk than certified mail plus return receipt parcels. Window costs for certified mail plus return receipt may be higher because two individual services are being sold as opposed to one service. Merchandise parcels with return receipt are generally larger than certified mail parcels. Thus, the delivery person is more likely to have to approach the door to make a delivery. This may lower the cost for return receipt for merchandise because some of the cost for return receipt for merchandise parcels is already being absorbed by the delivery person approaching the door with the host parcel. Return receipt for merchandise has the option of waiving the signature requirement. This may lower costs because the delivery person does not have to obtain a signature for a portion of the parcels, and delivery can be made on the first attempt. Finally, the Postal Service set the return receipt for merchandise cost coverage deliberately low to avoid sticker-shock because of the magnitude of the proposed price increase. Tr. 14/5790-2, Postal Service Reply Brief at VII-19-20. The Commission finds there are significant differences between certified mail plus return receipt and return receipt for merchandise.

[6082] The Commission has no recommendation for advertising certified mail, assuming that the Postal Service is, or has implemented corrective actions to address the alleged problems.

[6083] The magnitude of the proposed increase in the fee for certified mail is large enough that the Commission is concerned with the possibility of an adverse impact on consumers. The Commission is also concerned that the service problems described by the intervenors affect the value of service of certified mail.¹⁴ However, the magnitude of the rising cost of this service requiring an increase to cover costs is balanced with the concern for value of service. Therefore, the Commission recommends a rate less than the rate proposed by the Postal Service to mitigate the effect on the consumer. The Commission recommends a fee of $1.90 for certified mail. This rate represents a considerable increase of over 35 percent. At this rate, test year costs will equal $1.60, producing a 119 percent cost coverage. The Commission also recommends the proposed classification change to the DMCS language.

[6084] In its last omnibus rate case opinion, the Commission suggested that the Postal Service examine incorporating some of the information technology and infrastructure used for delivery confirmation to modernize certified mail. It should help lower costs compared to certified mail's relatively high cost of collecting delivery information manually. PRC Op. R97-1, para. 5924. In this docket, the Commission reiterates its concern with the rising cost of providing certified mail service. Although the Service has introduced electronic signature capture, the Commission suggests that the Postal Service explore other methods to reduce the cost of this service.

[6085] Registered mail is the most secure method of sending mail offered by the Postal Service. There is a system of receipts used to monitor and account for the flow of the mailpiece from acceptance to delivery. The sender is provided with a mailing receipt, and the Postal Service retains a delivery record of the item mailed. USPS-T-39 at 12126.

[6086] The Postal Service proposes a 23 percent total increase in the fees for this service. The fee for registered mail without insurance is proposed to increase from $6.00 to $7.25. Registered mail may also be purchased with insurance on a graduated scale up to the actual value of the mailpiece (from $0.01 through $25,000). The fee for the first step of registered mail with insurance would increase from $6.20 to $7.50, and each successive step is proposed to increase from $0.55 per step to $0.75 per step. A handling fee is imposed on items valued over $25,000. The Postal Service proposes to increase the handling fee from $0.55 per $1,000 to $0.75 per $1,000. Special arrangements must be made for items valued at over $15,000,000. Fees are assessed based on the special arrangements. The proposed total cost coverage for registered mail is 111 percent.

[6087] The Postal Service claims that registered mail provides a very high value of service, but aims for a moderate cost coverage to alleviate the impact that a more substantial rate increase would have on customers of this service. The Commission concurs and recommends the fees proposed by the Postal Service. These rates produce a 131 percent cost coverage.

[6088] This special service provides up to $5,000 in indemnity coverage for lost, rifled or damaged articles. The service is available for Express Mail, Package Services, and First Class Mail or Priority Mail that contains items that could be sent as Standard Mail. Express Mail provides for $500 of indemnity coverage free-of-charge. Above $500, insurance fees for Express Mail are charged based on each $100 increment, or fraction thereof. For Package Services and Standard Mail matter mailed at First-Class Mail and Priority Mail rates, no automatic insurance is provided; hence any level of indemnity coverage from $0.01 to $5,000 is assessed a fee. A per-piece discount is available for bulk mailers who mail a minimum of 10,000 insured mail pieces annually.

[6089] The Postal Service proposed increases in all existing fees for insurance. The current incremental fee of 95 cents between value levels is proposed to increase to $1.00. This proposed incremental fee increase also applies to Express Mail insurance value levels above $500. The proposed overall cost coverage for insurance is 138 percent. USPS-T-39 at 59.

[6090] OCA witness Collins argues that the Postal Service fails to provide any credible justification for the proposed increase and recommends no increase in the per $100 increment fee and a modification of the interval to $250 or $500 for insured value over $100 with a corresponding adjustment in the per increment fee. Collins implies that the Postal Service did not gather and analyze data on USPS insurance purchases. Collins states that without this data the Postal Service established whatever price it felt justified. Tr. 29/14197-8.

[6091] On brief, OCA implies witness Mayo's fee for unnumbered insurance (items valued at or below $50) is based upon erroneous cost information and must be reduced, and low insurance usage for values exceeding $900 may be due to a high fee relative to a low valued service. OCA comments that the value to the customer has not increased, the amount of competition has not changed, and the level of service has not improved. OCA Brief at 209-210.

[6092] Before hearings, witness Davis presented errata reducing the cost of unnumbered insurance to $0.95. USPS-T-30 at 14, revised April 17, 2000. Assuming the proposed unnumbered fee of $1.35, the cost coverage for this increment would increase from 104.5 to 138.6 percent, approximately the same as the overall insurance cost coverage. Mayo states that if she had known of the lower cost when developing the original fee, she would have proposed a lower fee, though a decrease in proportion to the cost decrease would not be warranted. Tr. 39/17726-27. Therefore the Commission could recommend a lower fee. Mayo indicated that a fee of $1.20 would eliminate an undesirably steep increase in the unnumbered insurance fee. Ibid.

[6093] The Postal Service proposes two classification changes applicable to bulk insurance. The first would establish separate bulk discounts for numbered and unnumbered insurance. The second classification change would extend bulk insurance to Standard A Mail. The intent is to meet the needs of more parcel mailers for insurance. The Commission agrees the change would provide a measure of safety to a mail class that does not have a high degree of speed of delivery, and that mailers should find this an enhancement to Standard A service.

