oca-reply-brief.pdf

UNITED STATES OF AMERICA
Before The
POSTAL RATE COMMISSION
WASHINGTON, D.C. 20268-0001
SEP22
Docket No. R2000-1
Postal Rate and Fee Changes
REPLY BRIEF
OF
THE OFFICE OF THE CONSUMER ADVOCATE
TED P. GERARDEN
DIRECTOR
EMMETT RAND COSTICH
SHELLEY S. DREIFUSS
KENNETH E. RICHARDSON
ATTORNEYS
September
22,200O
TABLE OF CONTENTS
TABLE OF CONTENTS
...................................................................................................
TABLE OF AUTHORITIES..
............................................................................................
i
v
SCOPE OF THE OCA REPLY BRIEF.. ...........................................................................
1
EXECUTIVE
SUMMARY .................................................................................................
2
ARGUMENT..
.................................................................................................................
.5
I.
II.
THE
COMMISSION
SHOULD
ADOPT
THE
OCA’S
CONTINGENCY
PROPOSAL. ...................................................................
5
A.
The Postal
Service’s
Attempts
To Minimize
the
Magnitude of the Increase Are Unpersuasive _._.__._._.__,___._..............
5
9.
Questions Raised by the Postal Service about its Equity
Position Are Specious _._.__._.._..__._.._............................................... 10
C.
The Record Is Devoid of Substantial Evidence in Support
of the 2.5 Percent Contingency Proposed by the Postal
Service
11
D.
Periodicals Should Not Be Relieved of the Contingency
Provision or Other Incurred Costs ___..__.__....................................... 13
E.
The Commission
Is Authorized
to
“Reasonable Provision for Contingencies”
Determine
the
_._.._._.__._,__._,_................
17
MAIL PROCESSING VOLUME VARIABILITY SHOULD BE 100
PERCENT .,,..,.....,..,.,.._.............,.,..,.,....,..................................................
A.
The Postal
Service’s
Initial Brief Contains
Two
Fundamental Flaws .._._,..,.,..,.,,......................................................
23
The Postal Service’s
initial brief erroneously
assumes
its model
accurately
reflects
mail
processing operations .,...,,.__..___._........................................
24
The Postal Service must demonstrate more than a
probability that variability is less than 100 percent
25
The Postal Service Has Not Demonstrated
that Mail
Processing Costs Are “Unambiguously”
Less Than 100
Percent ..,.,.,.,......,.,._...,.,....,..,........................................................
29
1.
2.
9.
23
_ ii _
Docket No. R2000-1
1.
2.
3.
C.
2.
F.
Ill.
Witness Degen’s explanation
of how mail is
processed does not prove 100 percent volume
variability
35
Volume variability
is not unambiguously
100
percent volume variable when witness Neels’
presentation is recognized
36
Is Incorrect
37
The appropriate manner to scrub the data remains
unresolved .___._____................................................................
38
A short-run analysis is irrelevant
38
Witness Bozzo’s Mail Processing Model Is Not Based on
the Appropriate Economic Theory
1.
E.
32
Witness Bozzo’s Analysis
1.
D.
The Postal Service has not demonstrated
that
setup and takedown
times will not increase
proportionately over the longer run
Witness
variables
Bozzo
has
not
selected
appropriate
40
2.
Network and location-related
factors do change
with volume, but that fact is not reflected in the
assumptions underlying the Postal Service model ___._........41
3.
In maintaining that TPF is a proper measure of
mail processing
volumes,
the Postal Service
ignores that TPF is not an accurate measure of
the volume of mail
42
The Cross Sectional “Between” Model Is the Only Model
that Considers Differences Between Facilities Based on
Size and Volumes
43
A Working Group To Address Modeling, Economic and
Estimation Problems Is Recommended ____..__...___..__._....................
44
THE COMMISSION
SHOULD
APPLY ITS ESTABLISHED
ATTRIBUTION
METHODS TO THE ES-BASED ESTIMATE OF
ACCRUED LOAD TIME COST ._._............................................................
A.
40
44
The Commission Should Accept the Postal Service’s ESBased City Carrier Street Time Proportions
.. .. . . .. . . . . . . .. . . . . 45
- 111
-
Docket No. R2000-1
IV.
V.
9.
The Commission
Based Elasticities
C.
The
Commission
Coverage-Related
Stop Ratios
Should
Continue
To
Load Time Using Single
FIRST-CLASS
9.
D.
E.
F.
SHOULD
BE SET
48
RATE ISSUES .................................................................
52
The 33-Cent Single-Piece
First-Class Rate Should Be
Retained.. .....................................................................................
.52
CEM Should Be Recommended..
.57
.................................................
1.
The classification
2.
The Postal Service’s reliance on the Ellard survey
is still misplaced.. ...............................................................
issue has long been decided .................. 57
.59
3.
Education
4.
The confusion
5.
Enforcement
6.
Administrative
costs are exaggerated
and based
on faulty reasoning .............................................................
61
The CEM proposal
is the most researched
proposal ..............................................................................
62
7.
C.
Attribute
Subclass
47
COST COVERAGES
FOR SUBCLASSES
AT A MINIMUM OF 105 PERCENT
A.
VI.
Should Continue To Employ LTVTo Derive Elemental Load Time _..__.__.._____.._.....
45
funds have already been set aside.. ................. .60
issue is a red herring.. ................................
can be managed.. .........................................
.60
.61
The Commission
Should Recommend the One-Pound
Priority Mail Rate for the Flat Rate Envelope ................................
63
The Commission
Should
Rely on the “As-Filed”
Methodology
to Forecast the Number of Additional
Ounces in First-Class ....................................................................
67
The Commission
Should Eliminate the Nonstandard
Surcharge for Low Aspect Ratio ....................................................
70
Criticism of the OCA Rate Stability Proposal
THE OCA COST MODEL IS ACCURATE
Is Misguided............7
3
AND RELIABLE.. .................. .77
Docket No. R2000-1
A.
9.
C.
VII.
CONCLUSION
- iv -
The OCA Approach to Rounding Is an Improvement to
the Postal Service’s Cost Model _.,.................................................
78
The Manual Editing for Standard (A) Single Piece Costs
Is Clear . .._.___..__.____._.....................,.................................................
80
The OCA “Scratch Pad” Component Numbers Are Not
Confusing . . ..__.__......_._....................................................................
81
NEITHER THE POSTAL SERVICE’S NOR CSA’S FEE FOR
BULK
PARCEL
RETURN
SERVICE
SHOULD
BE
RECOMMENDED
BY THE COMMISSION
_._._______.,.,.....,....,...,,..,.,.,..,.,..,....................,..,,,,...,,..,..,....,.,..,.,....,.,...,.,..
81
85
Docket No. R2000-1
-“-
TABLE OF AUTHORITIES
FPC v. Louisiana Power & Light Co., 406 U.S. 621 (1972) ._.__._.____._...,....,............,..,.,..19
FPC v. Transcontinental
Gas Pipe Line Corp., 365 U.S. 1 (1961)
._.__._. ._._.
_. 19
National Association of Greeting Card Publishers v. U. S. Postal Service, 462
U.S. 810 (1983) .__._._......................................................................................................
26
Newsweek, Inc. v. USPS, 663 F. 2d 1186 (2d Cir. 1981), aff’d sub nom. Nat’/
Ass’n of Greeting Card Publishers, 462 U.S. 810 (1983)
passim
Panhandle Eastern
507 (1947)
Pipe Line Co. v. Public Service
Commission,
332 U.S.
19
United Parcel Service v. USPS 184 F.3d 827 (D.C. Cir. 1999)
Administrative
19
Cases
PRC Op. MC95-1 ...................................................................................................
.54, 58
PRC Op. MC97-4 ...................................................................................................
passim
PRC Op. R84-1 ......................................................................................................
passim
PRC Op. R90-1 .............................................................................................................
PRC Op. R94-1 ...........................................................................................
58
12, 48, 49, 50
PRC Op. R97-1 ......................................................................................................
passim
Federal Statutes
Postal Reorganization
Act, 39 U.S.C. ~3621...........................................................
Postal Reorganization
Act, 39 USC.
Postal Reorganization
Act, 39 U.S.C. 93622(b)(l).
......................................................
.I4
Postal Reorganization
Act, 39 U.S.C. 53622(b)(3)
........................................................
49
33622.. ..............................................................
18, 19
.I9
UNITED STATES OF AMERICA
Before The
POSTAL RATE COMMISSION
WASHINGTON,
D.C. 20268-0001
Postal Rate and Fee Changes, 2000
Docket No. R2000-1
)
REPLY BRIEF OF THE
OFFICE OF THE CONSUMER ADVOCATE
The Office of the Consumer
of Practice and Procedure
s3001.34,
and pursuant
Advocate
(“OCA”), pursuant to Rule 34 of the Rules
of the Postal Rate Commission
to Ruling No. R2000-l/71,
(“Commission”),
hereby submits
the Request of the Postal Service for a Recommended
39 C.F.R.
its Reply Brief on
Decision on Rates and Fees for
postal services.
SCOPE OF THE OCA REPLY BRIEF
Thirty-eight
this proceeding.
not possible
follows,
initial briefs were filed by the Postal Service and the participants
Under the time constraints
to reply to or comment
the OCA will discuss several
OCA does not respond
discussed
presented.
key issues in this case.
participants
argument
In the brief that
To the extent that the
related
to an issue
in the OCA Initial Brief, however, that does not indicate that the OCA’s views
have changed.
consistent
for a reply brief in a postal rate case, it is
on each argument
to an individual
in
The OCA urges the Commission
with the evidence and arguments
to issue its recommended
presented
by the OCA in this case.
decision
Docket No. R2000-1
-2-
EXECUTIVE
The excessive
challenged
level of the Postal
The importance
the credible
contingency.
the
urges
needed
requested
Commission
The Commission
The OCA
contingency
evidence
to the 2.5 percent
record,
has been
every class and category
of
contingency
to change
the existing
one percent
by the Postal Service simply is not present.
has
no choice
has the authority
the Commission
but to recommend
a smaller
to do so and must exercise
to recommend
the existing
it in this
one percent
provision.
The Commission
mail processing
should not be stampeded
operations.
that mail processing
short-run
analysis.
A voluminous
a change in the existing methodology.
should
Many of the flaws
record in this proceeding
substantial,
The Commission
to the affected mail processing
city carrier street time, the Commission
but
unambiguously
of the model, the data scrubs performed,
translate into one that provides the necessary
percent volume variability
below 100 percent.
of
in Docket No. R97-1 and the OCA have carried over to the
current study, including the specification
the continuing
into reducing the volume variability
The Postal Service has not demonstrated
costs must be substantially
noted by the Commission
percentages,
representing
request
As the OCA has shown in its evidence and argument,
and substantial
contingency
contingency
of a careful review of the Postal Service’s
request cannot be overstated.
case.
Service’s
by the OCA and by 27 intervenors
mail user.
On this
SUMMARY
continue
does not
credible evidence to support
should continue to apply 100
cost pools.
With respect to
should accept the Postal Service’s
to employ
and
LTV-based
elasticities
ES-based
to attribute
Docket No. R2000-1
elemental
-3-
load time.
Finally, the Commission
should continue
to attribute
coverage-
related load time by means of single subclass stop ratios.
The Commission
applied
to any service.
percent of attributable
short
should
of
covering
Reorganization
carefully
By reducing
consider
cost coverage
costs), the Commission
their
Act.
costs,
the minimum
the
to near-zero
to be
(approaching
100
runs the risk of services actually falling
primary
The OCA recommends
cost coverage
statutory
mandate
that the minimum
of
the
Postal
cost coverage
for any
service be 105 percent.
The
OCA urges that the 33-cent
declining
costs of handling
increasing
the institutional
The answer
mailers,
letter mail should
cost burden
larger discounts
benefits of Courtesy Envelope
rate be retained.
be recognized,
of First-Class
The
and the trend toward
letter mail should
for large mailers
Mail should be recognized
The Postal Service should overcome
First-Class
basic
be reversed.
lies in holding the line on the basic rate for the benefit of all First-Class
not in granting
mailers.
First-Class
Mail to be valuable
and relevant
only.
The cost-savings
by a discount for single-piece
its obduracy.
to consumers,
If the Postal Service wants
it must give consumers
choice and value.
The Priority Mail flat rate should be set at the new one-pound
Service has introduced
a one-pound
rate in recognition
rate.
The Postal
that the jump in the proposed
two-pound
rate is too high.
Yet, for the flat rate envelope,
consumers
for very low weight
items, the Postal Service proposes
two-pound
rate.
frequently
used by
to use the higher
Imposing a 20 percent cost increase for flat rate envelopes
is a lose-
-4
Docket No. R2000-1
lose result.
It will penalize
consumers
who want the convenience
of the envelope
provided and it will cause a reduction in the use of this uniform, machinable
The Postal Service’s attempt to change the additional
be rejected.
increasing
veiled
The “as-filed”
additional
effort
overpayment
ounces.
to offset
method
correctly
reflects
ounce methodology
the long-term
trend
The belated effort to change the methodology
the correction
the
Postal
of postage that had been erroneously
Service’s defense of the anachronistic
also is unconvincing.
piece.
nonstandard
Service
surcharge
The data does not support the surcharge
toward
is a thinly
had to make
omitted from its filing.
must
to reflect
The Postal
for low-aspect
ratio mail
for this limited segment
of mailpieces.
