BEFORE THE
POSTAL RATE COMMISSION
WASHINGTON, DC 20268-0001
POSTAL RATE AND FEE CHANGES,
2006
)
)
Postal Rate Commission
Submitted 12/21/2006 4:17 pm
Filing ID: 55484
Accepted 12/21/2006
Docket No. R2006-1
BRIEF OF
NATIONAL ASSOCIATION OF PRESORT MAILERS
AND NATIONAL POSTAL POLICY COUNCIL
ON RATES FOR PRESORT FIRST-CLASS MAIL
David M. Levy
Richard E. Young
SIDLEY AUSTIN LLP
1501 K Street, N.W.
Washington, DC 20005-1401
(202) 736-8000
[email protected]
Counsel for National Association of
Presort Mailers and National Postal
Policy Council
December 21, 2006
TABLE OF CONTENTS
Page
I.
II.
III.
INTRODUCTION........................................................................................1
A.
Interests Of NAPM And NPPC ........................................................1
B.
Summary Of Argument....................................................................2
THE COMMISSION SHOULD RECOMMEND THE PRESORT
RATE DIFFERENTIALS PROPOSED BY THE POSTAL SERVICE
FOR FIRST-CLASS MAIL. ........................................................................4
A.
The Efficient Component Pricing Rule (“ECPR”) Is The Proper
Standard For Setting Presort Rate Differentials. .............................4
B.
The Relevant Presort Benchmark Under ECPR Is Single-Piece Mail
With The Cost Characteristics Of Collection Mail, Not BMM. ..........7
1.
The relevant presort benchmark under ECPR is
single-piece mail at the margin of conversion. ......................9
2.
The marginal piece of single-piece mail has the cost
characteristics of collection mail, not BMM. ........................12
3.
The BMM benchmark does not achieve a uniform perpiece contribution to institutional costs. ..............................21
4.
Basing presort discounts on the marginal piece of
mail does not give existing presort mail an
inappropriate reward...........................................................24
5.
Delinking of presort and single-piece rates is a
reasonable proxy for an ECPR-compliant presort
benchmark. .........................................................................24
C.
EPCR Requires Analysis Of All Presort-Related Cost Pools.........26
D.
The Presort Rate Differentials Proposed By The Postal Service
Would Pass Through Less Than 100 Percent Of Avoided Costs. .28
THE COMMISSION SHOULD RECOMMEND OTHER CHANGES
IN FIRST-CLASS RATE DESIGN PROPOSED BY THE POSTAL
SERVICE .................................................................................................29
A.
Greater Recognition Of Shape-Related Cost Differences .............29
B.
Reduced Extra-Ounce Rate. .........................................................30
CONCLUSION....................................................................................................31
BEFORE THE
POSTAL RATE COMMISSION
WASHINGTON, DC 20268-0001
POSTAL RATE AND FEE CHANGES,
2006
)
)
Docket No. R2006-1
BRIEF OF
NATIONAL ASSOCIATION OF PRESORT MAILERS
AND NATIONAL POSTAL POLICY COUNCIL
ON RATES FOR PRESORT FIRST-CLASS MAIL
(December 21, 2006)
The National Association of Presort Mailers (“NAPM”) and the National
Postal Policy Council (“NPPC”) respectfully submit this brief on rates for Presort
First-Class Mail.
I.
INTRODUCTION
A.
Interests Of NAPM And NPPC
NAPM represents presort mailers, the majority of which are presort
bureaus, on postal issues and policies that affect the businesses of NAPM
members. Members of NAPM now presort over 25 billion pieces of First-Class
Mail annually, of which more than 23.75 billion are prebarcoded.
NPPC is a trade association representing large business users of First
Class Mail, including those in the telecommunications, banking and financial
services, utilities and insurance industries. The members of NPPC rely heavily
on Presort First-Class Mail.
B.
Summary Of Argument
The Commission should recommend the rates and rate differentials
proposed by the Postal Service for presort First-Class Mail.
The Efficient
Component Pricing Rule (“ECPR”) requires that discounts offered to mailers for
presort and other private sector activity that reduces Postal Service costs should
be set equal to the costs thereby avoided by the Postal Service.
Rate
differentials smaller than cost avoidances violate ECPR and frustrate the goal of
achieving lowest combined costs, because such prices encourage mailers to
demand from the Postal Service a variety of activities that the private sector
could supply at a lower cost.
In this docket, the Postal Service has proposed to sever the traditional
linkage between single-piece and presort First-Class rates, and set rates for the
two subclasses so that the average contribution per piece in each subclass is
approximately the same. This approach produces lower presort rates than the
traditional approach, which relies on estimates of the costs of processing Bulk
Metered Mail (“BMM”), a hypothetical category of single-piece First-Class Mail
consisting of “machinable, homogeneous, non-barcode pieces with machine
printed addresses that are properly faced and entered in trays.”
The Commission should recommend the presort rates proposed by the
Postal Service, and can do so either by approving delinking or by preserving
linking but adopting an appropriate cost benchmark. Single-piece First-Class
Mail varies greatly in its physical characteristics and costs. Many of the factors
that cause this heterogeneity are not recognized fully—or recognized at all—in
the rate structure. Because the resulting rate structure is so heavily averaged,
the proper presort benchmark under ECPR is not the cost of an average piece of
-2-
single-piece mail, let alone the cost of unusually low-cost pieces exemplified by
the BMM paradigm, but the cost of single-piece mail that is at the margin of
converting to presort today. As NAPM witness Bell and USPS witness Taufique
have testified, the latter mail is far costlier than assumed by the hypothetical
BMM paradigm.
APWU objects that abandonment of the BMM benchmark would result in
nonuniform
contributions
to
institutional
costs
by
individual
First-Class
mailpieces, but the BMM benchmark produces the same kind of nonuniformity.
No presort benchmark can yield uniform per piece contribution to cost from a
mail class as heterogeneous as First-Class Mail as long as its rate structure is so
heavily averaged. Rather than pursue this chimera, the Commission should set
rates that give mailers proper signals for decisions that result in the lowest
combined cost of the Postal Service and private sector suppliers.
APWU’s further contention that the rates proposed by the Postal Service
would give an undue reward to mail that would be presorted anyway is equally
without merit. That result is the normal (and appropriate) outcome of basing
prices on marginal cost in any market with an upward sloping supply curve.
Compliance with ECPR also requires that rate differentials reflect all costs
avoided by presorting, not just the cost pools that the Postal Service has
arbitrarily chosen to model.