[6094] Commission Analysis. As a whole, insurance covers its own costs and makes a moderate contribution to institutional costs. The fee increases are modest and should not have a negative impact. The Commission recommends an increase in the unnumbered insurance fee from $0.85 to $1.10 resulting in a 29 percent increase. For numbered insurance (items valued above $50.01) the Commission accepts the rate proposed by the Postal Service. The initial fee for items valued between $50.01 and $100.00 will increase from $1.80 to $2.00. The fee for each subsequent $100.00 in value will increase from $0.95 to $1.00.

[6095] The Commission also recommends new discounts for bulk insurance. The discount for unnumbered pieces will change from $0.40 to $0.60 with a pass through of 63 percent. The net fee for bulk unnumbered insurance will thereby increase from $0.45 to $0.50. For the numbered pieces, the bulk discount will increase from $0.40 to $0.80 which provides a 59 percent implied pass through. The overall cost coverage is 126 percent with an overall rate increase of 13.1 percent.

[6096] Collect on delivery (C.O.D.) is a service a mailer may use to mail merchandise prior to the recipient paying for the item. Upon delivery, the recipient pays for the merchandise plus the applicable C.O.D. fee. The Postal Service then transmits the amount collected back to the sender. Collect on delivery mail is insured. USPS-T-39 at 44-49.

[6097] The Postal Service proposes a classification change to increase the maximum value of the C.O.D. item from $600 to $1,000. The purpose of the increase is to attract a potential new customer base-those who purchase items using the Internet. This will give Internet purchasers who are reluctant to divulge credit card information over the Internet an optional payment method. This also provides Internet businesses that do not accept credit cards an alternate method of accepting payments. The Service further claims that increasing the maximum value will aid small businesses without a credit card payment option, that require payment prior to shipping, and that accept payment by check, by providing faster delivery of goods. Faster delivery results from eliminating the delay in receiving a check and waiting for the check to clear prior to shipment.

[6098] The fees are based on the C.O.D. amount to be collected, or insurance coverage desired for the particular item. The fee for items up to $50 in value would increase from $4.00 to $4.50. The fee for items up to $100 in value would increase from $5.00 to $5.50. The fee for each successive $100 increment, up to $1,000, would increase by $1.00 per step. The Postal Service does not propose changes to the fee for registered C.O.D. of $4.00, the notice of non-delivery fee of $3.00, or the alteration of C.O.D. fee of $3.00. A fee increase is proposed for restricted delivery C.O.D. from $2.75 to $3.20.

[6099] The Commission recommends the classification change to allow C.O.D. items valued up to $1,000. There may be a market for C.O.D. for items of this value given the rapid expansion of Internet services. The proposed rates are not opposed by any party. Although the Commission recommends expanding C.O.D. to a higher value limit, a low cost coverage remains appropriate for this service. See, PRC Op. R97-1, para. 5941. The Commission finds the proposed rates reasonable and thus, recommends the proposed rates for this service. These rates produce a 118 percent cost coverage.

[6100] Return receipt service is designed to provide customers with proof of delivery of mail. The return receipt customer receives the signature of the recipient, the delivery date, and the address where the mailpiece was delivered if it differs from the address on the mailpiece. A box is checked on the return receipt to indicate if the delivery address is the same as the address on the mailpiece.

[6101] "Regular return receipt" can be used with Certified, COD, Registered, Express Mail, and Numbered Insurance. "Return receipt for merchandise" can be used with Priority Mail and Package Services mail.

[6102] The Postal Service proposes two fee increases and one fee decrease to the return receipt fees. The regular return receipt fee proposal is an increase by 20 percent, from $1.25 to $1.50, with a cost coverage of 116 percent. The return receipt for merchandise fee would increase 68 percent from $1.40 to $2.35 with a cost coverage of 101 percent. The third change is a decrease in the return receipt after mailing fee by 50 percent, from $7.00 to $3.50 with a cost coverage of 153 percent. The overall cost coverage is 116 percent.

[6103] Intervenor Carlson asked witness Davis to explain the differences between the cost study for return receipt for merchandise conducted for Docket No. R2000-1 and the cost study conducted for Docket No. 97-1. Witness Davis indicates the new cost study includes updated wage rates, piggyback factors, labor times for clerk and carrier review functions, weighting factors, and retrieval time for return receipts after mailing. See USPS-LR-I-108 at 47-55. The new study bases the costs of return receipt for merchandise on the costs of certified mail, since the operations are similar. The costs of obtaining the return receipt signature, printing the return receipt, and returning the return receipt through the mailstream are added to the unit cost of certified mail. An adjustment is made to reflect the unit cost savings from the electronic signature capture process.

[6104] Intervenor Popkin questions the quality of service that return receipt provides to the mailer. Popkin requested "details of all studies and tests that have been performed or conducted by the Postal Service in the past ten years (since Docket R90-1) to determine the mailing public's needs and desires for return receipt service." Tr. 14/5417. In response Mayo refers to Plunkett's response to DBP/USPS-33 in R97-1. Id. at 5418.

The Postal Service utilizes many methods to ensure that employees and managers provide the services customers expect. The fact that studies have not been performed at a national level to determine the level to which the Postal Service has been able to meet this goal vis a vis return receipts should not be construed as meaning that there has been no concerted effort toward this end. Local managers have access to customer feedback via consumer service cards and other means. They are expected to utilize these data to improve their performance not only as regards return receipts, but for all products and services.

[6105] Commission Analysis. Since Docket 97-1 the costs for return receipt requested at time of mailing have increased by 29.9 percent because of the combined increases of various cost inputs. Return receipts are potentially a high value service, but persistent problems with the quality of service imply a lower cost coverage. The service fills an important function by providing the mailer with delivery information. Although the total return receipt service cost coverage suggested by the Postal Service is low, the proposed fees cover the costs of the service and contribute modestly to other costs.