Attacks by intervenors
stability concept
on the OCA’s innovative single-piece
are unwarranted.
A technique
longer period of time is a win-win situation
The intervenors
costs, limit litigation, or otherwise
Finally,
promoted
and for the Postal Service.
by the OCA would not shift
unfairly affect other mailers.
the OCA cost model provides
an accurate
replication
Service’s cost model and improves upon that model by eliminating
of the Postal Service’s
rounding
at each step.
OCA’s model, and the Commission
decision.
Mail rate
for holding the stamp rate level for a
for consumers
fears are baseless; the concept
First-Class
should
of the Postal
the distorting
impact
The Postal Service should adopt the
rely upon it in reaching
its recommended
Docket No. R2000-1
-5-
ARGUMENT
I.
THE
COMMISSION
PROPOSAL
The Postal Service’s
SHOULD
excessive
ADOPT
contingency
THE
OCA’S
request has unleashed
and severe criticism by almost all of the active participants
billion
increase-to
evidence,
a 2.5
has not been adequately
postal and mailer interests.
evidence
percent
contingency-is
explained,
CONTINGENCY
to this proceeding.’
not
supported
and would
The $1
by substantial
be counter-productive
The OCA, on the other hand, has presented
that the most prudent
widespread
course of action is to maintain
to
convincing
the contingency
at its
current one percent level. The OCA’s position should be adopted by the Commission.
A.
The Postal Service’s Attempts To Minimize the Magnitude
Are Unpersuasive
In its brief, the Postal Service characterizes
to the contingency*
its proposal
as a “slightly higher level of protection”
of the Increase
for a $1 billion addition
and a “simple mathematical
1
The Postal Service’s contingency request is actively opposed by the OCA, the Direct Marketing
Association, Inc., ADVO, Inc., the Alliance of Independent Store Owners and Professionals, the Alliance of
Nonprofit Mailers, Amazon.com,
Inc., American Business Media, American Library Association, the
Association for Postal Commerce, the Association of Priority Mail Users, Inc., the Coalition of Religious
Press Associations,
Dow Jones 8 Company, Inc., the Florida Gift Fruit Growers Association, the
Magazine Publishers of America, the Major Mailers Association, the McGraw-Hill Companies, Inc., the
Parcel Shippers Association, Time Warner, Inc., the Greeting Card Association, Hallmark Cards, Inc., the
American Bankers Association, the National Association of Presort Mailers, the American Association of
Publishers, Val-Pak Direct Marketing Systems, Inc., Val-Pak Dealers’ Association, Inc., Carol Wright
Promotions, Inc., and the Classroom Publishers Association.
2
The Postal Service’s request for a $1.68 billion contingency
versus the OCA proposed contingency of $676 million (Tr. 41/18303).
provision
(Tayman
Exh. USPS-gA)
Docket No. R2000-1
phenomenon.“3
work.
-6-
Unfortunately
for the Service this public relations ploy simply does not
As the OCA pointed out in its brief (at 41) the Postal Service’s
proposal
in this proceeding
increase
in revenues.
distinction,
comprises
Furthermore,
an unprecedented
the 2.5 percent
60 percent of the requested
proposal
has another
Le., in every rate case from Docket No. R76-1 through
Service has proposed
case4 or proposed
this pattern
a contingency
a reduction
contingency
that either maintained
dubious
R97-1, the Postal
the level from the previous
in the level of the contingency.5
The sole exception
is found in Docket No. R64-1 when the Postal Service proposed
modest 17 percent increase in the proposed
far cry from the Service’s
proposal
contingency
to increase
a more
(from 3% to 3.5%).
the contingency
to
That is a
150 percent
in this
proceeding.
For
scrutinized
these
reasons,
the
Postal
Service’s
with greater care than past proposals.
contingency
instant
a misconception
case.
component,
Where
Service
be
is being urged (USPS Brief at II-
about the Postal Service’s
the Postal
must
The Postal Service’s arch suggestion
that a cynic might question whether a double standard
4) reveals
proposal
seeks
it is indeed subject to the requirement
burden
a substantial
of persuasion
increase
that the change
in the
in a cost
be supported
by
3
USPS Brief at 11-4, -2. The Postal Service’s description of the proportion of the contingency as a
fallout of a modest operating deficiency and growing budget adds further weight to the conclusion that the
Service selected 2.5 percent ($1.68 billion) as a plug figure. See OCA Brief at 67-73.
4
In Docket No. R77-1. the Postal Service proposed the same contingency as it had proposed in
Docket No. R76-1; likewise in Docket Nos. R84-1, R87-1, and R90-1, the proposed contingencies were
the same.
5
The Postal Service’s Docket No. R80-1 contingency proposal was a reduction from its Docket No.
R77-1 proposal, and the Service’s Docket No. R94-1 proposal was a reduction from its Docket No. R90-1
proposal.
-7-
Docket No. R2000-1
credible
and substantial
Brief at 12-14.
evidence,
Earlier proposals
not just subjective
management
to reduce or maintain then-current
selection.
OCA
contingency
levels
do not require the same degree of support or examination.
Citing the Commission’s
Service’s
requested
opinion in Docket No. R94-I, which approved the Postal
reduction
convince the Commission
the greater showing
in the contingency,
in R2000-I,
is unfair.
Service has cited Commission
statements
indicating
evidentiaty
circumstances
passages
burden
extant
in the
Service
attempts
that the lesser showing made in that case, as compared
required
a heavier
the Postal
in a case such
in Docket
Docket
No.
No. R94-1.
R94-1
Ironically,
with
in its brief, the Postal
that the Postal Service does bear
as this,
The
opinion
to
in sharp
Service
that
contrast
quotes
describe
the
from
with the
important
distinguishing
circumstances:
In this case, the Postal Service has limited the magnitude of its proposed
contingency provision for the overt purpose of constructing a constrained
revenue requirement in order to restrain the overall level of rate increases
as a business objective.
*
l
*
witness
Ward defends~the proposed
“other valid business considerations.
Postal Service’s concern about the
base” [and]
postal managements
to create “the opportunity
to build
In direct testimony,
expectations
6
that regulators
OCA witnesses
*
l
2 percent allowance on the basis of
.” [witness
Ward voices the
“long-term viability of its customer
goal of restraining rate increases
[the Service’s] revenue base.‘@
Burns and Rosenberg
outline the contrasting
have for utilities that evidence a determined
USPS Brief at II-6 quoting from PRC Op. R94-1, m2036 and 2030
effort to control
-8-
Docket No. R2000-1
costs, as opposed to utilities that strive to manufacture
avoid aggressive
a large cushion allowing them to
measures to control costs. Witness Burns cautions that:
Where managers are sheltered from the effects of future events, they
make less effort to take actions to control costs.
A larger than
necessary contingency
reserve creates a similar perverse managerial
incentive.
Managers cushioned from the consequences
of controlling
costs will tend not to act as aggressively to cut costs and waste. They
become lax. This could, and likely would, happen for Postal Service
managers if the contingency reserve were raised to a level that exceeds
reasonable provision for future, uncontrollable events and thus acts to
cushion managers from the consequences
of failing to curb controllable
costs.’
Witness Rosenberg
counsels:
[l]f the allowed contingency provision is too large, the cushion may result
in a tendency toward slackness.
This is a form of what economists call
moral hazard.
[l]t means that the structure of incentives and rewards
may not lead to cost minimizing behavior.
If the contingency
provision is too generous, managers can still meet their breakeven goal in
the face of adverse circumstances,
without having to make tough
decisions.
A contingency provision that is overly generous can relieve
Postal Service management of the pressure to manage economically and
efficiently.’
Moreover,
pockets
the $1 billion dollar cushion
of postal customers,
and individuals
profitable
causing
is not costless-it
them to suffer an opportunity
who use the mails at unnecessarily
investments
rates of interest far above Treasury
8
Tr. 2219826-27
Businesses
rate levels must forego
rates.
and may be forced to borrow money at
Tr. 2219828.
if the Postal Service chooses to narrow the contingency
Tr. 2219713-14.
loss.
out of the
(with rates of return higher than the rates the Postal Service must
pay if it borrows money from the U.S. Treasury)
7
bloated
comes
The opposite condition ensues
from higher levels.
-9-
Docket No. R2000-1
Unintentionally
passage
adding further proof of this propensity,
from a House
Postal Reorganization
Report that constitutes
the Postal Service cites a
part of the legislative
could
appropriations”
easily
view of the contingency
consequence
overruns
“contingency
as a subjective,
risk that can be tolerated””
recognized
substitute
in this quote and do no violence
large contingency
of the
Act:
So long as postal management operates with a general
congressional
appropriations are always available
to
shortfalls of revenues or overruns of costs, there is little
make the best possible use of resources and efficiency is
honored in the speech than in the observance.’
One
history
provision”
for
to its meaning-the
intuitive “policy judgment
demonstrates
awareness that
make good any
real incentive to
sure to be more
that the Service’s
“congressional
Postal Service’s
regarding
the level of
object is to manufacture
a
cushion whose purpose is to shield postal mangers from any adverse
arising
from
an inability
to control
in the House report, establishment
and revenue
shortfalls
destroys
expenses.
Just
of a bloated contingency
any incentive
as Congress
to cover cost
to manage economically
and
efficiently.
In short,
contingency
evidence
9
a double
proposals
standard
must
is not presented),
is eminently
be carefully
reasonable-dramatically
scrutinized
while contingencies
proposed
(and
rejected
to advance
higher
if compelling
sound business
USPS Brief at 11-16, quoting from House Report No. 1104, 91” Cong., 2d Sess. (May 19, 1970) at
16-17.
“Responses of United States Postal Service to Questions of the Office of the Consumer Advocate
Regarding the Provision for Contingencies,” filed May 17, 2000 at 2.
-lO-
Docket No. R2000-1
objectives
and lighten the burden that must be borne by postal customers
should be
applauded.
B.
Questions
Specious
Raised
by the Postal Service
In its initial brief, the Postal Service emphasizes
has declined to $226 million through AP 12.”
about
its Equity Position
that its year-to-date
This is followed
Are
net income
by the rash speculation
that a “typical” AP 13 and AP 14 loss will occur soon after. Also, an allusion is made to
the Patelunas
speculation
roll-fonvard
calculation
estimating an FY 2000 loss of $325 million.
such as this should be immediately
dismissed,
particularly
Rank
since there is
ample record evidence that these losses may not occur.
First, the OCA established
in its initial brief that the Postal Service’s failure to use
actual AP cost data for FY 2000 has likely produced
a gross overestimate
costs.
its calculation
The OCA cites record evidence
$64.454
billion for interim year FY 2000 (OCA Brief at 18-21).
overstatement
22).
to support
of costs by witness Patelunas
Since witness Patelunas’
estimate
of $65.1715
improbable.
billion
computation
(Exh.
Witness Rosenberg
such an occurrence.
of a likely cost of
This indicates a likely
of up to approximately
$700 million (id. at
of a $325 million deficit stems from a cost
USPS-ST-44A),
expressed
of FY 2000
a $325
strong reservations
million
deficit
is highly
about the likelihood of
Tr. 41/18307.
Second, witness
Rosenberg
income figure less favorable
points out in his rebuttal testimony
than the PFY figure is not inevitable.
that a GFY net
In the case of FY
See “Notice of United States Postal Service of Errata to Initial Brief,” filed September 19, 2000.
Docket No. R2000-1
1999, for example,
-11.
the GFY net income was higher than the PFY net income.
Id. at
18307-08.
Third, the Postal Service is taking pains to curb expenses
year.
OCA Brief at 57. If these measures
be able to keep costs below budget,
loss for the accounting
12.
equal to budget projections,
then the Postal Service may
as it did in AP 12,” and generate
period than was budgeted.
Also, if the Postal Service’s
clearly exaggerated
are successful,
at the end of the fiscal
a smaller
This, too, was accomplished
luck holds out, then revenues
net
in AP
will be approximately
as they were in AP 12. In its brief, the Postal Service has
the bad news and avoided any mention of recent good news.
Finally, the Postal Service’s equity position is far better today than it has been in
the past.
While it is true that its accumulated
approximately
$35 billion (through
losses since reorganization
now stand at
FY 1999),‘3 it should be recalled that at the end of
FY 1994 they stood at $9 billion. The Postal Service’s fiscal health is sound.
C.
The Record Is Devoid of Substantial Evidence in Support
Percent Continaencv Proposed bv the Postal Service
In its brief, the Postal Service blusters that it has supplied substantial
support
being
“postal managements
accurate.
proposal,
13
Few reasons
have been advanced
in the future.
Financial and Operating Statement for AP 12.
Tayman Exh. USPS-9L.
evidence to
USPS Brief at 11-2. This claim is far from
in support
of the 2.5 percent
and these few amount to no more than vague apprehensions
adverse tendencies
12
judgment.”
of the 2.5
about possible
-12-
Docket No. R2000-1
The Postal
Service
Strasser testimonies.
been accounted
testimonies.
for any
sets forth a short
list of worries
A modest one percent contingency
misestimation
and
USPS Brief at 11-6-7. For the most part, these uncertainties
for in the various Postal Service witnesses’
possible
from the Tayman
misestimation
has been greatly
economic forecast information
of these
diminished
cost, volume, and revenue
is all that is necessary
Moreover,
amounts.
have
to compensate
the
due to the substitution
likelihood
of more
of
recent
for interim year 2000 and test year 2001 cost estimation.