Careful analysis of all relevant cost pools, as
undertaken by Pitney Bowes witness Buc and MMA witness Bentley, shows that
the presort rate differentials proposed by the Postal Service are all well below
100 percent of actual cost avoidances.
-3-
II.
THE COMMISSION SHOULD RECOMMEND THE PRESORT RATE
DIFFERENTIALS PROPOSED BY THE POSTAL SERVICE FOR FIRSTCLASS MAIL.
The Commission traditionally has set presort rate differentials from the top
down. First, the Commission has constructed a hypothetical category of SinglePiece First-Class to serve as the rate benchmark for the workshared categories
of First-Class Mail.
Rates for the workshared rate categories are set by
subtracting from the benchmark rate any differences in attributable costs that the
Commission finds to be avoided by the worksharing. 1 This methodology has
traditionally led to rate differentials between the single-piece and presort
subclasses that are smaller than average inter-subclass cost differences. See
MC95-1 PRC Op. & Rec. Decis. ¶ 4221.
Both the threshold choice of a rate benchmark and the estimation of
presort-related cost avoidances have been subjects of dispute in this docket. We
discuss each area in turn. First, however, we discuss the Efficient Component
Pricing Rule (“ECPR”), which by general consensus is the appropriate
conceptual framework for resolving these issues.
A.
The Efficient Component Pricing Rule (“ECPR”) Is The Proper
Standard For Setting Presort Rate Differentials.
The Efficient Component Pricing Rule, a rule for achieving lowest
combined costs through the pricing of individual components of vertically
integrated goods or services offered by a regulated monopolist, requires that a
1
See Notice of Inquiry #3 at 2-4 (quoting R84-1 Op. & Rec. Decis. ¶ 5132 and
R90-1 Op. & Rec. Decis. ¶¶ 5066-67, 5088); Panzar Direct (PB-T-1) at 35-36.
The Commission often used a “cost difference” approach until 1995. See R87-1
Op. & Rec. Decis. ¶ 5195. Since MC95-1, however, the Commission has
adhered to a “cost avoidance” approach. See MC95-1 Op. & Rec. Decis. ¶ 4222.
-4-
vertically integrated firm offer potentially competitive components at the marginal
cost of supplying those components.
Applied to the Postal Service, ECPR
requires that the discounts offered to mailers for private sector activity that
reduces Postal Service costs should be set equal to the per unit avoided costs of
the Postal Service. Prices that satisfy this standard induce mailers to engage in
such private sector activity if (and only if) the savings to the Postal Service
exceed the added costs to society of the additional private sector activity.
Panzar Direct (PB-T-1) at 16-26; Buc Direct (PB-T-2) at 4-5; accord, PRC Op. &
Rec. Decis. MC95-1 ¶ 4256.
Setting rate differentials for worksharing that are smaller than the costs
avoided by the Postal Service violates ECPR and causes productive inefficiency,
by signaling mailers to demand, and the Postal Service to supply, a variety of
components of Single-Piece Mail service instead of substitute components that
third-party vendors, or the mailers themselves, could supply at a lower cost to
society. For example:
•
Mailers demand that the Postal Service supply mail sorting instead of
substitute presort activities that the private sector could supply at a
lower cost.
•
Mailers demand that the Postal Service provide intercity transportation
that could be bypassed at a lower cost to society through destination
entry and/or distributed printing.
•
Mailers demand that the Postal Service provide mail acceptance
services for multiple mailings—services that could be bypassed in
substantial part, at a lower net cost to society, by consolidation of
mailings into fewer, larger mailings.
-5-
•
Mailers purchase postage through sales channels and methods of
postage evidencing whose unit costs could be bypassed in substantial
part by making the investments needed to make use of lower cost
sales channels and methods of postage evidencing.
See Panzar Direct (PB-T-1) at 16-26; Buc Direct (PB-T-2) at 6-7; see also
Comments Of Association For Mail Electronic Enhancement et al. In Response
To Notice Of Inquiry No. 3 (August 17, 2006) at 15-17 (discussing restaurant
analogy); John C. Panzar, “Efficient Worksharing Discounts With Mail
Heterogeneity,” in M. Crew and P. Kleindorfer, eds., Liberalization of the Postal
and Delivery Sector (2006) at 121-134.
The Commission has repeatedly held that ECPR-compliant prices
maximize both efficiency and fairness. PRC Op. & Rec. Decis. MC95-1 (Jan. 26,
1996), ¶ 3074; PRC Op. & Rec. Decis. R2000-1, ¶ 5060; accord, Docket No.
R2006-1, Notice of Inquiry No. 3 (“NOI #3”) (issued July 26, 2006) at 2. The
Commission’s judgment is supported by the consensus of the economic
testimony before the Commission in this docket. See Panzar Dir. (PB-T-1) at 1626; 35 Tr. 11724, 11739-41 (Crowder); Sidak Direct (NNA-T-1) at 6-11. 2
Even APWU and the OCA, while adopting a more restrictive interpretation
of ECPR than the other participants, do not dispute that ECPR is the governing
economic benchmark. APWU witness Kobe agreed that “similar letters being
provided First Class service should bear the same amount of the institutional
2
Valpak witness Mitchell contends that in some circumstances appropriate rate
differentials may be larger than 100 percent of cost avoidances. See Mitchell
Direct (VP-T-1) at 118, 178; 35 Tr. 11911-12 (Mitchell).
-6-
costs of the Postal network” by setting discounts equivalent to the “clearly
capturable cost avoidance.” Kobe Direct (APWU-T-1) at 4 & n. 4 (quoting MC741 PRC Op. & Rec. Decis. at 16); id. at 5, lines 1-2 (presort discounts should be
set by “determining [the] cost savings due to worksharing activities”). See also
20 Tr. 7131 (Kobe response to NAPM/APWU-T1-1) (“I think the pass through of
100 percent of costs avoided provides the correct economic signals.”).
Similarly OCA witness Thompson acknowledged that worksharing
discounts are, “to the extent practical, set equal to the costs avoided by
worksharing.”
Thompson Direct (OCA-T-4) at 18; accord, 20 Tr. 7378
(Thompson) (“Discounts should be based on the costs avoided by the activities
of mailers that justified the creation of the discount.”).
B.
The Relevant Presort Benchmark Under ECPR Is Single-Piece
Mail With The Cost Characteristics Of Collection Mail, Not
BMM.