[6106] The Commission recommends the fees for return receipts proposed by the Postal Service. The Commission also recommends a classification change to extend return receipt for merchandise service to Standard Mail Regular and Nonprofit parcels. The intent is to meet the needs of more parcel mailers for return receipt for merchandise service, which should in turn make Postal Service products more attractive. Providing this additional service option to those businesses that mail lightweight parcels should benefit mailers as well as the Postal Service. § 3623(c)(2) and (5). Also, the Commission recommends a language change in DMCS Section 945.25 from "duplicate return receipt" to "evidence of delivery from the delivery record." The new language more accurately describes what the Postal Service provides to customers.

[6107] Restricted delivery is a special service that allows a mailer to direct mail delivery only to the addressee or the addressee's authorized agent. The addressee or agent must be an individual (natural person) specified by name. Restricted delivery is only available in conjunction with certified mail, C.O.D. mail, mail insured for greater than $50.00, or registered mail. Restricted delivery is available at the time of mailing, or after mailing. If purchased after mailing, additional charges may be assessed to cover the cost of contacting the delivery post office. USPS-T-39 at 127-130.

[6108] The fee for restricted delivery is charged in addition to postage and any other applicable fee. The Postal Service proposes increasing the fee for restricted delivery from $2.75 to $3.20. Test year costs will equal $2.02. The cost coverage of 158 percent is close to the system wide average and is appropriate for a high value special service that is desirable to either mailers or the recipients concerned with controlling or restricting the delivery of mail. The Commission recommends the proposed fee of $3.20 for restricted delivery.

[6109] A certificate of mailing provides a mailer with evidence of mailing. It does not provide proof of delivery or insurance against damage or loss. Certificates of mailing are provided on Form 3817 for individual mailpieces, on Form 3877 (firm mailing book) for three or more mailpieces, and on Form 3606 for bulk mailings. A duplicate certificate of mailing is available for a fee at the time of the mailing or upon presentment of the original certificate. USPS-T-39 at 35-39.

[6110] The Postal Service proposes to increase the individual piece certificate from 60 cents to 75 cents, the individual piece duplicate fee from 60 cents to 75 cents, the bulk up to 1,000 piece certificate from $3.00 to $3.50, and the bulk duplicate fee from 60 cents to 75 cents. The proposed cost coverages are 124 percent, 156 percent, 118 percent, and 188 percent respectively. The firm mailing book fee is proposed to remain at 25 cents and the additional 1,000 piece bulk fee is proposed to remain at 40 cents. The associated cost coverages are 122 percent and 134 percent respectively. The proposed overall cost coverage for certificates of mailing is 123 percent. The fees are summarized in Table 5-23.

[6111] Intervenor Proposal. Intervenor Popkin argues that the fee for individual piece certificate of mailing should not be higher than the proposed retail fee for delivery confirmation. He alleges that delivery confirmation also provides proof of mailing, in addition to providing the confirmation of delivery. Thus, it is hard to understand why the fee for certificate of mailing is greater. He proposes a fee of 30 cents or less for certificate of mailing. Popkin Brief at 6-7. The Postal Service explains that window acceptance costs are greater for certificate of mailing, while delivery confirmation utilizes newer technology which exhibits greater cost efficiencies. Postal Service Reply Brief at VII-20.

[6112] Commission Analysis. The Commission finds that differences in the services and cost structures between individual piece certificate of mailing and retail delivery confirmation make comparison difficult. Delivery confirmation has approximately four times the volume of certificates of mailing. Delivery confirmation utilizes modern technology whereas certificate of mailing is basically a manual service. Certificate of mailing has approximately two-and-one-half times the acceptance costs of delivery confirmation. A mailing receipt is not provided for more than 20 percent of delivery confirmation articles. The total cost for delivery confirmation is less than the acceptance cost for certificate of mailing. Thus, the services and costs are different, and the rates for the two services can be set independently.

[6113] Intervenor Popkin's concern regarding the higher fee for a service of lower value is particularly telling because delivery confirmation is not available with First-Class Mail. Individual mailers could benefit if the Postal Service should make delivery confirmation available with First-Class Mail (including single-piece First-Class Mail). First-Class mailers would then have an option for a lower cost alternative to Certified Mail. The Postal Service may benefit also by eventually replacing what is essentially a manual service with a service that utilizes more modern technology, at a lower cost. Mailers should be eager to use delivery confirmation to obtain a mailing receipt since that would also get the added bonus of confirmation of delivery. The Commission suggests that the Postal Service consider expanding delivery confirmation to make it available with First-Class Mail. Consideration also should be given to offering this service to mailers sending large First-Class or Standard Mail envelopes.

[6114] Witness Mayo states that certificates of mailing provide a high value of service to individuals requiring proof of mailing. She acknowledges that the overall proposed cost coverage is moderate, which is low for a high value service. USPS-T-39 at 39. In R971, the Commission recommended an overall rate increase of 15 percent that corresponded with an overall cost coverage of 132 percent. The rate was approved noting that the increase brought the cost coverage closer to the system wide average. PRC Op. R97-1, para. 5953.

[6115] The Commission recommends the fees proposed by the Postal Service. The recommended fees result in an overall cost coverage of 124 percent. Although higher rates could be justified to bring the cost coverage closer to the system wide average, the recommended rates adequately cover costs and are selected to moderate the impact of the rate increase on mailers.

Table 5-23
Summary of Rates for Certificates of Mailing
	Current	Proposed	Recommended

Individual Pieces
Original Certificate (Form 3817)	$0.60	$0.75	$0.75
Firm Mailing Book (Form 3877)	$0.25	$0.25	$0.25
Duplicate Copy	$0.60	$0.75	$0.75
Bulk Pieces
Up to 1,000 Pieces (Form 3606)	$3.00	$3.50	$3.50
Each Additional 1,000 Pieces	$0.40	$0.40	$0.40
Duplicate Copy	$0.60	$0.75	$0.75
Source: Adapted from USPS-T-39 at 14.

[6116] Delivery confirmation provides a mailer with information about the date and time that an article was delivered, or, if delivery was not successful, the date and time of attempted delivery. Delivery confirmation can be purchased as a manual (retail) or an electronic (commercial) service. A mailing receipt is provided with the manual service. Mailers can access manual delivery information over the Internet or through the toll-free corporate call management system by telephone. Electronic delivery confirmation requires a mailer to apply a barcode to the mailpiece, and provide an electronic manifest to the Postal Service. An electronic link must be established with the Postal Service to exchange electronic acceptance and delivery data. Delivery confirmation is currently available with Priority Mail and Standard Mail B. USPS-T-39 at 54-59.