One reason given by witness Strasser in defense of the $1.68 billion contingency
is that Docket No. R2000-1
January
2001.
proposed
took
rate increases
that
rates are in effect for all of the Test Year.“14 In reliance on the exchange
that
between
appearance,
However,
Chairman
Gleiman
the OCA emphatically
billion contingency
requested
“[t]he Postal Service
any earlier than
filing assumes
place
Tr. 46AI20199.
will not be implemented
and
contends
witness
that roughly
by the Postal Service
Tayman
during
Tayman’s
$425 million of the $1.68
must immediately
be disallowed.
The Postal Service’s express intent to use the contingency
in lieu of higher rates for the
beginning
pointed out, an unforeseen
of the test year was not, as Chairman
Gleiman
risk or event.15 This means that the Postal Service’s
contingency
request really is for
$1.255 billion.
The failure of the Postal Service to support
even
14
15
to the $1.255
billion
level-leaves
open
Tr. 21/9273 (Postal Service response to interrogatory
Tr. 2/560-63.
The Chairman’s
an increase
the question
in the contingencyof what
PSAAJSPS-1).
statement is consistent with PRC Op. R94-1, 72043
the correct
-13-
Docket No. R2000-1
contingency
provision
should
be.
evidence.
As discussed
witnesses
Burns and Rosenberg
continuation
It is to this question
in the OCA’s Initial Brief (pp. 42-58)
provides
substantial
of the present one percent contingency
D.
that the OCA directed
the evidence
and credible
that the Periodicals
toward the contingency
of OCA
support
for the
provision.
Periodicals Should Not Be Relieved of the Contingency
Incurred Costs
The suggestion
its
class be exempted
Provision or Other
from any contribution
was first made in the testimony of American
Business Media, et
a/., witness Morrow.”
The joint brief of the Alliance of Nonprofit Mailers, et al., defends
this novel proposition
on brief.
witness
Morrow’s
contingency
ANM Joint Brief at 50-59.
novel proposal
for exempting
and the effort by Periodicals
The OCA strongly opposes
Periodicals
from any share
mailers to shift all risk of unforeseen
of the
events to
the other classes of mail.
Witness
warranting
include
Morrow contends
the proposed
proposed
exceptional
rate increases
reductions
for processing
increases
for periodicals.‘g
that exceptional
for them.
According
that are difficult to quantify,”
ABM-T-l; Tr. 29/13542-13560.
Magazine Publishers Association
view in her supplemental testimony, MPA-ST-1 (Tr. 38/17072-81).
17
Tr. 29/13549 (ABM-T-l
18
Id. at 13555-57.
Id. at 13554.
at 7).
attach to Periodicals,
to Morrow, these
far above the average for periodicals,”
periodicals
16
treatment
circumstances
planned cost
and unexplained
witness
cost
Cohen joined in this
Docket No. R2000-1
-14-
The factors
contingency
for periodicals
Postal Service’s
explaining
cited by witness
proposed
Morrow
are essentially
in defense
of the proposal
for a zero
the same factors that were weighed
low cost coverage
for this subclass.
her choice of a near zero cost coverage,
Witness
in the
Mayes,
in
explicitly states that the proposed
cost coverage for Outside County Periodicals
has been further reduced due to consideration
of the effect of rate
increases
Without this consideration,
the large increases in unit
costs would have led to even higher percentage
rate increases for
Outside County Periodicals.
Despite the objectives of both the Postal
Service and the Commission
in previous cases to move the cost
coverages
for Periodicals
upward to provide a more meaningful
contribution to other costs, the recent increase in costs precludes doing so
at this time. USPS-T-32 at 33.
She apparently
also took into account
what factors may have contributed
for Periodicals”
Periodicals
equity
(1) Postal Service efforts “to understand
to increases
in flats mail processing
and (2) that “[t]he Postal Service
is also committed
mailers to reverse the cost trends of recent years.”
required
of every
set of proposed
periodicals
both from the contingency
institutional
costs of the Postal Service.”
rates,”
however,
As is clear from the Postal Service’s
already made extraordinary
39 USC.
21
to working
with
Id. The fairness
and
clearly
bars exempting
and from more than a token contribution
There is, of course, no reason to grant Periodicals
contingency.
costs, especially
to the
mailers special relief from the
Initial Brief, the Postal Service has
efforts to modify its filed case to the exclusive benefit of the
5 3622(b)(l).
In Section IV of this Reply Brief, the OCA recommends that the minimum cost coverage for any
class of mail be established at 105 percent to avoid the potential for any class of mail to under-contribute
to the Postal Service’s institutional costs.
-15-
Docket No. R2000-1
Periodicals
class.
filed position
While the Postal Service doggedly
against
other mailers,
defends
it urges the Commission
changes to its filed case that benefit only the Periodicals
VII-137-148.
Service’s
direct case.
decrease
Periodicals
for Periodicals,
Id. at VII-139.
affect other classes of mail.
so long
as a corresponding
variability
however,
regressions
affect other mailers.
circumstances
are
each subclass
circumstances
share of the contingency.
Commission
overall
can be
revenue
The request to use the
from the new ES study, for instance,
The OCA strongly opposes
is
do not adversely
supported
in the
should be rejected.
Id. at VII-144.
would
the use of the untested
even apart from the Postal Service’s cost relief efforts, there is no
reason to single out Periodicals
exceptional
is made
The Postal
If the Commission
that are adequately
reduction
to
OCA Initial Brief at 135-43.
Furthermore,
contingency,
to make all of these changes.”
Cost reductions
in the Postal
$200 million.
it should be careful that the changes
Other requests,
Baron load-time
singular
Postal Service Brief at
relief effort must be closely examined.
inclined to accept any changes,
regressions.
of
“would allow the Commission
costs by a large amount-approximately
This extraordinary
adversely
class.
beyond those contained
These changes
Service strongly urges the Commission
requirement.
to adopt a number
As the Postal Service states, it is now willing to accept “a number of test
year cost savings opportunities
made,
the vast majority of its
for special relief from the contingency
examined
to
determine
and service must be similarly
call for an average,
Periodicals’
examined
above-average,
provision.
share
of
the
to see whether
or a below-average
Since the record is devoid of any such assessment,
cannot make the requisite factual findings to do so.
If
the
Docket No. R2000-1
Moreover,
Class Letters
-16-
if such an assessment
and Sealed
parcels,
were to be performed,
more than almost
it is clear that First
any other subclasses
should be relieved of more than just a token share of the contingency.
the cost coverage
that proposed
proposed
the Periodicals
class.
Postal Service witness Mayes requests a First-Class
April 21, 2000, in Exh. 328.
By contrast,
revenues will not be sufficient
costs.
conclusion,
First-Class
contribution
cost coverage
for Outside
is to protect against the possibility
to cover estimated
costs.
In light of the
that this mail will fail to pay for all of its estimated
to follow
would
the logic of the Periodicals
have
County Periodicals
of only 101.37 percent,
to cover estimated
to be applied
22
Mail
she proposes
to
be exempted
from
mailers
any
to its
significant
toward the contingency.
The Outside
insufficient
Perforce,
at 20, as corrected
Letters and Sealed Parcels will include such a high
it is virtually impossible
attributable
Mail Letters
Id. at 32.
of the contingency
fact that the rates for First-Class
cost coverage,
See USPS-T-32
the cost coverage
is a mere 101.37 percent.
One of the chief purposes
that estimated
is far higher than
and certainly much higher than for
and Sealed Parcels cost coverage of 197 percent.
County Periodicals
This is because
by the Postal Service for this subclass
for the majority of other subclasses,”
of mail,
to different
attributable
classes
subclass,
on the other hand, with a proposed
runs a substantial
costs.
risk of generating
revenues
If different levels of contingency
of mail, Periodicals
should
therefore
were
be made
Express Mail, with a proposed cost coverage of 222.2 percent, and Enhanced Carrier Route Mail,
with a cost coverage of 208.8 percent, are hvo notable exceptions. USPS-T-32 at 28 and 38, respectively.
-17-
Docket No. R2000-1
responsible
for an above-average
share of the contingency.
from any share of the contingency
would be cross-subsidized
Periodicals
the odds that Periodicals
by other classes of mail, such as First-Class
Periodicals
would disproportionately
attributable
cost estimates
In its Docket
would vastly increase
Exempting
contribute
to the depletion
Mail, or that
of the contingency
for Periodicals are too low.
No. R97-1
opinion,
the Commission
articulated
the bedrock
principle that: “The intent of the framers of the Act was to prohibit cross subsidies”
that “[rlecommended
reasonable
periodicals
rates for each subclass
contribution
to institutional
run afoul of both of the requirements
For the foregoing
proposal for Periodicals
reasons,
and
costs, plus a
proposal
to exempt
would cause the rates for Periodicals
by the Commission,
to
i.e., (1) not bearing
increases the risk that the Periodicals
contribution
cost coverage
toward recovery of institutional
the OCA asks that the ABMIANM
subclass
prevents
costs.
contingency
be rejected
The Commission
for Continqencies”
Is Authorized
Several parties have discussed
23
The ABM/ANM
costs and (2) paying a negligible
this mail from making a meaningful
adjustments
costs.“23
articulated
share of the contingency
will not cover its attributable
E.
must recover attributable
from any share of the contingency
a proportional
if
to the Postal Service’s
To Determine
the “Reasonable
on brief the authority of the Commission
requested
provision
for contingencies.24
Provision
to make
These
PRC Op. R97-l.nn4026 and 4025.
24
See, e.g., DMA et al. Joint Brief Concerning the Revenue Requirement at 3-6: GCA & Hallmark
Joint Brief at 29-31; VPlCW Brief at 75-76.
Docket No. R2000-1
parties
correctly
Commission
included
-18-
argue that the Postal Reorganization
authority
to determine
the “reasonable
in the revenue to be derived
Commission
showsZ5
agrees,
as its analysis
requested
authority
provision
from rates.
decision
to recommend
a contingency
in Docket
provision
to be
Certainly
the
No. R84-1
the OCA will discuss a further rationale
different
for the
from that
by the Postal Service.
If one starts from the assumption,
contingency,
that the Commission
implied in the Postal Service’s
has no authority
That is because
contingency
the Postal Service
requirement.
The
contingency
in omnibus
well have encouraged
Commission
has
not
in the revenue
Quite possibly
itself
are
against
not the adverse
since Docket No. R80-1.
a contingency
the “contingencies”
revenue
external
its authority
requirement
requirement
provision,
as a whole.
by simply inflating the
in other elements
exercised
the Postal Service to propose
to make.26
Commission,
reductions
rate cases decided
offset any other reductions
protecting
could “game the system”
provision to offset anticipated
view of the
over the contingency
then, as a practical matter, it has no authority over the revenue requirement
choose
upon the
for contingencies”
39 U.S.C. § 3621.
of the Newsweek
In this section of argument,
Commission’s
Act directly devolves
of the revenue
to reduce
This may very
large enough
that the Commission
that the Postal
adjustments
the
made
to
might
Service
is
by
the
events cited by Postal Service witnesses.
In
*s
See PRC Op. R84-1 m 101947, discussing Newsweek, Inc. v. USPS, 663 F. 2d 1186 (2d Cir.
1981), affd sub nom. Nat’/ Ass’n of Greeting Card Publishers, 462 U.S. 810 (1983).
26
In Docket No. R97-1, for instance, the Commission made various adjustments to the requested
revenue requirement for a net reduction of $745 million from the Postal Service’s request. Docket No.
R97-1 Opinion at ii; App. C.
Docket No. RZOOO-1
United Parcel
-19.
Service
v. USPS:’
the Court
of Appeals
described
a similar
Postal
Service gaming scheme that was laid bare in the Docket No. R97-1 proceeding:
After it became apparent that its original revenue estimates were overly
pessimistic, the Postal Service reported to the Commission that it would
face more costs than it had initially predicted.
[T]he Service predicted
that it would need $300 million more than it had initially requested for
discretionary programs
The Commission rejected what it viewed as
attempts to avoid the full impact of the Service’s bright economic situation,
labeling the new 1.5% contingency number “a plug figure” used by the
Service to counterbalance
the decrease in the size of its contingency fund
[T]he Commission pointed to a Postal Service documentinadvertently included as evidence and initially disavowed by the Service
as inauthentic-that
identified the Service’s updating “strategy”
The most reasonable
should be applied
interpretation
to avoid a regulatory
often been recognized
by the courts.
Pipe Line Corp. 928the Supreme
jurisdiction
to deny a certificate
Gas Act of 1938. This conclusion
of 39 U.S.C. 5s 3621 and 3622 is that they
gap.