In this docket, the Postal Service proposes to sever the traditional linkage
between single-piece and presort First-Class rates.
Instead of determining
presort rate by subtracting “avoided costs” from a hypothetical benchmark
(traditionally “bulk metered mail” or “BMM”), the Postal Service proposes to set
rates for presort and single-piece First-Class Mail so that the two kinds of mail
will cover their attributable costs and make about the same average unit
contribution per piece. Taufique Direct (USPS-T-32) at 12-17; Bentley Direct
(MMA-T-1) at 6. This approach produces lower presort First-Class rates than
would the BMM benchmark. Thompson Direct (OCA-T-4) at 10.
-7-
The intervenor testimony on First-Class rate design is generally supportive
of the de-linking proposal. See Panzar Direct (PB-T-1); Buc Direct (PB-T-2);
Bentley Direct (MMA-T-1); Bell Reb. (NAPM-T-1); Kent Reb. (ABA-RT-1). 3
Two participants, the Office of Consumer Advocate and the American
Postal Workers Union, AFL-CIO, have submitted testimony opposing the
proposal.
OCA and APWU ask the Commission to maintain the traditional
linkage of single-piece and presort rates, and the traditional use of BMM as the
cost and rate benchmark for implementing this linkage.
Their reasoning,
although not spelled out in detail, appears to be as follows:
1.
ECPR requires that worksharing discounts be limited to the costs
avoided by the worksharing activity; otherwise, mail differing only in
its presort condition will make differing contributions to the Postal
Service’s institutional costs.
2.
In determining the costs avoided by “presort” mail, the Commission
should adhere to its previous assumption that the single-piece mail
most likely to convert to “presort” mail is bulk metered mail
(“BMM”), a kind of mail that is already relatively clean and low-cost
to process.
See Kobe Direct (APWU-T-1) at 4-8; Thompson Direct (OCA-T-4) at 10-26.
3
Indeed, one intervenor, Pitney Bowes, has asked the Commission to
recommend presort rate differentials deeper than those proposed by the Postal
Service. Buc Direct (PB-T-2) at 31-34.
-8-
NAPM and NPPC believe that much of the debate over delinking in this
docket misses the point. The fundamental issue is not whether rates for the
single-piece and presort subclasses are formally linked or not, but whether the
resulting rate differentials satisfy the Efficient Component Pricing Rule (“ECPR”).
ECPR also requires that presort cost avoidances be determined by reference to
the costs of the marginal piece of single-piece mail—i.e., mail that is currently on
the margin of conversion to presort—not the average piece of the single-piece
mail has already “converted” to presort, or an usually “clean” (low cost) subset of
that single-piece mail. Presort cost avoidances determined in compliance with
these standards are clearly larger than all of the presort rate differentials that the
Postal Service is proposing.
While the delinking proposal achieves
approximately the same rates as would the use of an appropriate rate
benchmark, the important issue is whether the resulting rates are ECPRcompliant.
1.
The relevant presort benchmark under ECPR is singlepiece mail at the margin of conversion.
The dispute in this case involves not the general desirability of ECPR, but
its specific application to First-Class Mail. The problem arises because the costs
of single-piece First-Class Mail vary widely, but many of the most important
causes of the cost variations are unrecognized in any rate element.
If single-piece mail were homogeneous, the choice of a cost benchmark
would be moot, for any benchmark would give same result. Regardless of the
starting point on the presort ladder, subtracting the cost savings produced by the
specified level of presorting should arrive at the same rate for the latter presort
category.
-9-
The choice of a rate benchmark would also be immaterial if the First-Class
rate structure fully recognized all of the various non-presort cost drivers that
cause the costs of single-piece First-Class Mail to vary.
As long as the
Commission determined presorted-related cost avoidances with a comparison
that controlled fully for all of the non-presort characteristics, the results should be
the same regardless of the presort category used as the benchmark.
Unfortunately, however, neither of these conditions holds. First-Class Mail
service is “heterogeneous”—i.e., its costs vary with multiple dimensions of quality
other than the amount of mail sorting performed by the Postal Service. This
heterogeneity produces a wide spectrum of attributable costs:
There are a myriad of reasons why the pieces of mail within [single
piece First-Class Mail] have varying costs. For example, they are
sent different distances; they are sent in different parts of the
country; they are to be delivered to rural or urban areas; they are
addressed in different ways; the paper used is different; the mailpiece is shaped differently; the list goes on and on.
MC95-1 Op. & Rec. Decis. ¶ 3064; accord, Notice of Inquiry No. 3 (July 26, 2006)
(“NOI #3”) at 3 (quoting R84-1 Op. & Rec. Decis. (Sept. 7, 1984) ¶¶ 5093 and
5129); Panzar Direct (PB-T-1) at 35-36; response of APWU witness Kobe to
NAPM/APWU-T1-4 (20 Tr. 7135).
Many of these non-presort quality attributes are recognized only partially,
if at all, in individual rate elements. For example:
•
First-Class rates include neither destination-entry discounts nor
distance based rate zones, and thus do not recognize the cost effects
of the distance between the entry point and the addressee.
- 10 -
•
Until now, the First-Class rate structure has offered little recognition of
the cost effects of the shape of the mailpiece (letter, flat, or parcel).
Although the USPS is proposing in this case to give increased rate
recognition to shape as a cost driver, the recognition is far from
complete; the proposed passthroughs of shape-related costs are
substantially below 100 percent.
•
The unit cost of First-Class Mail is also affected by the number of
pieces in a mailing, and the total volume of all mail pieces entered by a
given mailer in a year.
Greater mail volumes reduce the unit
transaction costs of mail acceptance and enforcement of mail design
and address quality requirements.
•
The First-Class rate structure recognizes none of the unit cost
differences caused by the sales channel (e.g., retail window or CAPS
account) or postage evidencing methods (postage stamps vs. meters
vs. permit indicia) used by the mailer.
•
Mail addressing requirements allow wide variations in the correctness,
completeness, and legibility of addresses, particularly for Single-Piece
Mail.
The lack of rate elements for many individual cost drivers might be of little
concern if the overall cost effects of these quality characteristics were distributed
evenly between the Single-Piece and Presort categories of First-Class Mail. In
fact, the non-presort characteristics are not evenly distributed. Compared with
Single-Piece Mail, the average piece of First-Class Presort Mail is less costly in
multiple respects, and has significantly lower unit costs than Single-Piece Mail
even after presort cost avoidances are netted out:
- 11 -
Presort mail tends to be “cleaner.” In comparison to the nonpresort
benchmark, presort mail is normally printed (or typewritten), and
more uniform than nonpresort mail. These factors, and possibly
others, which are unrelated to mailer preparation, may contribute to
the magnitude of the cost difference between presorted and
nonpresorted mail.