[6117] The Postal Service initially estimated total volume variable costs for delivery confirmation of: 17 cents for Priority Mail electronic, 52 cents for Priority Mail retail, 17 cents for Standard Mail electronic, and 52 cents for Standard Mail retail. USPS-T-30 at 3-7. Priority Mail includes the 17 cents of delivery confirmation costs within its base costs. Witness Davis subtracts the 17 cents accounted for with Priority Mail and adds the contingency to arrive at a net volume variable plus contingency cost for delivery confirmation of: 0 cents for Priority Mail electronic, 36 cents for Priority Mail retail, 17 cents for Standard Mail electronic, and 53 cents for Standard Mail retail. The Postal Service initially proposed fees of 0 cents for Priority Mail electronic, 40 cents for Priority Mail retail, 25 cents for Standard Mail electronic, and 65 cents for Standard Mail retail. USPS-T-39 at 55.

[6118] Intervenor Proposal. Parcel Shippers Association witness Zimmermann testifies that there should be "no charge to an electronic manifest Parcel Select mailer" for the use of delivery confirmation. He recognizes that there may be some volume variable costs, but argues that delivery confirmation should be a standard operating procedure. Tr. 29/14141-42.

[6119] In response to witness Zimmermann's testimony, Postal Service witness Davis reexamined the underlying cost study supporting the delivery confirmation fees. He concluded that there are costs associated with delivery confirmation service, but his original assumptions were too conservative and in many cases included activities that were not necessary, or that could be completed during the time for normal delivery activities. Therefore, witness Davis revised his cost estimates downward. Tr. 39/17426-31. As a result, the Postal Service would not object to a fee of 10 cents for Standard Mail electronic, and 50 cents for Standard Mail retail. Postal Service Brief at VIII-29-31.

[6120] Commission Analysis. Delivery confirmation is a relatively high value of service that conveniently provides delivery information to a mailer. In some ways, the value of service is not as high as return receipt for merchandise, or certified mail with return receipt, as these services provide for a signature upon delivery. On the other hand, major private delivery services that compete with the Postal Service have the cost of equivalent or better services than delivery confirmation built into the base price of their services. Delivery confirmation is a relatively new service that also can utilize a lower cost coverage to allow the service to develop to its fullest potential.

[6121] The Commission recommends a rate for the electronic service that has a cost coverage approximating the originally proposed cost coverage. In recommending rates for the manual services, the Commission also approximates the cost coverage originally proposed for Priority Mail manual service, and accepts witness Davis' revised cost estimates for Standard Mail retail. To achieve its cost coverage target the Commission recommends fees for delivery confirmation of 0 cents for Priority Mail electronic, 40 cents for Priority Mail retail, 12 cents for Standard Mail electronic, and 50 cents for Standard Mail retail.

[6122] The Postal Service proposes to extend delivery confirmation to Standard Mail Regular and Nonprofit pieces that pay the residual shape surcharge. The Service believes that mailers of these parcels have an interest in delivery confirmation. The revised fee proposed by the Postal Service for this new service is 10 cents-the same as Standard Mail B electronic. USPS-T-39 at 54-58, Postal Service Brief at VIII-31. The Commission recommends that this service be extended as proposed by the Postal Service, but at a rate of 12 cents consistent with the recommendation for Standard Mail electronic.

[6123] OCA Proposal. The OCA proposes that the Postal Service offer delivery confirmation service for Priority Mail to individuals through the Postal Service website at no charge. Tr. 29/14199-201. The OCA alleges that several Internet companies, including SmartShip.com, are currently offering no charge delivery confirmation service through their websites to individual shippers.

[6124] The Postal Service identifies costs that are associated with manual (retail) Priority Mail delivery confirmation that need to be recovered. See USPS-T-30 at 3-7. Furthermore, an individual does not perform the mail preparation tasks that qualify a high volume (commercial) mailer for the no fee Priority Mail electronic service. Although providing a free service to a mailer may be desirable from the mailer's viewpoint, it does not meet the § 3622(b)(3) requirement that delivery confirmation service bear its own direct and indirect postal cost. The Postal Service may, if it chooses, place delivery confirmation on its website at the recommended rate, or it may in the future propose a reduced rate for customers accessing delivery confirmation only through the internet. However, at this time the Commission does not recommend the OCA proposal to extend delivery confirmation service for Priority Mail to individuals through the Postal Service website at no charge.

[6125] The Postal Service proposes to establish signature confirmation as an independent service. Currently, delivery confirmation is a prerequisite to signature confirmation, with signature confirmation acting as a form of return receipt. The new service, as proposed, would include the functions of delivery confirmation within signature confirmation.

[6126] Signature confirmation provides a mailer with access to delivery confirmation information and an image of the recipient's signature. It can be purchased as a manual (retail) or an electronic (commercial) service. With the manual service, a receipt is provided containing the signature confirmation number. Mailers can access manual signature confirmation information over the Internet or through the toll-free corporate call management system by telephone. Electronic signature confirmation requires a mailer to apply a barcode to the mailpiece, and provide an electronic manifest to the Postal Service. Unlike delivery confirmation, users of electronic signature confirmation can access delivery information both over the Internet and through the corporate call management system. Signature confirmation may only be purchased at the time of mailing and is only available with Priority Mail and Standard Mail B. USPS-T-39 at 141-45.

[6127] The Postal Service's estimated total volume variable costs for signature confirmation are: $1.18 for Priority Mail electronic, $1.54 for Priority Mail manual, $1.18 for Standard Mail electronic, and $1.54 for Standard Mail manual. USPS-T-30 at 8-11. Priority Mail includes 17 cents of the delivery confirmation costs within the base delivery service costs. Witness Davis subtracts the 17 cents accounted for with Priority Mail and adds the contingency to determine a net volume variable plus contingency costs for delivery confirmation of: $1.04 for Priority Mail electronic, $1.40 for Priority Mail manual, $1.21 for Standard Mail electronic, and $1.58 for Standard Mail manual. The Postal Service proposes fees of $1.25 for Priority Mail electronic, $1.75 for Priority Mail manual, $1.25 for Standard Mail electronic, and $1.75 for Standard Mail manual. USPS-T-39 at 142. The corresponding proposed cost coverages are 120 percent for Priority Mail electronic, 125 percent for Priority Mail manual, 103 percent for Standard Mail electronic, and 111 percent for Standard Mail manual. The overall proposed cost coverage for delivery confirmation is 122 percent.