This tenet of statutory construction
For example,
in FPC v. Transcontinental
Court held that the Federal
of public convenience
Power Commission
and necessity
has
Gas
had
under the Natural
rested upon the principle that:
Congress did not desire an “attractive gap” in its regulatory scheme
[Tlherefore,
when we are presented with an attempt by the federal
authority to control a problem that
state regulatory commissions can
[not] be expected to deal [with], the conclusion is irresistible that Congress
desired regulation by federal authority rather than nonregulationZ9
27
28
29
184 F.3d 827,831 (D.C. Cir. 1999).
365 U.S. 1, 28 (1961).
The Court also noted that this conclusion was consistent with the “broader principle” that
Congress “meant to create a comprehensive and effective regulatory scheme,” id. at 19. citing Panhandle
Eastern Pipe Line Co. v. Public Service Commission, 332 U.S. 507. 520 (1947). This reasoning was
applied in a later case also involving Federal Power Commission jurisdiction-WC
v. Louisiana Power &
Light Co., 406 U.S. 621, 631 (1972): “FPC regulation was to be broadly complementary to that reserved
to the States, so that there would be no ‘gaps’ for private interests to subvert the public welfare.
‘[Wjhen a dispute arises over whether a given transaction is within the scope of federal or state regulatory
(continued on next page)
Docket No. R2000-1
Consequently,
they
-2o-
the most compelling
invest the Commission
contingency
proposal
interpretation
with authority
of Sections
to review
3621 and 3622 is that
and, if necessary,
modify
a
made by the Postal Service to ensure that it is “reasonable.“30
Any other view would open a gap through which the Postal Service could avoid review
of its overall revenue requirement
In Docket No. R84-1, the first omnibus
Court of Appeals’
Governors
decision
in Newsweek,
rate case following
of the Postal Service, rather than the Postal Rate Commission,
the Commission
could
not make adjustments
unless objective factual or mathematical
the
extreme
approved
claim that
by the Board.”
“the
disagreed
that the most reasonable
Commission’s
Commission
practices
observed
was
found
must
to the revenue
accept
the
with this restrictive definition
view of the Newsweek
than the Postal service claimed.
the
Also, according to the
errors had been unearthed.
Commission
has the so/e
requirement
The Service made
revenue
requirement
ld.
The Commission
on
Circuit
the Postal Service argued that “the Board of
authority to determine the amount of the revenue requirement.“3’
Service,
the Second
adopting
ld. at 7 1028.
the
role
Id.
noting
opinion was “much less far-reaching”
The Newsweek
holding was premised
of “disciplinarian.”
that its failure in that case to explain
to be arbitrary.
of its authority,
at fi1031.
Furthermore,
its departure
Weaving
the
from past
together
these
authority, we are not inclined to approach the problem negatively, thus raising the possibility that a “no
man’s land” will be created,“’ quoting Transcontinental Gas Pipe Line, supra, 365 U.S. at 19.
30
I.e., ‘“a reasonable provision for contingencies.”
31
PRC Op. R84-1 1 1021. The Commission
paraphrased the Postal Service’s initial brief.
Docket No. R2000-1
observations,
the
responsibility
-21-
Commission
concluded
that
to adjust the revenue requirement
as adjustments
proposed
are fully explained and do not encroach
Id. at lJlJ 1039, 1047.
Moreover, the Commission
rates that are accurate and supported
that are distinguishable
1192, 1203.
4.3 percent
Docket
requested
Docket
No.
R2000-1
revenue
constituting
a much more modest adjustment
The second important distinguishing
adjustment
to an effective
Id. at fi 1045.
of the Newsweek
in Docket
case
describes
the
663 F. 2d at
of
One billion dollars constituted
requirement
a $1 billion (rounded)
requirement
of $67
of $23 billion.
reduction
billion,
in the
USPS-SA,
of 1.5 percent.
circumstance
is that the Commission
from the Docket No. R77-1 contingency
1.8 percent
lawful
about the large proportion
No. R80-1 revenue
PRC Op. R80-1 at 3. The OCA also proposes
a downward
record.
The Court of Appeals
comprised.
and
prerogatives.
to recommend
case ($1 billion) as “staggering.”
that the reductions
of the requested
authority
upon managerial
This suggests that the Court was concerned
the revenue requirement
the
by the Postal Setvice so long
circumstances
from Docket No. R2000-1.
at issue in the Newsweek
have
is duty-bound
by the evidentiary
It is also important to note three additional
reductions
it does
No. R80-1.
allowance
made
of 4 percent
This fact was given considerable
weight by the Court:
The PRC slashed the contingency fund to 1.8 percent
In view of the
four percent contingency provision approved by the Board in the last rate
case
we agree with the Board that the reduction of the fund to less
than half the percentage of the previous rate filing was arbitrary.
663 F.2d at 1205.
By contrast, the OCA proposes
current level, a level fully embraced
maintaining
the contingency
by the Board of Governors just two years ago.
at its
Docket No. R2000-1
-22
A third significant
eliminated
arbitrary
the recovery
by the Court.
difference
is that in Docket
of prior years’ losses without
ld. at 1204.
losses in this proceeding.
No. R80-1
explanation,
the Commission
an action deemed
No party has asked for elimination
In fact, witness Rosenberg
as one of the reasons that a moderate
contingency
of prior year
explicitly cites this cost element
provision
is feasible.
He outlines
several “safety nets” that are at the disposal of Postal Service management:
forma estimates
of revenues
(3) a new rate request,
and expenses,
if necessary,
(2) the Postal Service’s
(4) managements
recovery of prior years’ losses, and (6) the contingency.
options,
there
“necessarily
is no reason
to believe
that
another ground asserted by the Newsweek
ability to borrow,
ability to control expenses,
Tr. 2219819.
a reduction
have th[e] effect” of causing more frequent
(1) pro
(5)
In view of these
in the contingency
will
rate filings, 663 F.2d at 1204,
Court for reversing the Commission’s
action
in that case.
In light of the Postal Service’s failure to provide credible and substantial
in support
of its request,
determination
the Commission
of a reasonable
parties on brief, the Commission
The evidence
the present
presented
contingency.
recommendation
has the duty to make an independent
As discussed
above and by the other
has the authority to recommend
a lower contingency.
by the OCA and the other parties strongly supports
level of one percent.
would withstand
Given the facts of record
appeal.
evidence
retaining
in this case, such a
Docket No. R2000-1
II.
-23.
MAIL PROCESSING
In the following
VOLUME
VARIABILITY
sections, the OCA responds
support of its volume variability position.
errors in the approach
database
the
deficiencies,
economic
Service’s
The sections first discuss the two fundamental
and econometric
of significant
defects.
model relating
model the longer-run
Because
of the
Postal
issues in its Initial Brief, the OCA is unable to respond
on those important
to the OCA’s initial brief for a discussion
the Postal Service
in
the need for a long run study over the rate effective time period,
fully to the Postal Service argument
Commission
to the Postal Service arguments
taken in the Postal Service initial brief and other key issues:
methodologies,
omission
SHOULD BE 100 PERCENT
to QICAP
(III.E.l(b)
issues.
The OCA refers the
of the significant
and E.3.(a))
deficiencies
in
and the need to
(lll.E,2.(c)).
The Postal Service’s initial brief is more telling by the issues that it ignores than
by the issues that it discusses.
econometric
studies
“unambiguously”)
Although
The thrust of the Postal Service argument
in this record demonstrate
a volume variability
conclusively
for mail processing
(the Postal Service says
of less than 100 percent.
many, but not all, of the studies purport to show volume variability
100 percent, the Postal Service arguments
is that the
is less than
fail to consider important factors and are not
persuasive.
A.
The Postal Service’s Initial Brief Contains Two Fundamental
The Postal
accurate
underlying
Service’s
Initial Brief has two fundamental
model of mail processing,
and it assumes
flaws--it
Flaws
assumes
an
that the longstanding
Docket No. R2000-1
- 24 -
policy of the Commission
variability is probably
1.
must be changed if the Postal Service only shows that volume
less than 100 percent.
The Postal Service’s initial brief erroneously
accurately reflects mail processinq operations
The first of the two fundamental
deficiencies
the Postal Service erroneously
mail processing
operations.
theories and economics
incorrect application
underlying
primav
in the
No.
reflects
The Postal Service studies are based upon an
The Postal Service
Docket
is that
as the underlying
There also continue to be deficiencies
thrust of the Commission’s
compiled
model accurately
The model does not do so inasmuch
are deficient.
its model
in the Postal Service’s dissertation
assumes that its econometric
of economics.
the study.
assumes
has failed to adequately
objections
R97-1
with the data
respond
to the
to the fixed effects model previously
opinion.
criticized
the
It also criticized the model’s handling of variables;
both
appropriateness
of the model.
those variables
that vary with respect to volume
The
Commission
and those variables
that are not
affected by volume (the control variables).
The Postal
criticisms.32
Service’s
initial brief primarily
The brief is misleading
intervener witness offered conflicting
Neels.” USPS Brief at V-13.
Smith’s testimony
the proceeding
conflicted
and somewhat
attempts
disingenuous
UPS witness
knows very well that OCA witness
with witness Degen’s analysis.
not an intervenor.
Neels’
in stating, “Only one
analysis of Postal Service operation,
The Postal Service
and, technically,
to dispel witness
The OCA is a participant
The Postal Service’s statement
in
may
-25.
Docket No. R2000-1
be technically
accurate,
but, nevertheless,
it is misleading
and ignores
extensive
criticism in this record of witness Degen’s comments
The Postal Service brief ignores the significant
concerning
the inappropriate
specification
It also totally ignores discussion
for an understanding
important.
Without
both regarding
of the longer-run
which specifications
brief also ignores other significant
for
characteristics,
a complete
and physical
expansion
measurement
these theoretical
Another
Postal
percent
Service
are meaningless,
values of the variabilities.34
issues relating to the omission
analysis;
characteristics
e.g.,
capacity
The Postal Service
of variables that are
utilization,
of the processing
certain
facilities
only needs to demonstrate
network
such as floor
facilities.
more than a probability that
flaw in the Postal Service’s initial brief is the assumption
to compel the Commission
The Supreme
of costs (across facilities) is
The Postal Service must demonstrate
variabilitv is less than 100 oercent
fundamental
path that is so significant
issues, conclusions
space, age of facility, and number of stories in processing
2.
in the model, QICAP.
are preferable (e.g., models A, B, or C as defined in
NOI No. 4)33 and what are the computed
necessary
of the capital variable
of why the longer-run
resolving
issue raised during the hearings
that variability
is probably
to scrap its present policy.
that the
less than 100
That is not the case.
Court has made clear that the selected method of costing must attribute
32
The Postal Service devotes the following sections in response to witness Neels (USPS Brief,
Section V: pages V-13-21[§la]; V-23-26[§lc]; V-374O[§§lf-g];
V42-44[1ij; V-47-49(§ljiii);
V-5061(§11-0).
33
Notice of inquiry No. 4, “Concerning
34
NOI No. 4 response of witness Smith, Tr. 46EI22363-69.
Mail Processing Variability Models,” August 2, 2000.
- 26 -
Docket No. R2000-1
to a given class costs that are reliably identifiable.
the mail processing
costs attributable
Service to demonstrate
This model does not reliably identify
to the given classes.
The burden is on the Postal
that a change in policy is warranted.
The Commission
does not
have a basis on which to select a specific variability from this record with any assurance
that it is reaching the correct result.35
Assume,
disadvantages
The several
MODS pool.
arguendo,
that the Commission
of using an incomplete
studies
study are not outweighed
in the record compute
Some variabilities
100 percent,
prefers to adopt the Bozzo model.
computed
and witness
very different
The
by the advantages.
point variabilities
for each
by the Postal Service are not significantly
different
from
Neels found variabilities
percent.
In fact, even two of witness Bozzo’s ten recommended
in excess
estimated
of 100
variabilities
are very close to 100 percent (89.5 percent for BCS cost pool and 95.4 percent for the
LSM cost po01).~~
Postal
percent
Service
is significantly
witness
Greene
testified
that, in his view, a variability
different from 100 percent,
95.4 and 89.5 are not percentages
but he agreed the percentages
that are substantially
whether they are the right variabilities for those pools.
35
of 85
of
less than 1. He did not know
Tr. 46EI22088.
He also testified
National Association of Greeting Card Publishers v. U.S. Postal Service, 462 U.S. 810, 820
(1983). See a/so, OCA Brief at 92-3. The Periodicals Mailers’ brief contends that if the Commission
declines to recognize that a considerable portion of mail processing costs are not volume variable, it is not
free under the National Association of Greeting Card Publishers case to apply a 100 percent volume
variability as there is no substantial evidence to support such attribution. Rather, it says, the costs must
be treated as institutional costs. Periodicals Mailers Brief at 40-I.
The Periodicals Mailers fail to
recognize the substantial record evidence in this case supporting the 100 percent variability and do not
give adequate deference to the Commission’s long-standing policy of mail attribution.
Bozzo, Direct testimony,
USPS-T-15, Table 6 at 119.
Docket No. R2000-1
-27-
that he just did not have the expertise about certain specific activities in mail processing
to know
whether
the recommended
percent.
Tr. 46E/22089.