R90-1 Op. & Rec. Decis. ¶ 5067.
These circumstances require a more robust and precise specification of
ECPR. Because the costs of single-piece mail are heterogeneous, maximizing
incentives for productive efficiency requires the Commission to set presort rate
differentials equal to the unit cost differences between Presort Mail and the
marginal piece of Single-Piece mail. Panzar Direct (PB-T-1) at 36-39. Using the
typical, median or average piece of Single-Piece mail as the benchmark
produces incorrect results. Id. And using a low-cost subset of Single-Piece mail
produces results that depart even further from ECPR. Id. at 37-39 (responding to
Cohen et al.).
2.
The marginal piece of single-piece mail has the cost
characteristics of collection mail, not BMM.
The foregoing analysis should make clear that bulk metered mail (“BMM”)
can no longer be considered an appropriate rate benchmark. BMM is a fictional
category of low cost single-piece First-Class Mail loosely defined as bulk mailings
of letters that are “machinable, homogeneous, non-barcode pieces with machine
printed addresses that are properly faced and entered in trays.” Kobe Direct
(APWU-T-1) at 15 (quoting R2005-1 Tr. 4/952 (Abdirahman)).
BMM is best
described as a result—lower costs—in search of supporting facts. It has no
- 12 -
precise definition, 4 and even its proponents are unsure of how much mail with
the characteristics of BMM actually exists. 5
Advocates of the BMM paradigm have defended its use as a rate
benchmark on the theory that BMM best models the characteristics of the
average single-piece mail that is assumed to have migrated to the Presort
category. 6 As explained above, however, the appropriate cost benchmark for
mail as heterogeneous as single-piece First-Class Mail is not the average piece
of single-piece mail submitted for presorting, or the average piece of presorted
mail after presorting, but the marginal piece of single-piece mail.
The record provides no evidence that the marginal single-piece mailpiece
is remotely akin to BMM—or, indeed, that mail converted from BMM represents
more than a tiny share of the existing presort mailstream. To the contrary, the
most credible evidence in the record on this issue, from NAPM witness Elizabeth
4
See 20 Tr. 7078 (response to APWU witness Kobe to MMA/APWU-T1-3(a))
(defining BMM as mail tending to be “at the cleaner end of the continuum,” and
defining “clean mail” as “mail which, for a variety of reasons, is cheaper than
average to process . . . there is no precise definition of this term”).
5
“To my knowledge the Postal Service does not provide volumes of BMM letters
nor am I aware of any source of data that provides the conversion information
that you seek.” 20 Tr. 7093 (Kobe response to MMA/APWU-T1-11); id. at 7093
(there are no data showing how much BMM is expected to convert to presort
First-Class Mail in TY 2008); id. at 7199 (“I haven’t personally seen it [BMM], but
I know it exists.”).
6
Kobe Direct (APWU-T-1) at 15, lines 2-5. Ms. Kobe also suggests that BMM is
the most appropriate benchmark because it most closely resembles the
characteristics of presort mail after it has been processed. Kobe Direct (APWUT-1) at 6, lines 13-15. This is clearly illogical. To estimate the costs that the
Postal Service avoids by presorting, one must begin with the costs of a piece that
has not been presorted. Using the characteristics of mail that already has been
presorted as the starting point would define the entire cost savings out of
existence.
- 13 -
Bell, (NAPM-RT-1), makes clear that the marginal piece of Single-Piece mail
currently sought and obtained by presort bureaus has costs characteristics akin
to collection mail.
Ms. Bell has owned and operated a presort bureau in Florida for the past
20 years.
Bell Reb. (NAPM-RT-1) at 3 (38 Tr. 12946).
Before starting her
company, she worked for several years at the Postal Service, performing a
variety of tasks including collection of mail from collection boxes. 38 Tr. 12946,
12989. Her career has made her familiar with the characteristics of presorted
First-Class Mail before it is presorted. 38 Tr. 12946. In her experience, the FirstClass Mail received by her company for presorting from customers who have not
previously used a presort bureau—i.e., from new customers—typically has
physical characteristics similar to collection mail. 38 Tr. 12946. She has not
seen any BMM coming in from a new customer. 38 Tr. 12999, line 25, to 13000,
line 1.
There a number of reasons why the single-piece mail that converts to
presort is far less clean than BMM:
•
Most customers do not know what sizes of envelopes are acceptable
for automation mail. 38 Tr. 12947. Even business mailers often use
self-mailers and other alternatives to standard business envelopes. 38
Tr. 12988-89.
•
Most customers do not know what type faces and color of envelope
stock can be read by optical character readers. 38 Tr. 12947.
•
Many pieces tendered to presort bureaus have handwritten addresses,
not pristine typed or computer-generated addresses like those on
- 14 -
BMM. 38 Tr. 12947; id. at 12990-91 (noting that hospitals tend to
generate significant volumes of hand-addressed mail from doctors).
•
Most mailers do not know what a barcode clear zone is, and would
have no reason to provide one. 38 Tr. 12947.
•
Most do not know what Move Update is. Without the involvement of a
presort bureau, these mailers would have no reason to comply with
Move Update requirements, especially for First-Class Mail, which the
USPS forwards free of additional charge. 38 Tr. 12947.
•
Most mailers do not understand “loop mail” (mail that is designed in a
way that causes the optical character reader—whether owned by the
Postal Service or a presort bureau—to read the return address and
send the mailpiece back to the sender) or how to avoid it. 38 Tr.
12947.
•
A significant number of mailpieces deposited by employees in
workplace collection boxes arrive at presort bureaus in envelopes that
are unsealed or stuck together. Id. at 12947, 12992. Mailpieces stuck
together must be separated by hand before they can be fed into mail
sorting equipment. Id. at 12992.
•
Most new mailers do not know how to properly use tabs and wafer
seals. 38 Tr. 12947.
Without the intervention of presort bureaus, the mail entered by these customers
would cost the Postal Service far more to process than BMM-like mail.
12950-52.