[6128] Commission Analysis. The Commission recommends the classification change to establish signature confirmation as an independent service, including the proposed changes to the DMCS language and the addition of Fee Schedule 949. The addition of signature capture makes signature confirmation a higher value service than delivery confirmation. Signature confirmation provides a mailer with comparable information to the return receipt service at a lower overall cost, when considering return receipt must be purchased in addition to another special service. The Postal Service cost should also be lower for signature confirmation given the electronic technology employed to capture the signature and delivery date. Because signature confirmation is a new service, a moderate cost coverage is justified. A higher cost coverage, to match signature confirmation's true value of service, may be appropriate in the future.

[6129] The Postal Service estimates cost for signature confirmation in a similar fashion to delivery confirmation. During the proceeding, witness Davis revised his cost estimates for delivery confirmation downward. Tr. 39/17426-31. On brief, the Postal Service states that the signature confirmation cost study already includes a 50 percent absorption factor for carriers' retrieval and replacement of scanners. Thus, the revised delivery confirmation cost assumptions may have little effect on the signature confirmation cost estimates. Postal Service Brief at VIII-46. In light of witness Davis' new assumptions for delivery confirmation and a 50 percent absorption factor for signature confirmation, the Commission estimates costs for signature confirmation of: $0.97 for Priority Mail electronic, $1.50 for Priority Mail manual, $0.97 for Standard Mail electronic, and $1.50 for Standard Mail manual. The Commission recommends the rates for signature confirmation as proposed by the Postal Service. This produces a 124 percent overall cost coverage.

[6130] Parcel airlift provides air transportation of parcels to military post offices outside the 48 contiguous states, for onward dispatch to other overseas military post offices, or for parcels from overseas military post offices to post offices within the 48 contiguous states. The service is available on a space available basis for Standard Mail B parcels limited to 30 pounds in weight, or 60 inches in length and girth combined. The Postal Service is expecting no volume or revenue from this service during the test year. Customers are choosing Priority Mail over parcel airlift, often at a lower price. The Postal Service proposes retaining the current fees for parcel airlift while it considers the ramifications of eventually eliminating this service. USPS-T-39 at 88-91. The Commission recommends retaining the current fees for parcel airlift.

[6131] Special handling provides preferential handling of a mailpiece during processing and transportation, but does not provide preferential delivery. A mailer may request the service with First-Class Mail, Priority Mail, and Standard Mail B. Special handling is required when shipping honeybees or baby poultry using Standard Mail B. USPST39 at 146-148.

[6132] In Docket No. R97-1, the Postal Service proposed to triple the special handling fees, provoking a negative response from several beekeeper and poultry organizations. The Commission questioned the accuracy of the IOCS costs reported for special handling and noted the devastating impact that such a large increase in fees could have on small businesses that rely upon this service, a service with few alternatives. The Commission concluded that pending the completion of a special study of costs, no increase could be justified based on the R97-1 record. PRC Op. R97-1, paras. 5989-5997.

[6133] The Postal Service, in the instant docket, proposes retaining the current fees for special handling. Special handling transactions have decreased dramatically from 15 million pieces in 1970 to 39 thousand pieces in 1998. The low volume exacerbates the problem of accurately tracking costs and brings into question the reliability of the cost and volume estimates in the CRA reports (there were only two tallies with encircled special handling activity codes in FY98). The Service considered whether a special study could be designed to quantify the special service costs. From field observations however, the Service concludes that costs would be difficult to measure through a special study. Those observations also cause the Postal Service to conclude the BY98 CRA estimate of $57 per transaction overstates the actual special handling volume variable unit cost. USPS-T-28 at 30-31.

[6134] The cost estimates for special handling, as was the case in Docket No. R97-1, are questionable. Without sufficient record evidence, the Commission cannot recommend a change to the special handing rates.

[6135] Stamped envelopes, i.e., envelopes with postage pre-attached, may be purchased from the Postal Service as individual envelopes, in household quantities of 50, or in bulk quantities of 500. Size 6-3/4, size 10, and intermediate size envelopes are available. Envelope formats include basic (no window), single window, and double window. They may be purchased with printed personalized information, or purchased plain. Envelopes may be printed with different postage values, including pre-cancelled.

[6136] The Postal Service proposes to increase the fees for all categories of stamped envelopes, and to make three classification changes to the service. Fees are based on modeled costs, the contingency, and an applicable markup. The modeled cost components for stamped envelopes are manufacturing costs, distribution costs, and selling costs. Contract costs from July 1, 1999 through June 30, 2000 are used in the model as a proxy for test year manufacturing costs, because a manufacturing contract is not in place for fiscal year 2001. Costs were modeled for 59 different style/quantity characteristics of envelopes. USPS-T-29 at 31-37. The Service then groups the different envelope characteristics under 10 different classifications. Fees, and the associated cost coverages, are determined using the highest cost style/quantity characteristic within each classification. The fee changes are summarized in Table 5-24. The proposed stamped envelope overall cost coverage is 128 percent. USPST39 at 152-61.

[6137] The Postal Service proposes three classification changes to the stamped envelope special service. The first change is to merge the household #6-3/4 and #10 categories into a single household basic category. The costs for the two current categories are similar. The second proposal is to eliminate the banded #6-3/4 and #10 categories. This will reflect that banded envelopes are not sold in box lots. They are only sold in vending machines at the single sale fee. The final proposal is to rename the hologram designation to "special." Special will reflect the higher value, and higher cost, of an envelope that may present a fancier appearance. However, manufacturing has been discontinued for the hologram envelopes because the patched in stamps are not recyclable. The special envelope category includes patched stamp envelopes in the event that these envelopes are manufactured in the future.