Some of the alternative
variabilities
“between”
for the letter and flat sorting operations
regression
were
substantially
results computed
100
by witness Bozzo
for the most significant
cost
pools (BCS, OCR, FSM, MANL, and MANF) generally
well above 90 percent.37
The
variabilities
different from 100 percent.
from those regressions
the vast majority of mail processing
close to 100 percent,
easily understood
have variabilities
less than
are not significantly
costs relate to cost pools with probable variabilities
there is no justification
variabilities.
of computational
Variabilities
technique.
to as high as 100 percent,
Nevertheless,
witness
policy that is so
re-modeling
to witness Bozzo’s model would change the
change with the choice of variables
but the true impact of corrective
could increase
Even witness
Bozzo’s
and the choice
would not be increased
adjustments
even the low variabilities
is unknown.
recommended
by
are now in the 70 or 80 percent range) upwards so
that they would probably not be significantly
37
the current
Witness Bozzo claims variabilities
Bozzo (the most significant
standard
to change
and applied.
The record is clear that adjustments
computed
If
recommended
errors, are either not significantly
different from 100 percent.
variabilities,
with a distribution
of two
different from 100 percent or very close to
Bozzo. Direct Testimony, Appendix E, “Principal results from the ‘between’ regression model
Table E-l. This is also true for Table E-3 at 155 in the same appendix except that the small LSM pool has
an estimated variability of 87.3 percent, extremely close to 90 percent. See also. witness Neels’ Direct
testimony, Tr. 27/12834-35, 1283940.
Docket No. R2000-1
-28-
being not significantly
variabilities
different from 100 percent.
The following chart compiled from the
with two standard errors included in witness Bozzo’s testimony
cost Pool
Standard
Error
Lower
Range
Range
0.895
0.751
0.817
0.954
0.772
0.735
0.030
0.038
0.026
0.022
0.027
0.024
0.835
0.675
0.765
0.910
0.718
0.687
0.955
0.827
0.869
0.998
0.826
0.783
USPS-T-15,
In addition,
contention.
other record evidence
12834-43.
those determined
is directly
contrary
Witness
Neels also concluded
to the Postal Service’s
in excess of 100 percent.
that variabilities
higher than
Mailers’ proposal for selecting volume variability estimates
of values
between
the pooled
and fixed effects
models
Mailers Brief at 45-48.
The models are incorrectly
bracketing
of estimated
ignores the ranges of the point estimates
estimates
“between”
variabilities
the 99 percent
confidence
bands
specified,
based
is flawed.
Periodicals
each model:
Tr.
by the Postal Service cannot be ruled out. Tr. 27/12839.
The Periodicals
on the range
I
Table 6 at 119.
For instance, witness Neels found variabilities
27/12807-08,
Upper
Variability
Point Est.
BCS
OCR
FSM
LSM
MANFLATS
MANLTRS
Source:
is illustrative.
and such a
for many of the volume
within
variability
are very close to 100 percent volume variable in the case of the pooled and
models,
Periodicals
variability goals is wrong.
Mailers’
selective
choice
of models
to advocate
- 29
Docket No. R2000-1
For these reasons,
computing
the Postal Service is in error if it expects that, by merely
volume variabilities
of less than 100 percent,
the Commission’s
long-standing
as an estimator
is subject
database.
should
Accordingly,
not be accepted
policy.
This is particularly
to so many deficiencies
the variability
it necessarily
studies
by the Commission
justifies altering
true where the model used
and relies on a questionable
recommended
as support
by the Postal Service
for altering
its 100 percent
volume variability policy.38
B.
The Postal Service Has Not Demonstrated
“Unambiauouslv”
Less Than 100 Percent
The Postal
Kingsley
in support
Service
has presented
of various
aspects
witnesses
did not collectively
unambiguously
succeed
presented
Degen,
of the volume
Service’s initial brief details how their testimony
Although each of the witnesses
that Mail Processing
study.
Greene,
that
study.Jg
relevant to mail processing,
mail
processing
and
The Postal
supports the volume variability
information
in establishing
variability
Bozzo,
Costs Are
variabilities
they
are
less than 100 percent.40
Witness Degen, “who has studied Postal Service operations
since 1983” (USPS
Brief at V-l 1) is cited as having provided an analysis of Postal Service mail processing
38
Just as the Commission and the Chief Examiner found in Docket No. R71-1, the statistical
analysis in this case must be rejected in favor of the 100 percent volume variability because no party has
presented a viable alternative. USPS-T-15 at 8.
39
The Postal Service Brief discusses the testimonies of Degen at V-11-21,
Greene at V-35, 45, 52 and 55-59, and Kingsley at V-31 and 37.
40
Bozzo at V-23-62,
Postal Service witness Greene testified that he really could not say that the data analyzed in the
studies contain the specific value of volume variabilities. Tr. 46E/22086. He also confirmed that from the
testimony he has read, a specific value for any of the activities has not been conclusively determined from
this record. Tr. 46E122085.
Docket No. R2000-1
operations
- 30 -
and concluding that mail-processing
variable.
ld. Actually, witness
costs are less than 100 percent volume
Degen’s testimony
is largely of the descriptive,
anecdotal
genre.
Witness
Degen
provided
comprehensive
operations,
focused
operations,
and the effect of the characteristics
changes
10-54.
on mail flows in the processing
presenting
Degen’s
“the operational
analysis
foundation
that network and location-related
”
describes
of
mail
processing
plants, the characteristics
on the ability of the workers
in volume with less than proportional
Witness
descriptions
changes
operating
in workhours.
practices
for volume variability.”
of the
to absorb
USPS-T-16
at
on the factory floor,
Witness Degen “explains
factors affect costs, but do not change with volume
In short, he ignores changes
in activities
of mail processing
in a mail processing
plants.
facility and the
addition
and/or modification
Clearly such an analysis can
correctly
identify costs over a relatively short period of time (at the lower limit, less than
a day), but is irrelevant for the longer run analysis of cost causation.
Likewise,
operations
in the case of gateway staffing, witness Degen explained
are generally capable of more throughput
The time spent waiting for mail in gateways
Rather,
handle.”
it will tend to be converted
USPS
than the downstream
does not increase
into processing
Brief at V-16, citing USPS-RT-5
would
not require
is more mail to
at 14, Tr. 36/17319.
more work hours, as Dr. Neels implies,
operations.
as volume increases.
time, as there
Service states that, “Mr. Degen does not claim that a sustained
“that gateway
increase
The Postal
in workload
but rather describes
the
means by which the change in work hours need not be 100 percent volume-variable.”
USPS Brief at V-17.
The Postal Service also references
both witnesses
Stralberg and
Docket No. R2000-1
-31-
Kingsley to corroborate
many of the facts cited by witness Degen.
The amount of effort devoted to discussing
and sorting leads to the conclusion
operating
that witness
windows,
USPS Brief at V-18.
setup and breakdown,
Degen is considering
One would expect to see such costs have less than 100 percent
short-run
costs.
volume variability;
however, they are not the relevant costs which should be analyzed.
On a short-run
basis, witness Degen is obviously
have less than 100 percent variability.
correct in assuming
In fact, his analysis is descriptive
that costs
of a one day,
short-run time frame.
Contrary
to the
Postal
Service’s
claim,
of witness
Degen, cited to support the Postal Service
unambiguous.
The conclusion
argument,
“volume-variabilities
is
supplied).4’
Note that witness Degen’s testimony
variabilities
are “generally”
that a further qualification
less than 100 percent.
lies within witness
states that witness Degen concludes
less than 100 percent.
This testimony
100 percent.
witnesses
41
are genera//y
witness
Degen’s
less than 100 percent”
ambiguously
Degen’s
testimony.
unambiguously
just the opposite:
are not certain of the conclusions.
Postal Service Brief at V-II. citing USPS-T-16 at 51
(emphasis
hedges, stating only that
The Postal Service
it is “likely” that “volume-variabilities
does not demonstrate
is not
The Postal Service notes in its initial
_‘I USPS Brief at V-l 1, USPS-T-16
In fact, it demonstrates
testimony
are generally
at 2.
that variability
is less than
that even the Postal Service
- 32 -
Docket No. R2000-1
1.
The
Postal
The Postal Service has not demonstrated that setup and takedown
times will not increase oroportionatelv over the longer run
Service
proportionately
contends
that
with increases
times cannot possibly
setup
and
takedown
in mail volume and thus concludes
vary 100 percent with mail volume.
OCA would agree that the nature of mail processing
setups and takedowns
basis.
Witness
associated
times
may lead to variabilities
Kingsley
sorting
operations.
not increase
that mail processing
USPS Brief V-14-5.
sort schemes
The
accompanied
by
less than 100 percent on a short-run
is quoted on the unavoidability
with bundle
will
of setup and takedown
USPS-T-10
at 21.
Witness
times
Stralberg
indicated that with larger volumes the runs of each scheme would be longer. This would
result in the cost of setups
However, those situations
Witness
takedown
24.
Neels
being spread over additional
described
has
are of a short-run
contested
witness
mail pieces. Tr. 38/17280.
nature.
Degen’s
contention
times insure that volume variability is less than 100 percent.
Before responding
Neels, it is necessary
to the Postal Service’s
to first emphasize
included in Neels’ testimony.
specific
comments
that the straightforward
that
setup
and
Tr. 27/12820refuting
witness
and direct answer
is
The Postal Service is looking at the short run rather than
the longer run. If one machine requires one hour of setup, then as volume increases
so
that a second
Tr.
machine
is required,
two total hours of setup will be required.
27112822-24.
The Postal Service cites three aspects of witness
Neels on the setup and takedown
too much or miss the point.
times.
Degen’s responses
None are responsive,
The argument
to witness
as they either concede
of the Postal Service is that setup and
-33-
Docket No. R2000-1
takedown
follows,
times involve relatively fixed costs that do not change with new volume.
goes the argument,
proportionately.
that when volumes
increase,
It
all costs do not increase
The cost of the same number of setups would be spread over more
mail pieces.
The Postal Service agrees that setup times do sometimes
due either to response
deployment
of new automation
beside the point.
other factors.
changes
12.
to “growth”
in deliveries
and to the
which is “not always” tied to volume changes.
But this is
witness
that setup times never change
Degen
admits that setup times are responsive
In fact, some setup times do change
occur due to changes in volume.
than normal.
For example,
during ramping
establishes
times.
probably
In fact,
to some new deployments
to find a proportionate
that
increase
experience
does not necessarily
suggests
more
mean
complicated
has, in most cases, longer setup times.@ The Postal Service also
automated
or mechanized
volume where they had not previously
processing
activities
been established.
to meet increased
Tr. 27112781.
new setup times would clearly lead to a more than proportionate
42
at
could actually be greater than
New technology
common
USPS-RT-5
to
up periods, setup times can be greater
Also, the setup times for new equipment
setup
equipment
in response
It is not unreasonable
the setup time for the older equipment.
shorter
in response to
in volume, even if that is not always a reason for the change.
in setup times.
but that it is
more than to mail volumes
No one is contending
However,
change,
mail
The addition of
increase in setup and
Witness Neels depicts the cost effect of this process very clearly in Figures 1 and 2 of his direct
testimony, Tr. 271704-86.
Docket No. R2000-1
takedown
- 34 -
time for those activities
and is a reflection
of the impact of the longer-run
outlook.
The Postal
Service
increase in a stepwise
any other particular
refutes witness
fashion.
degree
Service’s statement
that setup
of variability.”
USPS Brief at V-14.
times in response
increase
in a stepwise
(emphasis supplied),
Neels’
to continuing
fashion
studies do not demonstrate
may
In fact, the Postal
that: “[rleplication
On the same
of setup
and
growth could create a situation in which costs
in direct proportion
citing Tr. 27/12822.
argument.
statement
times
imply 100 percent or
is in direct conflict with witness Neels’ own testimony.
takedown
to volume.”
USPS Brief at V-14
Thus, Neels’ testimony specifically
refutes the
The Postal Service further claims that witness
Neels’ own
variability
Again the Postal Service confuses
underlying
suggestion
It says this “does not specifically
page of its brief it cites witness
Postal Service’s
Neels’
is less than 100 percent.
the issue.
As witness
USPS Brief at V-14.
Smith has often said, the
models are flawed, and this is true even of witness Neels “alternative
level analysis” which will not measure the appropriate
longer-run variability.
shape-
There is no
reason to expect that the flawed model would produce 100 percent variabilities.
The Postal Service also says, in support
increase in proportion
takedowns
of its theory that setup times do not
to volume, that the direct determinant
is the “need to change mail processing
also states that “[nletwork
simple relationship
sort schemes,
not mail volume.”
It
affect the number of schemes, and there is no
between volume and scheme changes.”
Again this defense
analysis far enough.
considerations
of the number of setup and
misses the attack.
It admits that network
USPS Brief at V-14-5.
The Postal Service does not carry its
considerations
also affect the number
of
Docket No. R2000-1
- 35 -
sorting schemes, but forgets to recognize
that those network considerations
can be caused by increases
in volume.
instances where there are increases
they ignore situations
may increase
where
time increases
prevented
the setups
when necessary.
not necessarily
times may be volume related and
For instance,
disproportionate
could occur if there were mail processing
or takedowns
Also, additional
from occurring
and disproportionate
for particular
bottlenecks
machines
setup and takedown
being added in response
setup and
to new volumes.
that
precisely
times can
This may
last only during the ramping up phase, but could occur permanently
new, more complex equipment
2.
process.
in setup schemes that are not volume related, but
setup and takedown
be required to run new equipment
Witnesses
The Postal Service also points to individual
more than proportionately.“’
takedown
themselves
Stralberg
as
may require longer set up times than older machines.44
Witness Degen’s explanation of how mail is processed
prove 100 percent volume variabilitv
and Degen present
interesting
information
does not
on the mail handling
The Postal Service states:
Mr. Degen and Mr. Stralberg are both convinced that hours in mail
processing operations do not vary 100 percent with volume, and, along
with witness Kingsley, have cited numerous facts in support of that
conclusion. The Commission should accept their well-founded conclusion
that mail-processing
costs are not 100 percent volume-variable.