- 15 -
38 Tr.
Moreover, it would be difficult for most mailers to prepare BMM-like mail
even if they wished to do so. Most individual mailers lack access to a timely and
adequate supply of trays. 7 Many customers of presort bureaus generate too little
mail to justify using trays. The single-piece rate structure provides no incentive
for using trays – or for entering mail in containers at all. Nor does the singlepiece rate structure provide any incentive to bundle or sequence pieces, or orient
them consistently. To the contrary, it is easier to enter large quantities of singlepiece mail with “every other handful inverted, which increases the ease of putting
mail into a box without flared ends.”
38 Tr. 12948-49, 12996, lines 13-14 (Bell).
Furthermore, mailers that prepared large quantities of BMM would have
difficulty entering it in the Postal Service network. Because BMM is single-piece
mail, bulk mail entry units would not accept it, and it would need to be entered at
a retail mail entry facility, which typically would not be organized to accept bulk
mailings efficiently. The mailer “would need to go probably to the dock, which
with all the gates and everything you have to have an appointment to get in.
Otherwise you take it up to the front counter.” 38 Tr. 13001 (Bell).
7
During cross-examination of Ms. Bell, counsel for APWU suggested that the
flow of trays would be bidirectional, on the theory that large senders of FirstClass Mail are “highly likely” to be “receiving a substantial volume of mail as
well,” and thus are likely to receive a supply of trays sufficient for their outgoing
mail. 38 Tr. 12987. In fact, facilities that generate large volumes of outgoing
mail often do not receive equal quantities of incoming mail. Ms. Bell’s two largest
customers, for example, are a VA hospital center that sends out large volumes of
appointment notices, and a company mail processing center whose
corresponding payment center is located elsewhere. Both of these companies
send far more mail than they receive. 38 Tr. 13002 (Bell). “If they were getting
the trays in that they needed, I would never have to deliver trays to my
customers, and I deliver trays daily to some of my customers. I deliver a supply
weekly to every customer.” Id. (emphasis added).
- 16 -
What is true of the baseload volume of mail received by presort bureaus is
even more true of their marginal piece of mail. Volumes of single-piece mail from
new customers, when they appear at all, do not appear in trays of clean letters.
This is true not only in southwest Florida, but elsewhere in the United States. Id.
at 7 (38 Tr. 12950). Indeed, on an average day, Ms. Bell’s company returns to its
customers 200 letters that are so poorly prepared that they would be
undeliverable as addressed. Id. at 9 (38 Tr. 12952).
Ms. Bell’s testimony on the physical and cost characteristics of the single
piece mail that currently converts to presort is supported by the testimony of
USPS witness Taufique. In response to cross-examination by APWU counsel,
Mr. Taufique stated:
What we get now, and we have seen a lot of evidence of that, is
that large presort bureaus are picking up office mail, which is not
BMM mail, which is not trayed, which is not faced. . . . Large
presort bureaus are picking up office mail and converting that into
presort mail, which is not the same thing as the BMM mail which is
metered, which is faced and which is trayed.
16 Tr. 4938 (Taufique). Mr. Taufique’s testimony on this point, like Ms. Bell’s,
was based on first-hand observation of the presort mailstream:
As a pricing witness, I keep up with the industry. I try to visit both
our plants and the plants of the people who prepare mail for us, sir,
and talk to the costing people who see the mail coming in so it is
not [just] a feeling.
It is based on observations from the folks who work in the field,
folks who process the mail and our own personal experiences from
persons in the plant, the three or four bureaus’ activities, actually
the costing people.
I have actually visited more than one facility where we’ve seen that
the mail that is being converted into presort is not the clean mail of
the past. It is a heterogeneous stream of mail that is being
- 17 -
converted, and that is why we need to look at the benchmark issue
in a different light.
We are not abandoning the [efficient] component pricing in this
regard. That is based on the observations of the Postal Service
that this is happening, and we need to change the benchmark in
relation to the new industry that we’re looking at right now.
16 Tr. 4939 (Taufique).
APWU and OCA, the two defenders of the BMM benchmark in this case,
were unable to offer any credible evidence that BMM represents a significant
fraction of the mail currently handled by presort bureaus, let alone of the marginal
mail that now converts to presort. APWU witness Kobe acknowledged that she
was “unaware of any studies that provide details on which pieces [BMM or other]
shift from Single Piece mail to Presort mail”; unable to provide any studies
showing that the “mail that is converting from First Class single piece to presort is
not equivalent to the average collection mail” from households, nonprofit
organizations and small businesses; unable to state what percentage of the
growth in presort mail volume is due to conversion from single-piece mail; and
unaware of any “statistics kept on this topic” of “how much First Class single
piece mail . . . still ‘converts’ to presort mail”); unaware of any data provided by
the Postal Service showing “volumes of BMM letters”; and unaware of any data
showing how much BMM is expected to convert to presort First-Class Mail in
Test Year 2008). 20 Tr. 7074, 7084, 7093, 7097 (Kobe).
Witness Thompson asserts that “automation compatible letters, hand
addressed letters, non-machinable letters” and collection mail from households
and small businesses is unlikely to be converted to presort because presort
bureaus are unlikely to be interested in working this kind of high-cost mail.
Thompson Direct (OCA-T-4) at 11, lines 5-8; id. at 12, lines 5-8. In fact, presort
- 18 -
bureaus “would be thrilled to get collection box mail to process. It would not
present us with any challenges we do not face now on a daily basis.” Bell Reb.
(NAPM-RT-1) at 3 (38 Tr. 12946).
In rebuttal testimony, USPS witness Abdirahman claimed to have seen
“hundreds of trays of” BMM at the Southern Maryland processing plant in
September 2006. Abdirahman Reb. (USPS-RT-7) at 5, 35 Tr. 11952. He was
unable to state how the trays entered the plant, or whether they were entered by
one mailer or many. Id. at 12008, 12012. He also claimed to have seen BMM
before at other plants. Id. at 12012.
In fact, what Mr. Abdirahman saw on these occasions almost certainly was
mail processed by presort bureaus, but entered at undiscounted single-piece
rates because the mail already carried full rated postage 8 or could not be entered
at discounted rates for other reasons (e.g., its address could not be given a
proper 11-digit delivery point POSTNET barcode). DMM 233.5.1. The Postal
Service requires that presort bureaus perform extensive processing on all mail
they enter—even mail that does not qualify for presort discounts. 38 Tr. 1294950 (Bell). The uncompensated work that the Postal Service compels presort
bureaus to perform on this “full-paid mail” includes facing, traying and separation
of stamped envelopes from business reply envelopes (“BREs”), stamped mail
and BRE from metered mail, Express and Priority Mail from First-Class Mail, flats
from letters, and packages from other shapes. Bell Reb. (NAPM-RT-1) at 6-7 (38
8
Mail bearing indicia of payment at the full single-piece rate is ineligible for
value-added refunds to the presort bureau entering it. DMM 902.9.4.12(c);
accord, 38 Tr. 12994, lines 15-21 (Bell).