[6138] The Commission recommends the proposed classification changes to merge the household #6-3/4 and #10 categories into a single household basic category, to eliminate the banded #6-3/4 and #10 categories, and to rename the hologram designation "special." Parties have not opposed these classification changes.

[6139] The rates proposed by the Postal Service are high when compared with the Commission recommended rates in Docket No R97-1, which resulted in a 105 percent cost coverage. Witness Mayo bases rates and cost coverages on the highest costs in each category presented by witness Campbell. This distorts the proposed cost coverage to appear lower than the actual cost coverage. However, the Commission recommends the rates as proposed by the Postal Service. Stamped envelopes provides a relatively high value of service when viewed as a convenient method of purchasing a limited number of envelopes. The Commission estimates that the actual cost coverage is 148 percent for stamped envelopes.

Table 5-24
Summary of Rates for Stamped Envelopes
		Current	Proposed	Recommended

Single Sale:
	Basic	$0.07	$0.08	$0.08
	Special (Hologram)	$0.08	$0.09	$0.09
Printed Household:
	6-3/4 Basic (50)	$3.00	N/A	N/A
	10 Basic (50)	$3.25	N/A	N/A
	Basic (50)	N/A	$3.50	$3.50
	Special (Hologram)	$3.50	$4.50	$4.50
Plain 6-3/4: Banded (500)		$9.50	N/A	N/A
Plain 10: Banded (500)		$12.00	N/A	N/A
Plain 6-3/4 Basic (500)		$8.50	$12.00	$12.00
Plain 10 Basic (500)		$11.50	$14.00	$14.00
Plain 10 Special (500) (Hologram)		$15.50	$19.00	$19.00
Printed 6-3/4 Basic (500)		$14.00	$17.00	$17.00
Printed 10 Basic (500)		$15.00	$20.00	$20.00
Printed 10 Special (500) (Hologram)		$19.00	$25.00	$25.00
Source: Adapted from USPS-T-39 at 154.
Fees do not include the price of postage. Basic envelopes include regular, window, pre-cancelled regular, and pre-cancelled window. The special envelopes are those with patched in stamps.

[6140] Intervenor Argument. Intervenor Popkin argues, on brief, for the elimination of the shipping and handling charge imposed on orders placed with the Philatelic Fulfillment Service Center. Popkin Brief at 9. The Commission, in dismissing a complaint on the same subject matter, ruled "that the services involved - the handling and shipping of catalog orders placed with the Philatelic Fulfillment Service Center - are not closely related to the delivery of mail and, therefore, the charges for such services do not constitute "fees for postal services" within the scope of section 3662 of title 39, United States Code." See Docket No. C95-1, PRC Order No. 1075. Because the charges in question do not constitute "fees for postal services," and consistent with the prior ruling, the Commission declines to make a recommendation concerning the shipping and handling charges imposed by the Philatelic Fulfillment Service Center.

[6141] DMCS Fee Schedule 961. The Postal Service proposes to replace the current DMCS Fee Schedule 961 with a revised fee schedule. The proposed fee schedule omits information on envelope configurations and options that may be valuable to a customer purchasing envelopes. Descriptive terms such as regular, window, double window, pre-cancelled, multi-color printing, and savings bond, that are in the current fee schedule, are removed in the proposed fee schedule. Although the Commission recommends restructuring Fee Schedule 961 to streamline the information displayed, it also recommends retaining much of the information that alerts a consumer to the available stamped envelope options.

[6142] Stamped Cards are cards with postage pre-affixed to the card. They are available from the Postal Service for the price of First-Class postage, plus a fee for the card. The Postal Service proposes to double the fee for a single card from 1 cent to 2 cents, not including postage. Correspondingly, the fees for a double stamped card will increase to 4 cents, and for a sheet of 40 stamped cards to 80 cents. This results in a proposed cost coverage of 139 percent given the Postal Service projected cost of 1.4 cents for printing, materials, and distribution of a single card, including contingency. USPS-T-39 at 149-51.

[6143] Intervenor Comments. Intervenor Popkin argues that the sale of stamped cards for more or less than its face value is in violation of 18 U.S.C. 1721. Popkin Brief at 7. This view is not consistent with the Commission's interpretation of the statute. See, e.g., P.O. Ruling R97-1/31.

[6144] Intervenor Carlson argues against increasing the stamped card fee. He asserts that the mail processing costs for stamped cards is less than the mail processing costs for private post cards. Carlson calculates the total cost for a mailed stamped card at the proposed rates is 23 cents (21 cents for postage, plus 2 cents for the stamped card) versus 21 cents for a private post card (21 cents for postage). He infers that this cost relationship is backwards because the higher priced product incurs lower processing costs. Carlson also claims that because of the cost differences, stamped cards are unfairly subsidizing the larger cards subclass. Carlson Brief at 20, see also Popkin Brief at 9. The Postal Service points out that the stamped card special service fee just covers the printing and manufacturing costs of the stamped card. Postage and mail processing costs are First-Class Mail Cards rate design issues. Even if there is a mail processing cost difference between stamped cards and private post cards, the Postal Service asserts there is no record evidence in this docket to show the difference. Postal Service Reply Brief at VII-30.

[6145] Commission Analysis. Concerning intervenor Carlson's proposal to maintain the current stamped card fee, the Commission finds that the special service stamped card fee does not include mail processing costs. Thus, the postage rate for stamped cards is a separate issue from the issues being examined under this special service. Therefore, arguments based on mail processing costs do not affect the determination of the fee for this special service.

[6146] The Postal Service claims that stamped cards provide a high value of service justifying a moderate cost coverage. Businesses may save labor costs by not having to affix postage, and individuals may benefit by purchasing their stationary and postage at the same time. In PRC Op. R97-1, para. 6005, the Commission expressed an interest in a relatively low cost coverage so that this service will provide a low cost method by which individuals can send mail. In this docket, the Commission is constrained from setting a lower rate by the whole cent rounding criteria. This service has costs above one cent per card. Therefore, the Commission recommends the rates for stamped cards as proposed by the Postal Service. These rates produce a 295 percent cost coverage.