USPS
Brief at V-21.
43
It is well settled that Postal Service equipment
27112777-82.
44
is added to meet increases
in volumes.
See Tr.
Even if setup times on a new machine are longer, it does not follow it is less efficient.
throughput may be so much greater that the unit costs will decrease.
The
- 38 -
Docket No. R2000-1
In fact, the witnesses
have, instead,
focused
made a presentation
on the short-run
accommodations
analyses
have not presented
an analysis of volume variability;
on how mail is processed.
time frame, generally
addressing
made during a mail processing
shift.
from those of the longer run. The Commission
mail processing
costs are unambiguously
3.
analysis
is misleading
Brief at V-130.
corrected.
the types of operational
These are very different
cannot fairly conclude
that witness
significantly
One can turn to witness
cost
that the
Neels’
as to the level of variabilities
It is clear that neither witness
volume
Neels’ “own alternative
lower than 100 percent.”
direct testimony
well in excess of 100 percent for 7 of 8 cases.
can only speculate
are
less than 100 percent.
in contending
of those data yields variabilities
variabilities
Their comments
Volume variability is not unambiguously
100 percent
variable when witness Neels’ Presentation is recoqnized
The Postal Service
they
USPS
and find activity
Tr. 27/12834,
12840.
One
if witness Bozzo’s flawed analysis were
Neels nor witness Smith agree with witness
Bozzo’s conclusions.
The Postal Service maintains
record points to mail processing
by operation.”
that “[a]11 of the reliable statistical
variabilities
USPS Brief at V-23.
The Postal Service references
Bozzo’s direct testimony.
models
original
theoretical
witness
deficiencies
Neels.
statistically
of witness
mentioned
Bozzo’s
in witness
USPS Brief at V-25.
analysis
and
Smith’s testimony
The Postal Service has designated
reliable.
in the
being less than 100 percent and differing
the Notice of Inquiry No. 4 and in witness
are variants
evidence
various models in
In fact, all of the
are subject
to the
and also discussed
by
the pooled model results as being
Even using the pooled model, subject to the
- 37 -
Docket No. R2000-1
theoretical
deficiencies
of witness
Bozzo’s approach,
six of the ten variabilities
(BCS,
OCR, FSM, LSM, MANF, and MANL) are within, or close to, two standard deviations
90 percent.
90 percent,
purposes,
Postal
Loosely stated, the pooled model suggests
Dr. Greene
be considered
Service
has indicated
that 90 percent
to be 100 percent.
presentation
does
volume variabilities
variability
Tr. 46E/22088.
not strongly
support
of
of at least
could, for practical
Put differently,
variabilities
even the
less than
100
percent.
Furthermore,
it should
be noted that the Postal Service’s
reference witness Neels’ analysis that attempted
witness Bozzo’s study.
As previously
of 100 percent for some operations.
C.
USPS
Brief
at V-26.
by witness
less than 100 percent
precludes
In fact,
from
is
results of 100 percent
a theoretical
viewpoint,
Bozzo, has been cited for a variety of major
Bozzo’s database
of high R2 equations
that is theoretically
is inevitable.
has been scrubbed
In addition,
witness
to the point that the
He utilizes a microeconomic
flawed and which produces short-run
(one year or less), a matter of increasing
four years.
labor variabilities
Witness Bozzo’s Analvsis Is Incorrect
for instance, witness
production
in excess
well in excess of 100 percent are also possible.
The study, developed
flaws:
in
The Postal Service states, “[a]11 of the witnesses
Nothing in Dr. Bozzo’s modeling approach
volume-variability.”
variabilities
to correct some of the deficiencies
stated, witness Neels found variabilities
agree that the result of mail processing
possible.
initial brief did not
estimates
methodology
of volume variability
concern given that rates are in effect for two to
Bozzo uses an estimating
procedure
that ignores
Docket No. R2000-1
important
-38-
variables
exogenous
(such as capacity
utilization),
treats
endogenous
(such as the Manual Ratio and QICAP), and is conducted
variables
as
at the wrong level
of analysis (the activity rather than the plant or system level). This ignores the long-run
expansion
path and still has not resolved the issue of whether the fixed effects are truly
fixed.
1.
The appropriate
manner to scrub the data remains unresolved
The Postal Service has stated that “Dr. Neels does not expressly
Degen’s analysis is incorrect
,” Brief at V-13.
find that witness Bozzo’s study is deficient.
coefficient
estimates
data scrubs
substantial
However, witness Neels does, in fact,
He concluded that some of witness Bozzo’s
were subject to error in variables bias, even after witness
had been performed
Tr. 46EI22336.
issues with the specification
He also concluded
of the model and the approach
they don’t take into account
,” Tr. 46E/22356.
good solid consensus
out there, or something
would warrant
alone
appropriate
estimate
The uncertainties
sufficient
that “I have
and the things
“And I frankly don’t think there is a
that is robust enough
that it
to warrant
surrounding
rejection
the appropriateness
of witness
Bozzo’s
Tr.
of the data scrubs
variabilities
until the
handling of the data can be resolved.
2.
A short-run analysis is irrelevant
The Postal Service has again maintained
appropriately
Bozzo’s
a decision to move away from the 100 percent volume variability.”
46E/22356-57.
are
claim that Mr.
that, “[t]he variability
‘short run’ to fill its role in the Postal Service’s
USPS Brief at V-40.
The Postal Service brief discusses
analysis
must be
overall cost analysis.”
that labor planning
can occur
Docket No. R2000-1
-39-
on a daily basis, or over a year, thereby advocating
Brief at V-39.
A SPLY analysis,
mail processing
operation
is a cyclical analysis--looking
at where the
stood at the same time a year ago. The OCA has shown that
the longer run encompasses
a period bounded
however,
the use of a SPLY model. USPS
at least a period as long as the rate effective time period,
at the lower end by a time frame of the several years between
rate
cases.
The Postal Service witnesses
their
short-run
discussed
by witness
are different;
variability.
estimator
is identical
Smith.
have confused
to that
the discussion
associated
Tr. 38117384, 27/13189,
Witness
Smith testified, “Finally, the modeling
on a long-run basis, focused on the facility expansion
analysis of mail processing
investment decisions
costs.
that
path
In fact, the two
They do not have the same
should have been performed
path.” Tr. 46E/22367.
In short, the Postal Service has not presented
permits adequate
with the expansion
46E/22199-00.
one is short run and the other is long run.
by claiming
the type of cost analysis
that
Relevant volume, operational,
and
occur over a span of two to four years, the relevant time frame for
a rate case.45 Allowing
for some overlap with rates on both sides of the analysis,
the
relevant time period for analysis is two to four years.
45
Deputy Postmaster Nolan’s August 21, 2000 interview by PostCom indicated “We do not have a
plan to implement new rates every year, every two years, or every five years. There is no such plan
.We don’t have a specific plan to raise rates every two years or anything like that
I think people
would like more frequent, more predictable rate cases but they don’t want them every year.” USPS LR-I492 at 3.
- 40 -
Docket No. R2000-1
D.
Witness Bozzo’s Mail Processing
Economic Theorv
The Postal Service continues
Model Is Not Based on the Appropriate
to be wrong on the underlying
economic
theory.
The Postal Service maintains that the labor demand function should be specified at the
activity
rather than facility
level.
USPS
Brief at V-29.
contrary to the Postal Service’s own arguments
plant, that the continued
use of the manual
based on the interrelationship
of activities),
This contention
that workers
is directly
can be shifted around the
ratio is maintained
(the manual ratio is
and that the capital investment
variable,
QICAP, is based on facility wide, not activity wide, capital.
1.
Witness Bozzo has not selected aooropriate
It is stated that witness
gap between
generic theory
Bozzo identifies “the factors that sufficiently
and operational
ignores capacity utilization (a potentially
been
cited as a possible
described
driver
Bozzo analyzed
equipment
USPS Brief at V-30.
important cost driver).
of costs
(Tr. 27/13185)
cost causation,
Capacity
Yet he
utilization has
and witness
Degen
letters arrive (thereby
idle time) on a short-term
the longer-run
bridge the
basis.
has
using
In neither case has
the period over which capital
decisions are implemented.
Witness
Bozzo has also ignored the use of a simultaneous
as needed for the estimation
of the labor demand function.
not cause network characteristics,”
the configuration
sortation
reality.”
how labor usage rises and falls as additional
capacity) or do not arrive (creating
witness
variables
of machines
requirements.
equations
approach
In stating that “volume does
USPS Brief at V-31, the Postal Service ignores that
in the internal plant network
is largely driven by mail
-41 -
Docket No. R2000-1
2.
Network and location-related
factors do change with volume, but
that fact is not reflected in the assumptions underlying the Postal
Service model
The Postal Service’s
location-related
factors
that determine
Bozzo’s
volumes,
appropriately
characteristic
of the configuration
no consideration
recognize
Degen’s
that network
Postal Service’s
Brief at V-12.
is the number
of networks
of delivery
factors
In fact, witness
points.
There
is no
plant based on mail
and plants to the network
of the interaction
to
of ‘mail processing
is minimal.
of national
volume
related factors change
growth
fails to
with volume.
The
Initial Brief states very clearly that this is Mr. Degen’s position.
USPS
This is contrary to common sense and ignores the interrelationship
of the
in the network
Tr. 27/12793.
Volume growth at the local level can result in
both within the facility and across facilities.
volume grows within a plant so that expansion
plant modifies its facilities, the operations
For instance,
to an annex is necessary,
there will be an impact on the internal network operations
constructed
The Postal Service states
volume and non-volume
of machines
and
at least, the Postal
in a mail processing
of the effects
and other location
that network
USPS Brief at V-37.
studied.”
of machines
analysis
activities within the facility.
changes
assumes
(although,
distinguishes
and no consideration
In short, the discussion
Witness
with volume
of the addition
changes,
erroneously
that these factors affect costs).
costs in the operations
meet volume
plants.
analysis
only network
consideration
analysis
do not change
Service does now recognize
that “Dr. Bozzo’s
network
must be modified.
of a particular
as
most certainly
facility.
As a
Likewise, if a new facility is
as a result of local volume growth to eliminate a processing
particular facility, the old facility may modify its network or coordinate
bottleneck
at a
its processing
with
-42-
Docket No. R2000-1
the new facility in a way that modifies the network.
Witness Smith discusses
All this is a result of volume growth.
this in his testimony:
Mail processing activities and sites do not stand alone in terms of the
network of originating and destination nodes. There seem to be three
types of network issues. First, there is the intra-plant network of activities
that feed mail to each other. One gets the impression that this network
could change based on a variety of factors, including network volumes. A
second type of network effect is apparently the delivery configuration of
the service territory. Dr. Bozzo measures this network configuration with a
variable measuring the number of possible deliveries. Finally, the position
of the plant in the mail flow between other mail processing plants also
seems to be a type of network relationship. According to an interrogatory
response, the size of facilities and their mail processing
operations
depends not only on the volume of mail processed, but also their position
in the Postal Service’s network. Tr. 27/l 3193 (footnote omitted).
Witness Neels also discusses
part in the response
The omission
networks
and notes that new plant construction
of the Postal Service to an increase in mail volume.
of these variables
is significant;
for as witness
stated, “Omitting relevant variables from a regression
3.
plays a
Tr. 27/12790.
Bozzo has so clearly
leads to bias.” Tr. 46E122154.
In maintaining that TPF is a proper measure of mail processing
volumes, the Postal Service ignores that TPF is not an accurate
measure of the volume of mail
In maintaining
that TPF is a proper measure
of mail processing
volumes,
the
Postal Service ignores that TPF is not an accurate measure of the volume of mail. TPF
may be measured
volume of mail.
with great precision,
Various
criticisms
of the precision
have been made; such measurement
However,
the inability
to measure
inability to measure accurately.
but it is not accurate
in portraying
with which FHP can be measured
is clearly open to improvement
precisely
the actual
the amount
See USPS Brief at V-32.
if necessary.
of mail is different from the
-43.
Docket No. R2000-1
The Cross Sectional
Differences Between
E.
The Postal
sectional
model.
Service
“Between” Model Is the Only Model that Considers
Facilities Based on Size and Volumes
correctly
USPS Brief at V-34.
mention that the cross-sectional
current case that adequately
the statistical
problems
with the cross-
However, the Postal Service conveniently
model is the only one offered by witness
considers
on different sizes and volumes.
notes
the differences
The Commission
fails to
Bozzo in the
in costs between facilities based
has discussed
the desirability
of such
an approach:
In particular, the rate cycle production period implied by the theoretical
framework
described in the previous section requires an estimation
procedure which relies on persistent differences in the mode of operation
across facilities, rather than differences in mode operation within the same
facility over short time horizons. Differences in the operation of facilities of
different sizes would more accurately reflect the relationship between
volume and costs that would occur over a production period as long as a
rate cycle than would differences in the operation of the same facility over
adjacent time periods.