- 19 -
Tr. 12949-50). 9 Although the result may look like BMM to the uninitiated, it is not.
As Ms. Bell testified:
Q:
In your professional opinion, have you seen mail inside the
Postal Service that looks like bulk metered mail?
A.
I’ve seen processed mail like [that] coming from presorts.
I’ve never seen large volumes of anything other than going
on the facer cancelers from the collection boxes.
Q.
In your professional opinion, what is the most likely source of
mail that looks like bulk metered mail inside the Postal
Service?
A.
I would say presort service bureaus.
38 Tr. 13001 (Bell). Needless to say, mail that has already received extensive
processing from a presort bureau is not a valid benchmark of the cleanliness of
single-piece mail before conversion to presort. 10
Finally, bulk mailings of clean First-Class Mail that still remain in the
single piece mailstream are likely to be resistant to conversion.
Government-
originated mail is a good example, as Mr. Abdirahman acknowledged during
cross-examination by MMA counsel:
9
Ms. Bell’s company alone enters about 8,000 pieces of such full-paid mail every
day—an amount equivalent to 40 trays of mail, or a little more than five percent
of the total mail volume entered by her company. Bell Reb. (NAPM-RT-1) at 6-7
(38 Tr. 12949-50).
10
It is also possible that some mail observed in Postal Service facilities as having
the appearance of BMM may in fact be collection mail that was trayed and faced
by Postal Service employees at an associated facility before transport to the
observation point. See, e.g., R2000-1 Tr. 45/19699 (Miller) (some of the mail
volume characterized as BMM “is trayed by the Postal Service itself rather than
mailers”). Needless to say, mail that has already received processing by the
Postal Service is also not a legitimate benchmark of the cost characteristics of
single-piece mail before conversion to presort.
- 20 -
Q.
Do you recognize that there’s certain mail out there, which
may look like BMM, smell like BMM, walk like BMM, but is
never going to convert to pre-sorted mail?
A.
Give me an example.
Q.
Well, a government agency, for example, that might have
considerations other than just controlling costs, and they
might not want to be involved in all it takes to workshare mail
and to live up to the requirements that the Postal Service
has and to meet the deadlines that the Postal Service has.
A.
It could. In a daily business, people mail things in half tray,
full tray, depending on the business.
Q.
Right. And so this could have been going on for years and
years and may never convert to presorted mail, right?
A.
Yeah.
35 Tr. 12013 (Abdirahman).
3.
The BMM benchmark does not achieve a uniform perpiece contribution to institutional costs.
APWU witness Kobe contends that because First-Class mail is “very
heterogeneous,” the BMM benchmark is “the only way to make certain that two
mailers with identical pieces of mail are paying the same contribution to overhead
costs, irrespective of whether they workshare.” Kobe Direct (APWU-T-1) at 15;
id. at 7, lines 2-5. But the BMM benchmark does not achieve uniformity of cost
contribution. Under Ms. Kobe’s proposed rate design, for example, for example:
•
The mailer of a “clean” Single Piece letter would pay a larger
contribution to overhead costs than the mailer of a “dirty” Single Piece
letter.
- 21 -
•
The average piece of presort First-Class Mail would pay a larger
contribution to overhead costs than would the average piece of singlepiece First-Class Mail. Bell Reb. (NAPM-RT-1) at 10 (38 Tr. 12953).
•
A First-Class letter entered at a destination facility would pay a larger
contribution to overhead costs than an otherwise identical letter carried
a longer distance. 20 Tr. 7194-96 (Kobe).
•
A letter with a computer-generated address typically costs the Postal
Service less to process than an otherwise similar piece “that is
handwritten and has to be remote barcoded by a manual operator.” 20
Tr. 7196, lines 14-18 (Kobe cross-examination). Presort First-Class
mail must have a machinable address; single-piece mail need not, and
often does not. Id. at 7196, line 19, to 7197, line 4. APWU is not
proposing to offer a discount for putting a machinable address on a
single piece letter. Id. at 7197, line 5-8. Hence, under APWU’s rate
design, “first class letters with handwritten addresses are unlikely to be
making the same contribution to the Postal Service institutional costs
as otherwise identical letters with typed or printed addresses . . .
That’s part of the averaging of the first class rate.” Id. at 7197, lines 915.
•
APWU’s proposed rate design would not offer First-Class mailers
discounts for buying postage through a less costly channel than a retail
counter, or entering mail in a Postal Service facility rather than a street
collection box. 20 Tr. 7197, line 16, to 7198, line 11 (Kobe crossexamination). These mailer activities save the Postal Service money.
Id. at 7197, lines 16-20. Because the APWU rate design does not offer
discounts to single-piece mailers for engaging in these activities, the
- 22 -
result could be “differences in institutional cost contributions for mail
that varied only in that attribute.” Id. at 7197, line 24, to 7198, line 2.
The BMM benchmark is not alone in failing to achieve uniformity of
contribution; nonuniformity of cost contribution is inherent in the highly averaged
nature of the First-Class rate structure. Until the First-Class rate structure is fully
deaveraged for each of the numerous mail characteristics that make First-Class
costs so heterogeneous, no presort benchmark will achieve uniformity of
contribution across the rate class. Taufique Reb. (USPS-RT-18) at 5-6, 38 Tr.
13346-47; Panzar Direct (PB-T-1) at 28-39. As Ms. Kobe conceded in response
to an interrogatory, “whenever there is an averaging process, there will be some
mail within that category that pays more towards institutional costs than other
mail.” 20 Tr. 7076 (Kobe response to MMA/APWU-T-2(a) & (b)); accord, 20 Tr.