[6147] The Postal Service offers a domestic money order, an APO/FPO (military) money order, and an inquiry service. Traditionally, domestic money order users have been thought to be people with modest income levels, people without checking accounts, or people without credit cards. Postal money orders are popular in rural areas that do not have access to alternative money order services. Money orders also may become a popular means of transacting Internet business. APO/FPO money orders are generally sold to military personnel at military installations. Both domestic and APO/FPO money orders can be issued up to a maximum amount of $700. Inquiry service verifies whether a postal money order was cashed, and provides a copy of the paid money order. USPS-T-39 at 73-79.

[6148] The Postal Service proposes to increase the domestic money order fee from 80 cents to 90 cents, the APO/FPO money order fee from 30 cents to 35 cents, and the inquiry fee from $2.75 to $3.00. The corresponding overall proposed cost coverage is 198 percent. The Service contends that because incremental costs are high in relation to volume variable costs, the ratio of revenue to incremental costs is 142 percent. The fees are designed to create a moderate overall cost coverage, while keeping fee increases reasonable.

[6149] The Postal Service foresees the Internet as a potential new source for money order customers. For example, Internet auction cites may require payment by money order, customers may be reluctant to provide personal information such as credit card numbers over the Internet, and business can potentially be lost by waiting for traditional checks to clear. This new market for money orders is in contrast to the accepted view that money order customers may be people of modest income. It suggests this new customer base may support higher fees for money orders. Id. at 78.

[6150] OCA Proposals. The OCA proposes lowering the fee for domestic money orders by 5 cents, reducing the current 80-cent fee to 75 cents. Tr. 29/14187-193. The OCA alleges the current and proposed fees are too high in relation to those recommended in previous dockets. In making its recommendations, the OCA notes that money orders have a large difference between volume variable and incremental costs, and recognizes that different treatment of the non-fee revenue will result in different cost coverages. In support of arguments that the Postal Service money order fees are high, and that lowering the fees will not hurt competition, the OCA performed a limited survey in the Washington, D.C. area. The survey shows the majority of money order outlets charging less than the fee proposed by the OCA for money orders. Id. at 14191. Finally, the OCA cites the Commission's historic preference for mitigating increases in money order fees since a large proportion of money order users are people with modest incomes, people lacking access to financial institutions, or people living in rural areas.

[6151] The OCA also proposes lowering the fee for APO-FPO money orders by 5 cents, reducing the 30-cent fee to 25 cents. Id. at 14193-194. These money orders are available to military personnel at APOs and FPOs. The retail transaction costs associated with APO-FPO money orders are borne by the military. The OCA presents a philosophical argument that military personnel serving this country have limited banking options available outside the United States, and also because of their service, should pay the lowest possible fee.

[6152] Commission Analysis. The Commission views money orders as a vehicle for people of modest means or limited access to financial alternatives to pay for necessary goods and services. For military personnel, isolated rural residents, or people lacking access to credit cards and checking accounts, money orders facilitate these groups transacting business. Rates should not be raised, and in fact should be lowered, for this service when attributable costs are more than adequately covered. The Postal Service argues for a cost coverage that is high when this service can justify a very low cost coverage. The Service attempts to justify higher rates for money orders by requesting the Commission to consider the positive potential impact that the Internet may have on money order sales. The desirable aspects of using money orders for Internet transactions would suggest that money orders provide a high level service. While there is merit in this argument, increasing rates would abandon the people that most need this service.

[6153] Therefore, the Commission recommends reducing the money order fees by 5 cents to reduce the money order cost coverage to a more appropriate level. The information on other money order providers contained in the OCA proposal assures the Commission that this recommendation results in rates that are in line with the marketplace. The Commission recommends a fee of 75 cents for a domestic money order and 25 cents for an APO/FPO money order. The Commission also recommends that the inquiry fee remain at $2.75. The corresponding overall cost coverage is 153 percent.

[6154] OCA Internet Proposal. The OCA requests that the Postal Service consider offering a new service for the sale of money orders over the Internet. With merchandise sales over the Internet increasing rapidly, the OCA sees consumers and small businesses conducting Internet business as a potential source of new money order customers. Several suggestions are presented as to how this new service can be offered. Tr. 29/14194-196. While the Commission can not recommend a proposal for money order sales over the Internet on this record, the Postal Service might explore this alternative revenue generating idea.

[6155] Mailing Online is a three-year experiment under which individuals and businesses transmit documents to the Postal Service via the Internet for printing, finishing, and posting as hard copy mail. The Mailing Online Opinion and Recommended Decision, PRC Docket No. MC2000-2, issued June 21, 2000, reflects the Commission's recommended rates and classifications for this special service. The Decision of the Governors of the United States Postal Service on the Recommended Decision of the Postal Rate Commission on Mailing Online Experiment, Docket No. MC2000-2, issued August 7, 2000, approved the Commission's Opinion and Recommended Decision.

[6156] The Postal Service initial filing does not contain rate or classification proposals for this service. However, the Postal Service has proposed a change to the Mailing Online certification fee on brief. This fee is discussed along with the other annual permit fees. The Commission does not recommend rate or classification changes to Mailing Online, other than the level of certification fee.

[6157] The Postal Service proposes fee increases and three classification changes to permits. The Postal Service proposes to increase the permit fee from $100 to $125 for: (1) Business Reply Mail (BRM), (2) bulk parcel return service, (3) First-Class presort, (4) merchandise return, (5) permit imprints, (6) destination entry Standard Mail B, (7) Standard Mail A bulk, and (8) Standard Mail B special and library presort. USPST39 at 96-98. The Commission recommends an increase in the permit fee to $125. The resulting low cost coverage of 106 is appropriate for a permit fee that allows access to other higher value services. The Commission also recommends grouping all permit fees under Fee Schedule 1000.

[6158] Subsequently, on brief, the Postal Service proposes an increase to the certification fee for Mailing Online from $100 to $125. The purpose of this increase is to maintain consistency with other permit fees proposed in this proceeding. Postal Service Brief at VIII-41. The impact of the proposed increase is minimal because the foreseeable universe of potential customers that might desire certification for Mailing Online is limited. Although the Postal Service made this proposal extremely late in the proceeding, the overall impact of the proposal, is small. Therefore, the Commission recommends consistent increases in all certification fees.