Consequently,
an estimation procedure which
primarily relies on the cross-sectional
dimension of the panel dataset is
preferred to one that relies on differences over time within the same
facility, such as the fixed-effect estimator. PRC Op. R97-I, Appendix F at
13-14.
The volume variabilities
statistically
Recognizing
different
from 100 percent
that the cross sectional
from the “between”
model are, in general,
not
in many cases, and are high in other cases.46
“between”
model shares many of the deficiencies
of witness Bozzo’s other models, it has been proposed as the “least bad” model, not as
a good
46
model.
The variabilities
Problems
in specifying
a “good”
model highlight
are in Appendix E of witness Bozzo’s Direct Testimony,
the need for the
USPS-T-15,
-44-
Docket No. R2000-1
formation
of a working
group to resolve modeling
and other issues in a collaborative
environment.
F.
A Working Group To Address
Problems Is Recommended
Numerous
deficiencies
Modeling,
Economic
and estimation
this effort,
it would
Accordingly,
problems.
be desirable
for this reason
to salvage
modeling,
as much
of the work
the establishment
to
as possible,
of a working
modeling of mail processing
THE COMMISSION
SHOULD APPLY
ITS ESTABLISHED
ATTRIBUTION
METHODS TO THE ES-BASED ESTIMATE OF ACCRUED LOAD TIME COST
The OCA has proffered a consistent
carrier load time costs in this proceeding.
approach
flowing
addressed
Given the amount of effort and time devoted
OCA has recommended
group to consider the appropriate
Ill.
Estimation
in witness Bozzo’s study have been outlined in testimony
and the OCA’s initial brief. The Postal Service has not adequately
economic,
and
includes
from
Standards
the
the following
carrier
street
and reasonable
As outlined
components:
initial brief, this
(1) use the accrued
load time costs
time proportions
developed
the Load Time Variability
study (“LTV”) regressions
and (3) attribute
load time costs after elemental
(i.e., coverage-related
costs)
single subclass
This approach
stops.
street time percentages
from the
Engineered
(2) apply the elasticities
to estimate
based on the proportion
combines
city
in the 004’s
(“ES”) as the basis for the load time analysis;
residual
approach for attributing
elemental
from
load time;
costs have been estimated
of mail volumes
delivered
to
the use of updated and fully-reviewed
and related accrued load time costs with the application
of well-
-45
Docket No. RZOOO-1
established
methods for assessing
them to appropriate
A.
the volume variability of these costs and attributing
mail classes.
The Commission
Should Accept
Carrier Street Time Proportions
the Postal
In this case, the Postal Service has presented
percentages
used in the segment
completed
ES study.
Postal
supporting
testimony
submitted
convincing
arguments
in support
technology
to record these activities.
purposes
and represent
Kent,
Tr. 39/17877-88,
along with
provide
USPS Brief at V-
are reflective of current carrier street activities
study that employed
Although
street-time
an improvement
on the recently
and Baron,
proportions.
City
of the street-time
Raymond
of the new street-time
two-year
from the record that ES-based
witnesses
ES-Based
that are based
by NAA witness
70 to -80. In particular, these proportions
and based upon an extensive,
new estimates
7 cost analysis
Service
Service’s
advanced
data collection
certainly not without flaws, it is apparent
percentages
are reliable
over the percentages
for ratemaking
derived from the 1986
Street Time Sampling study.
B.
The Commission
Derive Elemental
Participants
varying alternatives
Should Continue
Load Time
in this proceeding,
for estimating
to ES-based
47
Additional
load time.
For example,
Baron, argued for applying
accrued load time costs to estimate elemental
Response of witness
Designations,
Elasticities
To
primarily the Postal Service and Advo, et al., offer
elemental
the Postal Service, through witness
To Employ LTV-Based
in its direct case,
LTV-based
load time costs4’
elasticities
The OCA
Baron to lnterrogatorry UPS/USPS-T12-13,
Tr. 46D121098-99; MPA
August 16, 2000. The Postal Service neglects to mention this in its initial brief.
-46.
Docket No. R2000-1
agrees with the Postal Service that it is appropriate
conjunction
to the filing and defense
after the filing of intervener
of witness
direct testimony,
however,
Service changed
the Postal Service’s proposal.
the Commission
to abandon
developed
by witness
Commission
chooses
the LTV regressions
the use of the ES-based
of these
parties
The Postal Service
specification
at V-81.
This
point
strong-arming
the regression
compared
fails
to recognize,
the
record
is not
into a state of apparent
Furthermore,
comprehensively
surprising,
it is important
however,
to recognize
and quantitatively
however,
nonsensical
regression
that
are
provides
regressions.
the
results.
original
USPS-LRproffered
by
data or simply a product of
reasonableness.
party filed testimony
regression.
given
regression
the final regression
with the underlying
of the ES-based
if the
Advo et a/.
to the LTV-based
Advo et a/. point to the fact that no intervenor
to the application
but only
percentages.
that the ES-based
feasible
correlated
USPS Brief at V-80-81.
of the ES-based
The record does not indicate whether
witness Baron is appropriately
regression
(“Joint Brief’) at 17.
of the model and related data produced
l-310, pp. 7-9.
and
Baron and the Postal
regressions,
street-time
in support
argues
results that are more operationally
Brief
witness
in favor of an alternative
City Carrier Cost Attribution
The arguments
opposition
in
The Postal Service is now encouraging
to employ the ES-based
Joint Brief Concerning
unpersuasive.
Baron’s direct testimony
Baron that is based on the ES data.
et al. also support
USPS
elasticities
with the ES data.
Subsequent
Advo
to use the LTV-based
the
Joint Brief at 17. This void in
late arrival
that no intervenor
why the Commission
expressing
of the
proposal.
filed testimony
showing
should employ the ES-based
-47.
Docket No. R2000-1
regression.
witness
Indeed, even the primary intervener
Crowder,
considered
witness
admits that she has not investigated
possible
alternative
specifications.
advocating
for its use, MPA
the ES model thoroughly
Tr. 32/16224;
nor
see also OCA Brief at
135-37.
In sum, the ES-based
assessing
the volume
regressions
regression
variability
with a route-level
represents
of load
analysis.
costs
an entirely
new methodology
by replacing
the stop-level
In assessing
its application,
for
LTV
the Commission
has nothing but a mixed bag of interrogatory
responses
submitted
While OCA agrees that it is important
by witness
Baron to rely upon.
update the LTV regressions,
the
ES-based
regression
the evidentiaty
for
ratemaking
to
record clearly does not support the use of
in this
Commission
should follow the general approach
case-apply
the LTV-based
elasticities
and evolving library references
Accordingly,
proceeding.
the
used by the Postal Service in its direct
to the ES-based
accrued
load time costs to
derive elemental load time costs.
C.
The Commission Should Continue To Attribute
Time Usino Sinole Subclass Stop Ratios
For the past four rate proceedings,
the Commission
portion of load time costs (i.e., those costs remaining
calculated)
Coverage-Related
has attributed
after elemental
on the basis of the single subclass stop methodology.
the Postal Service purports to have shown why this approach
V-82-97.
Its argument,
however,
depends
attributed
on a marginal cost basis.
upon the conclusion
Load
the residual
costs have been
In this proceeding,
is flawed.
USPS Brief at
that costs can only be
As such, the Postal Service attempts to develop a
Docket No. R2000-1
- 46 -
method for assessing
“residual”
approach
The Postal
the volume variability
of coverage-related
costs and critiques the
from that perspective.
Service’s
method
fails in two critical ways.
First, it clings to a
worldview
of cost attribution that has long been refuted in Commission
precedent
firmly establishes
that the Commission
volume variability analysis in identifying
obligation
of residual
coverage-related
technique
for establishing
It is irrefutable
are caused
variability
shown
by that subclass.
can be reliably identified.‘@ In the case
has identified
these costs and particular subclasses
Second,
analysis of coverage-related
that witness
flawed and incorrect.
and even if one assumes
load time is appropriate,
Baron’s
efforts to construct
Tr. 25/12032-43.
should not depart from the established
on
has a statutory
a useful
of mail.
that costs incurred to deliver a single subclass of mail to a particular
convincingly
definition,
and furthermore,
load time costs, the Commission
a link between
This
is not obligated to rely exclusively
cost causation,
to attribute costs when their causation
precedent.
that a volume
OCA witness Ewen has
such an analysis
For these reasons,
method of attributing
stop
are, by
the Commission
residual coverage-related
costs.
IV.
COST COVERAGES
OF 105 PERCENT
The Postal Service
subclasses
48
FOR SUBCLASSES
has requested
of mail. The Commission,
SHOULD
low cost coverages
for certain
however, should resist the temptation
to set rates
See, e.g., PRC Op. R94-1, m3095-3152.
extremely
BE SET AT A MINIMUM
-49.
Docket No. R2000-1
at levels so close to attributable
such
costs.
Attributable
Consequently,
attributable
costs that the revenues
cost
that
underlie
no one can ever know with certainty
whether
costs of a class or subclass.
changes in costing methodology
a long
list of costing
consideratior?
changes,
estimates
may not actually recover
are
not
perfect.
rates do in fact cover the
This problem is exacerbated
occur.
changes
rates
all
when last-minute
In its initial brief, the Postal Service has set out
that
it either
or “would not challenge.“50
commends
If the Commission
it should also set cost coverages
for affected
for the
Commission’s
should adopt any of these
subclasses
reasonably
above
100 percent.
The Postal Reorganization
attributable
costs.
39 U.S.C. § 3622(b)(3).
make a “reasonable”
ld.
Act requires
contribution
The Commission
The Act also requires each subclass
this section
preclude
PRC Op. R94-1 ~ 3084-92.
that cost coverages
percent.
it be said that
requirements
then
can
with a reasonable
USPS Brief at t-18 to -28
54
Id. at 24.
to
cross-subsidization.
has also recognized
that the
it from meeting these standards
The resolution
statute and reality requires
Only
as prohibiting
However, the Commission
realities of cost and revenue estimation
absolute certainty.
of mail to bear its own
toward covering the other costs of the Postal Service.
has interpreted
PRC Op. R97-1 77 4026-27.
each subclass
degree of certainty.
of this conflict between
be set reasonably
postal
rates
with
and fees
in excess of 100
meet
statutory
Docket No. R2000-1
- 50 -
Hardly any expense of the Postal Service can be traced directly from the books
of account
through
to a particular
subclass.
And it is impossible
the postal network to determine
each piece from origin to destination.
systems to determine
witness
the exact value of the resources
Consequently,
And the Commission
needed to get
the Postal Service uses sampling
the costs of the various classes and services.
Fronk testified, “Sampling
34/16583.
to follow each piece of mail
error usually accompanies
Unfortunately,
as
estimates.”
Tr.
sampling
itself has taken note
of the general proposition that cost coverages near 100 percent provide
The
little leeway in terms of meeting the attributable
cost floor.
Commission is aware that very low markups might, in fact, result in rates
below attributable costs. It is not possible on this record to adopt a formal
statistical technique to use as a guide. The Commission, however, has
avoided extremely low markups whenever possible.
PRC Op. R94-1 7 3088.
Data collection
or analysis
is not a perfect science.
IOCS or the CCS or any other cost data system.
computers
can be programmed
incorrectly,
Data can be recorded
In
this
nonsampling
proceeding,
In spite of its own warnings,
changes
hastily
constructed
Service
the Commission
update provided
Mistakes of these types are
error.
Postal
error and has warned
in the roll-forward
certain
the
incorrectly,
and analysts can blink at the wrong time.
Such errors arise whether one takes a sample or a census.
referred to collectively as nonsampling
Errors can creep into the
in response
has
acknowledged
that undiscovered
to Order No. 1294.
committing
errors may lurk
USPS Brief at l-9.
the Postal Service is urging the Commission
changes
are unlikely to survive
in cost allocation
closer
scrutiny
procedures.
by the Postal
to adopt
Some of these
Service
itself.
The
Docket No. R2000-1
Commission
-51-
should exercise great caution when evaluating
cost allocations,
especially
costing
Service raises the possibility
compromised.
changes
of nonsampling
to a particular subclass.
may benefit
By calling attention to
a particular
subclass,
that the integrity of its cost allocation
Even if no intentional
revised costing procedures
the Postal
systems
were devised should alert the Commission
to the possibility
error.
adjusting extremely
upward in order to account for sampling error.
The Commission
low cost coverages
rejected the proposal
because the data needed to implement
it did not exist.
Commission
of the need to avoid cost coverages
did express
100 percent whenever
probability
its recognition
possible.
Service.
Sampling
higher.
By whatever
the Commission
little to the substantial
error still exists in cost estimates,
institutional
and nonsampling
necessary,
the Commission
have cost coverages
reasonably
in excess of 100 percent.
thus avoiding
assigned a high coverage
at least something
cross-subsidy
probability
should
costs.
error is now
ensure
that
on those
all
A cost coverage
that all subclasses
to the institutional
and unfair burden
of their attributable
faces an
costs of the Postal
means
their true costs and contribute
close to
will lead to rates that do not cover costs.
floor of 105 percent will provide greatly increased
Service,
However, as noted above, the
In the current proceeding,
that low cost coverages
Certainly such rates will contribute
subclasses
has been
bias has crept into them, the haste in which the
In Docket No. R94-1, the OCA proposed
increased
revised
where the Postal Service calls attention to the fact that a
proposed change will reduce attributions
the fact that certain
the Postal Service’s
cover
costs of the Postal
classes
that are
Docket No. R2000-1
V.