7079 (Kobe response to MMA/APWU-T1-3(d)). 11
Rather than pursuing the chimera of uniformity of contribution, the
Commission should focus on the more important and achievable goal embodied
in ECPR: setting rates that promote the lowest combined Postal Service/private
11
Needless to say, complete deaveraging would be neither feasible nor
desirable. Averaging is “an integral part of postal ratemaking.” MC95-1 Op. &
Rec. Decis. ¶ 3063. The “essence of a classification” is a “grouping of mail for
which a certain rate is charged.” National Easter Seal Society v. USPS, 656 F.2d
754, 762-63 (D.C. Cir. 1981) (emphasis added). A “separate rate for every group
of mailers with special cost savings, no matter how small the group, would
produce a hopelessly complicated rate schedule.” Id. (quoted in United Parcel
Service, Inc. v. USPS, 184 F.3d 827, 845 (D.C. Cir. 1999); accord, Mail Order
Ass’n of America v. USPS, 2 F.3d 408, 426 (D.C. Cir. 1993). Accordingly, the
Commission must balance the potential efficiency gains of greater rate
deaveraging against the additional transaction costs. See MC95-1 Tr. 33/15011,
15023 (Schmalensee).
- 23 -
sector costs for First-Class Mail. That goal, as Professor Panzar has explained,
requires adoption of a presort benchmark based on the marginal piece of singlepiece mail. See Section II.B.1, supra.
4.
Basing presort discounts on the marginal piece of mail
does not give existing presort mail an inappropriate
reward.
Ms. Kobe also contends that replacing BMM with the marginal piece of
mail as the presort rate benchmark, by deepening presort discounts, would give
an reward or “windfall” to the “least expensive mail already being workshared”—
an outcome that APWU regards as undesirable. 20 Tr. 7062-63, 7193. That low
cost mail would receive additional discounts is irrelevant to the merits of the
proposal. This result is the normal (and appropriate) outcome in any market with
an upward sloping supply curve when the price for a good or service is set equal
to the cost of the marginal unit supplied. Under these circumstances, suppliers
of the “inframarginal” (i.e., lower-cost) outputs normally receive “rewards” by
definition. That is how ECPR, and efficient competitive markets, treat low cost
suppliers.
Panzar Direct (PB-T-1) at 17-19, 22 and 35.
Even Ms. Kobe
conceded this on cross-examination. See 20 Tr. 7205-10 (APWU witness Kobe).
5.
Delinking of presort and single-piece rates is a
reasonable proxy for an ECPR-compliant presort
benchmark.
As previously noted, the Postal Service has proposed to de-link rates for
presort First-Class Mail from single-piece Mail: i.e., to set rates for presort FirstClass Mail so that each category pays its attributable costs and makes an
approximately equal contribution per piece. This is a practical and reasonable
proxy for linking presort rates to a single-piece benchmark that reflects the
- 24 -
characteristics of the marginal single-piece mail that converts to presorting today.
Taufique Reb. (USPS-RT-18) at 7-9, 38 Tr. 13348-50.
The Commission suggested in NOI #3 that de-linking “implicitly treats
Single-Piece Mail and workshared mail as separate subclasses.” Cf. NOI #3 at
2.
This is incorrect.
The creation of separate subclasses would entail the
prescription of separate coverage ratios or contributions to institutional cost for
each grouping of mail, based on a separate analysis of the demand for each
grouping. See R80-1 Op. & Rec. Decis. ¶ 692, aff’d, Newsweek, Inc. v. USPS,
663 F.2d 1186, 1210 (2nd Cir. 1981); R84-1 Op. & Rec. Decis. ¶¶ 5090-5106;
R87-1 Op. & Rec. Decis. ¶ 5144; MC95-1 Op. & Rec. Decis.¶¶ 2073, 3017-18,
3022-25, 5030-34. The Postal Service has not proposed this more far-reaching
relief here.
Absent these additional steps, the modified cost avoidance
methodology embodied in the de-linking proposal does not rise to the creation of
a separate subclass. As the Commission held in R87-1:
From the outset, we reject the Postal Service’s implication that a
choice of a cost methodology other than its cost avoidance
methodology to design the First-Class Mail 5-digit presort discount
is tantamount to a designation of the presort rate as a subclass or
“de facto subclass.” From our reading of the statutory criteria and
history and of the presort discount, nothing suggests that a
restrictive view such as that advanced by the Service is mandated
or desirable. On the contrary, these sources and the record in this
proceeding support the view that there should be flexibility in the
design of any given discount rate and the development of the
discount should not be bound by one exclusive cost methodology
which must be used throughout the history of the discount.
R87-1 Op. & Rec. Decis. ¶ 5188 (emphasis added). De-linking likewise avoids
the need to determine the “market boundaries” of Presort and Single-Piece Mail,
or to worry about whether the two categories of service serve, or do not serve,
“overlapping markets.” Cf. NOI #3 at 5; id. at 6 (question 1).
- 25 -
C.
EPCR Requires Analysis Of All Presort-Related Cost Pools.
A second precondition for efficient component pricing to basing rate
differentials on estimates of presort cost avoidances that reasonably capture all
costs actually avoided by the worksharing at issue. Basic rate differentials based
on cost studies that systematically understate presort-related cost avoidances, or
that arbitrarily assume without analysis that most cost pools do not vary with the
level of presorting, do not produce ECPR-compliant presort rates. Moreover,
cost avoidance estimates are understated will cause presort discounts to appear
too large when they are actually correct, or even too small. See Buc Direct (PBT-2) at 6-7.
Unfortunately, the Postal Service has repeatedly underestimated presort
cost avoidances in the past, and has done so again in this case. The problem
stems in significant part, as in past cases, from the Postal Service’s failure to
model all cost pools affected by the degree of presortation. For the cost pools
left unmodeled, the Postal Service has simply assumed by default that presorting
avoids no costs at all.
This is a clear source of downward bias.
As the
Commission found in MC95-1:
If costs avoided by a worksharing operation are difficult to isolate,
they tend to be omitted by engineering models. Therefore, cost
differentials based on engineering models tend to be
underinclusive.
MC95-1 Op. & Rec. Decis. ¶ 4220; Bentley Direct (MMA-T-1) at 6-7.
The Postal Service’s presort cost studies in this case continue to suffer
from this deficiency—and, in many respects, to a larger degree than in the past.
In some respects, the Postal Service’s presort cost models in this case improve
on previous versions.
Buc Direct (PB-T-2) at 10-12.
- 26 -
For many cost pools,
however, the Postal Service’s analysis of the presort relatedness of individual
cost pools continues to be little more than guesswork. Buc Direct (PB-T-2) at 1329.
Written discovery and oral cross-examination of the Postal Service
witness Abdirahman have confirmed that the Postal Service did not model all
costs of mail processing operations.