[6159] The Postal Service also proposes changes to the periodicals applications fees contained in Fee Schedule 1000. It proposes to increase the original entry fee from $305 to $350, decrease the re-entry fee and registration for news agents fee from $50 to $40, and maintain the additional entry fee at $50. USPS-T-39 at 92-95. No participant commented on these proposals. The Commission recommends the changes to the periodicals applications fees as proposed by the Postal Service.

[6160] The Postal Service proposes three classification changes. The first classification proposal is to make the language consistent in DMCS sections 280, 380, and 581 through 584 regarding annual mailing fees. The Postal Service states that the changes will have no effect on the current administration of the payment of permit fees. The second classification proposal is to divide the "Special and Library Standard Mail Presort Mailing" fee into separate fees for "Media Mail Presorted Mailing" and "Library Mail Presorted Mailing", with no change to the rate. The final classification proposal is to establish a permit fee for destination entry Bound Printed Matter (BPM). The Commission recommends these changes.

[6161] The Postal Service performed a general review of each Domestic Mail Classification Schedule (DMCS) special services section and offers proposals designed to revise each section to improve clarity, consistency, and organization. USPS-T-39 at 165-166. During the proceeding, the Postal Service identified several other changes to the DMCS special services sections that augment their original proposals. Tr. 39/17673-76, Tr. 39/17681-85. The proposals generally improve the readability of the DMCS special services sections.

[6162] Included within the rewrite proposal, the Postal Service proposes to eliminate the listings of services that are available in conjunction with other services. The Postal Service proposes to list only services that have the first service as a prerequisite. The Service also states a preference to specify the listings of services that are available in conjunction with other services in the Domestic Mail Manual (DMM).

[6163] The Commission perceives a benefit to the mailing public in maintaining a complete listing of services that are available in conjunction with other services within the DMCS. The information contained in an accurate listing is helpful in understanding the scope of different mailing options. Therefore, the Commission does not recommend eliminating the listings as proposed by the Postal Service at this time.

[6164] The Commission has reviewed the current DMCS, the current DMM, and the Postal Service's proposed DMM changes presented in Federal Register, Vol. 65, No. 168, August 29, 2000, to determine if an accurate listing could be developed. Upon comparing the information contained in the above documents, along with the Commission's understanding of each special service, it became apparent that there are potential inconsistencies that need to be explained. For example, compare DMCS § 945.21, proposed DMCS § 945.21, DMM § S915.1.2, proposed DMM § S915.1.2 (Federal Register, Vol. 65, No. 168, August 29, 2000), and proposed DMM § S915.1.7 (Id.), concerning return receipt service. These sections contain several apparent inconsistencies. For instance, compare the different availability of registered mail, delivery confirmation, and restricted delivery with return receipt. Similar examples can be found with other special services. Because of this, a review is necessary before the Commission can recommend changes throughout the DMCS to the listings of services that are available in conjunction with other services. The Commission's DMCS recommendation maintains the current listings until this area can be thoroughly reviewed and clarified.

[6165] The Commission recommends two additional organizational changes that affect both the text and the fee schedules. The Commission recommends grouping all permit fees under Fee Schedule 1000 and grouping all accounting fees for advance deposit accounts fees under Fee Schedule 1000. This adds convenience in locating permit fee and accounting fee rates, and facilitates modifying the fee schedules when changes occur.

[6166] The Commission has incorporated a majority of the Postal Service proposals into its DMCS recommendations. The Commission recommendations for the DMCS text appear in Appendix Two and recommendations for the DMCS Fee Schedules appear in Appendix One.

The current groups box fees are assessed according to groups defined in the Domestic Mail Manual (DMM, § D910.5). Group A fees are the highest, Group B fees, the second highest, etc. Group E fees apply to customers who are ineligible for carrier delivery for postal policy reasons.

As a consequence of P.O. Ruling 11, which required certain underlying data that was applied to Yezer's model to be filed under seal, Popkin's initial brief also was filed under seal. This requires that certain facts not be disclosed in the decision.

Yezer defines Erents as the dependent variable in his equation #2. They are computed based on a forecast of new five-year leases with specific terms that will be executed in the future. He notes that actual rents may be for leases executed years ago, for time periods other than for five years, and under various lease terms, stating that Erents permit direct comparisons among facilities while actual rents do not.

⁹

In fact, Bentley's overall estimates may understate the efficiency of the handling techniques used in the test year for low and high volume accounts. The Service discloses that a national QBRM task force is now studying the development of "best practices" to be deployed locally. Tr. 21/9466-67. Further, it is reasonable to assume that cost-based accounting and per-piece fees should encourage QBRM volume. As volume increases, the ability to use relatively more efficient counting techniques should rise also.

¹¹

In telling cross-examination on the comparison between non-letter size BRM and QBRM, Campbell agrees that the non-letter BRM processing was 7,272 piece per hour. Tr.14/6173. He agreed that "this is a productivity for . . . big, bulky, nonletter size BRM pieces"; and that it would be less costly to apply weight techniques to letter size QBRM than it would be to apply weight techniques to nonletter sized bulky packages: "I would expect that productivity to be a conservative productivity." Id. at 6174, 6175. "[O]ne could apply this productivity and perhaps obtain a reasonable estimate as [sic] weight averaging productivities for letter size mail." Id. at 6175. In his rebuttal statement, Campbell, noting that letter-sized material is processed at 6,390 PPH while bulkier pieces are processed at a rate of 7,272, says: "The relationship between these two productivities is counter intuitive." Tr. 39/17501. He continues: "unlike for nonletter-size BRM, the Service has developed no standards or procedures for applying weight averaging to trays of letters." Id. at 17501-02. The Commission notes that the 1987 Reply Mail Study was conducted at just one site. Tr. 14/5992. This suggests the study was not necessarily representative.

¹⁴

See also Area Coordination Audit, Special Services, May 18, 1999, Case No. 040-1241887-PA(2), U.S. Postal Inspection Service Northeast Division, Final Report, filed as USPS-LR-I-200 in response to DFC/USPS-24. Tr. 21/8833. Although not admitted into evidence, this report discusses findings and recommendations of the U.S. Postal Inspection Service concerning problems with certified mail. The certified mail section of this report was also filed in response to DFC/USPS-T39-3. Tr. 14/5468-74.