Rather than perform a thorough and
rigorous analysis of the effect of sorting on costs, the Postal Service modeled
only 10 out of 53 cost pools, and simply assumed that the 43 remaining cost
pools are unaffected by the degree of worksharing. See 4 Tr. 519 (Abdirahman
response to ABA-NAPM/USPS-T-22-4b); R2006-1, Abdirahman Direct (USPS-T22), at 7, lines 19-21, and 8, lines 1-2. The cost pools labeled as “proportional”
were merely those containing costs for tasks that were actually modeled and the
cost pools labeled as “fixed” were merely those containing costs that were not
modeled. See e.g., 4 Tr. 609 (Abdirahman response to PB/USPS-T22-4); id. at
528 (Abdirahman response to MMA/USPS-T22-1); id. at 618-20, 660
(Abdirahman). See generally Buc Direct (PB-T-2) at 7-9. Moreover, the Postal
Service completely excluded any analysis of cost differences at delivery units.
Buc Direct (PB-T-2) at 12-13.
APWU witness Kobe and OCA witness Thompson adopted with only
limited changes the cost pool classifications developed by the Postal Service.
Kobe Direct (APWU-T-1) at 16-21; 20 Tr. 7073, 7115, 7118-22, 7136, 7142
(Kobe); Thompson Direct (OCA-T-4) at 18-22; 20 Tr. 7373 (Thompson response
to MMA/OCA-T4-2-8(b); 20 Tr. 7375 (Thompson) (“I did not analyze cost pools”);
20 Tr. 7424, lines 4-144 (Thompson). Accordingly, their presort cost models also
underestimate presort cost avoidances.
- 27 -
By contrast, the presort cost avoidance estimates of Pitney Bowes witness
Buc and MMA witness Bentley are well supported. Mr. Buc, performing a
comprehensive new analysis of the Postal Service’s cost pool classifications, has
shown that many of the pools left unmodeled (and therefore categorized as
“fixed”) by the Postal Service in fact contain costs that vary with presort level and
thus should be classified as “proportional.” Buc Direct (PB-T-2) at 14-30; see
also Library Reference PB-LR-L-1, Tab 5 (detailed analysis).
The separate
analysis of MMA witness Bentley has arrived at similar results. Bentley Direct
(MMA-T-1) at 12-19.
See also Panzar Direct (PB-T-1) at 40-44 (providing
support for Buc analysis of piggyback and overhead costs).
D.
The Presort Rate Differentials Proposed By The Postal Service
Would Pass Through Less Than 100 Percent Of Avoided
Costs.
A comparison of the presort cost avoidances developed by Pitney Bowes
witness Buc and MMA witness Bentley with the Postal Service’s proposed
presort discounts confirms the reasonableness of the discounts.
For each
presort tier, the rate differentials proposed by the Postal Service are smaller than
the incremental cost savings.
Indeed, for two of the presort increments—
MAADC to AADC and AADC to 3-digit—the incremental passthroughs equal less
than 60 percent of the incremental cost savings:
- 28 -
USPS Proposed Rates for Automation Letter Mail
Rate
Category
USPS
Proposed
Rate (cents)
MAADC
34.6
AADC
33.5
56%
3-Digit
33.1
57%
5-Digit
31.2
81%
Proposed
Passthrough
Buc Direct (PB-T-2) at 31 (Table 9). 12 MMA witness Bentley has obtained similar
results. Bentley Direct (MMA-T-1) at 10-12. Under either set of cost avoidances,
the presort rate differentials proposed by the Postal Service are quite
conservative.
III.
THE COMMISSION SHOULD RECOMMEND OTHER CHANGES IN
FIRST-CLASS RATE DESIGN PROPOSED BY THE POSTAL SERVICE
A.
Greater Recognition Of Shape-Related Cost Differences
In this case, the Postal Service has proposed greater recognition of the
costs differences caused by the shape of First-Class mailpieces. See Taufique
Direct (USPS-T-32) at 17-20. NAPM and NPPC support this initiative.
Economic efficiency requires that all significant cost-causing differences
between the attributes of two categories of mail be recognized in differentials
between the rates for those two categories. Shape has become a very important
driver of mail processing costs under automation. Thus, the Postal Service’s
proposal to recognize at least a portion of shape-related cost differences in the
12
Indeed, that those cost avoidances would reasonably support even greater
rate differentials, as Pitney Bowes proposes. Buc Direct (PB-T-2) at 31-34.
- 29 -
First-Class rate structure is an important step toward greater economic efficiency
and sending better price signals to mailers. Greater recognition of shape will
also enable the Postal Service to compete more effectively against electronic
competition by reducing First-Class rates for heavier-weight pieces, as the Postal
Service has proposed in this case. Bentley Direct (MMA-T-1) at 22.
B.
Reduced Extra-Ounce Rate.
The Postal Service has also proposed in this case to deemphasize weight
as a rate determinant by reducing the extra-ounce rate from 24 cents to 20 cents.
Taufique (USPS-T-32) at 4. NAPM and NPPC support this initiative as well.
The Postal Service proposal is consistent with the principle that, when a
mail characteristic has only a limited effect on cost, the rate element
incorporating that mail characteristic should not exceed the cost differential. No
other ratemaking considerations warrant making an exception to this principle
unless there is some other reason to do so—such as the recognition of demand
characteristics, or other considerations specified by 39 U.S.C. § 3622(b). No
such reasons exist in this case. Additional ounces in letters cost the Postal
Service very little, especially in an automated mail processing environment.
Moreover, a reduced additional-ounce rate for First Class bulk letters
would enable First-Class Mail to compete more effectively against the Internet,
and to attract volume from Standard Mail, by encouraging mailers to insert more
solicitation matter along with statements entered at First-Class rates.
By
reducing the cost of such a dual-purpose mailing to the mailer, the reduced
additional-ounce rate will enable the Postal Service to compete more effectively
- 30 -
against the Internet for bill presentment, financial statements and similar
communications.
Bentley Direct (MMA-T-1) at 24-26; Resch (DFS/MSI-T-1).
CONCLUSION
For the foregoing reasons, NAPM and NPPC respectfully request that the
Commission recommend the Presort First-Class rate design proposed by the
Postal Service.
Respectfully submitted,
David M. Levy
Richard E. Young
SIDLEY AUSTIN LLP
1501 K Street, N.W.
Washington, DC 20005-1401
(202) 736-8000
[email protected]
Counsel for National Association of
Presort Mailers and National Postal
Policy Council
December 21, 2006
- 31 -