- 52 -
FIRST-CLASS
A.
RATE ISSUES
The 33-Cent Sinole-Piece
In this proceeding,
Commission
First-Class
make
arguments
disproportionate
single-piece
the
Major
First-Class
Mailers
institutional
the fact that the institutional
effort
to reduce
the very
high
by retaining the 33-cent
(“MMA”),
while
cost burden, nonetheless
discounts
nor other parties opposing
cost burden on First-Class
is very high by historic standards.
of that institutional
are irrelevant,
single-piece
34-cent rate.5’ Those parties
Mail in this proceeding
Association
33-cent
highlighting
the
supports the 34-cent
to reduce that burden
business mailers.53
Neither the Postal Service,
deficient,
to OCA’s
rate, but seeks an increase in worksharing
for large-volume
existence
to retain the current
to the proposed
objecting
cost burden on First-Class
Similarly,
proceeding
proposal
Letter rate, as an alternative
institutional
rate.52
the Postal Service and certain other parties argue that the
should reject OCA’s
numerous
First-Class Rate Should Be Retained
cost burden,
precluding
current 33-cent rate.
However,
Letter Mail proposed
the Commission
in this
Rather, they attempt to rationalize
maintain that past Commission
or claim that OCA witness
thereby
OCA’s proposal, dispute
Callow’s
testimony
no amount of rationalizing
statements
is somehow
from recommending
or arguments
the
retention
legally
of the
can justify the
Initial Brief of the United States Postal Service at VII-29; Initial Brief of Association for Postal
Commerce and Mail Advertising Services Association International at 9; and Brief of the Direct Marketing
Association, Inc. at 5.
52
In fact, the institutional cost burden on First-Class Letters, as measured by the mark-up index,
exceeds that of Priority Mail-the
Postal Service’s supposedly premium mail service.
53
Initial Brief of Major Mailers Association
at 6
Docket No. RZOOO-1
-53.
very high First-Class
Letter institutional
the current
rate in order
33-cent
cost burden, and the Commission
to reduce
that institutional
should retain
cost burden
in this
proceeding.
Witness
Callow examines
the growth
Class Letter Mail since 1988 as measured
Using an average of the Commission’s
in the institutional
by the mark-up index.
recommended
from the four rate cases during this period, witness
Mail has contributed
Commission
in its past opinions.
are also apparent
Commission’s
cases.
revenues to institutional
of First-
OCA Brief at 144-47.
Letters mark-up index
Callow concludes
that First-Class
costs in excess of that intended
Id. at 152. The excess institutional
when the actual First-Class
recommended
First-Class
cost burden
by the
cost contributions
Letters mark-up index is compared
to the
mark-up index, rather than the average over the four rate
Id. at 152-53.
The Postal Service fails to address the historically
has proposed
percentage
for First-Class
Letters
rate increase for First-Class
high institutional
in this proceeding,
arguing
is below the rate of inflation.
cost burden it
instead
USPS Brief at VII-
30. It suggests the lower rate increase is caused by changes in the First-Class
prompted
automation
by new worksharing
program.
of First-Class
institutional
incentives
and the success
of the Postal
Id. In effect, the Postal Service acknowledges
Mail, but unless such
declining
that the
mail mix
Service’s
the declining
costs are reflected
costs
in the rate, the
cost burden will continue to grow relative to the institutional
cost burden of
other mail classes.
PostCom advances a similar argument.
cost coverages
of single-piece
and presorted
It explains that differences
First-Class
exist because
in the implicit
projections
of
Docket No. R2000-1
- 54.
the mix of mail differ from those actually observed,
coverage to increase.
These arguments
First-Class
institutional
miss the central point of witness
cost burden
continues
institutional
cost
coverages,
inexorably-even
that the
as costs
Moreover, the Postal Service
makes
cost burden to First-Class
an ever
higher
First-Class
Mail-expecting
share
the class of mail
of the Postal
Service’s
costs.
The Postal Service further
flawed
Callow’s testimony
knowing that the mail mix has changed,
used by the public to carry an ever-greater
institutional
cost
cost burden more palatable to mailers. 54 This ignores the overall effect of
shifting the institutional
widely
implicit
to advance
below the rate of inflation.
and PostCom seem to suggest that somehow
in different
the subclass-wide
PostCom Brief at 12.
decline and the rate increases
resulting
causing
argues that witness
in that he uses the test year cost coverages
recommended
by the Commission
each case were in effect. According
Callow’s
analysis
and resulting
is somehow
mark-up
indices
in past cases during each year in which the rates for
to the Postal Service,
the Commission’s opinions only address cost coverages during specific
test years for each omnibus rate proceeding and not the intervening
years. While the Commission’s intentions for the Docket No. R90-1, R941, MC951 and R97-1 test years are known, one can only speculate as to
what cost coverages the Commission might have found acceptable for the
intervening years. USPS Brief at VII-30.
54
PostCom notes that there are implicit cost coverages
subclass of mail. PostCom Brief at 12. This is beside the point.
Commission declined to formally recognize those separate cost
subclass status for single-piece and presort in its Docket No.
Decision, making any discussion of implicit cost coverages moot.
for different rate categories in every
With respect to First-Class Letters, the
coverages by recommending separate
MC951 Opinion and Recommended
Docket No. R2000-1
-55-
The Postal Service apparently
meaning
fails to understand,
of witness Callow’s testimony.
test year as a period that is intended
for) the period that recommended
concept
intentionally
or otherwise,
the
Witness Callow states that he “interpret[ed]
the
to be typical or representative
rates are in effect.”
Tr. 22/10173.
of a test year makes little sense if it is not considered
period during which rates are in effect.
of (i.e., an average
In fact, the
representative
That is, for any given multi-year
of the
period, if the
goal is to “break even” during the period in which rates are in effect, then the rates
based upon test year costs must permit the Postal Service, through
and economical
management,
must represent
an average.55
to “break even” in that period.
Indeed,
a common
make money in the first year, see revenues
during
the third year.
Commission
Consequently,
witness
measure
equal costs in the second, and lose money
for the test year should represent
Callow’s
period the recommended
cost contribution,
the average contribution
recommended
and its expression
33-cent
Commission
55
rate
by the
per year during
mark-up
index
as the
of intent for the entire
rates were in effect.
The Postal Service and DMA argue that the Commission
current
recommended
For this reason, there is nothing wrong with
use of the Commission’s
of institutional
In effect, the test year
pattern for the Postal Service is to
the relative contribution
the period in which the rates are in effect.
honest, efficient,
in this
proceeding
with no practical guidance
because
“witness
cannot recommend
Callow
provided
the
the
[with respect to the pricing criteria] for reaching
The easiest way to see this is to look at what happens if the test year is known, a priori, not to be
an average of the period that rates will be in effect: a deviation from break-even is then inevitable. But
such a deviation is prohibited.
Docket No. R2000-1
such a result.”
- 56.
USPS Brief at VII-31; DMA Brief at 6. OCA’s initial brief discusses
detail all the relevant
pricing criteria.
noted that the First-Class
mark-up
rate, based upon OCA’s proposed
OCA Brief at 155-60.
index inherent
in the current 33-cent
costs, incorporates
mark-up index is 1.353 in the test year after rates.
witness Callow’s
proposed
can be compared
to past Commission
OCA Brief at 156.
recommended
proposed
OCA-LR-I-3,
First-Class
it should
mark-up
That
In this manner,
to the Commission,
First-Class
be
single-piece
all the factors of the Act.
mark-up index provides guidance
well as the Postal Service’s
year at 34 cents.
Moreover,
in
mark-up
as it
indices,
as
index of 1.422 in the test
Part I, Table A.
The Postal Service argues that “[rletaining
the current 33-cent rate essentially
would require that all other mail classes and postal services bear much of the burden of
the increased
alternatives
shrinking
revenue
exist.
requirement.”
OCA has proposed
the Postal Service’s
Consequently,
USPS Brief at VII-l 1.
not all classes
reducing the First-Class
that the revenue
excessive
contingency
or postal services
Letter Mail institutional
cost burden is appropriate.
distributed
Standard
Standard Mail have roughly equivalent
in past opinions.
OCA Brief at 148-49.
Mail in order
other
be reduced
by
OCA Brief at 39-58.
have to bear the burden
cost contribution.
sharing of the institutional
to commercial
requirement
request.
would
To the contrary,
Nevertheless,
some
Thus, some burden should
that
First-Class
of
be
and commercial
mark-up indices, as intended by the Commission
- 57
Docket No. R2000-1
With respect to MMA’s desire to expand workshare
discounts,56 the OCA takes
no position on the proper level of discount, as OCA’s proposal
piece user.
OCA Brief at 160.
cent single-piece
Nevertheless,
it should be noted that retaining the 33-
rate would achieve the lower institutional
and provide very substantial
is focused on the single-
cost burden sought by MMA,
rate relief to First-Class business mailers.
CEM Should Be Recommended
8.
The Postal Service has again trotted out the same tired arguments
using against CEM since Docket No. R87-1.
These arguments
unsound
today.
in the past and they are unsound
documented
service that will be valuable
more optimally
and will give consumers
benefit directly from the Postal Service’s
1.
The classification
The Postal Service
Section 3623(c).
argues
CEM aligns cost and price
Individual
mailers will be able to
issue has lono been decided
that CEM does not meet the statutory
recommended
This is absolutely
untrue.
56
criteria of
In Docket No.
the following DMCS language to the Governors:
221.25
Courtesy Envelope Mail Rate Category.
envelope mail rate category applies to Letters and
subclass mail in envelopes that:
a. Are
approved
well-
investment in automation
USPS Brief at VII-34-37.
R97-1, the Commission
have been shown to be
CEM is a well-researched,
to consumers.
a choice.
it has been
preaddressed
and preprinted
by the Postal Service;
reply envelopes,
The courtesy
Sealed Parcels
of a design
The American Bankers Association and the National Association of Presort Mailers also seek
larger workshare discounts without taking a specific position on the appropriate single-piece First-Class
rate. ABA/NAPM Brief at 5-20.
- 58 -
Docket No. R2000-1
b. Bear a facing identification
mark as specified by the Postal Service;
c. Bear a proper barcode corresponding
specified by the Postal Service;
d. Bear an indication that the envelope
specified by the Postal Service; and
to the correct ZIP Code, as
is eligible for the discount,
e. Meet automation compatibility
criteria as specified
Service. PRC Op. R97-1, App. 2, pp. 9-10, 5 221.25.
The consistency
question.
of CEM with the Section
It was decided by the Commission
by the Commission
language
in Docket
was recommended
by the Postal
requirements
in Docket No. R97-1.
No. MC95-1.
by the Commission
MC95-1, App. 2 at 9. Clearly this argument
Surprisingly,
3623(c)
Indeed
as
is not an open
It was also decided
the exact same classification
in Docket No. MC951.
PRC Op.
is old news.
the Postal Service again makes the equity argument
to withhold
the
benefit of CEM from consumers.
For households, the reduced rate would be a complete windfall, because
in reality they would do nothing more than continue to mail the envelopes
that someone else has provided them. Equity suggests that households
should not receive the benefit of their low-cost CEM mail until they are
asked to pay a surcharge to cover the costs of their high-cost mail. USPS
Brief at VII-44.
The
envelope,
Commission
should
has made
clear that the mailer,
receive the benefit.
PRC Op. R90-1 at V-55-57.
have been saving the Postal Service processing
during
which
processing
the automation
even hand-written
program
not the provider
of the
CEM envelopes
and delivery costs for the entire period
has been in place.
mail has been going down.
In addition,
the cost of
Tr. 1415938-39.
Finalization
Center (“REC”) continues
to improve.
of mail before it is sent to a Remote
Encoding
The finalization
from 32.7 percent in FY 1998 to 58.5 percent
rates have increased
in
Docket No. R2000-1
- 59 -
FY 2000 (YTD AP 7). Tr. 7/3080.
processed
envelopes
2.
There is no basis to withhold
the benefit of cheaply
from consumers.
The Postal Service’s reliance on the Ellard survey is still misplaced
The Postal Service continues
dark by the Postal Service-has
to argue that the public-purposefully
kept in the
no interest in CEM.
As fundamental
a change as CEM would represent to the mailing
practices of the general public, it would seem imperative that the
Commission and the Governors have a very high level of assurance that
its adoption was something the public actually preferred.
USPS Brief at
VII-37.
The Postal Service
claims that witness
Docket No. R97-1 research
concerning
is completely different than acknowledging
Postal Service research.
inconclusive
public.57
and, therefore,
that it proved consumers
Postal
consumers
Service
the dreaded
agreed
“two-stamp”
with witness
problem.
Ellard’s
Agreement
that there may have been a valid point in the
Willette went on to explain that the research
no reason to withhold
Indeed, the Commission
was less than definitive
The
Witness
Willette
found the research
choice
did not want CEM.
has claimed
that the
of a lower rate from the
inconclusive
on CEM’s appeal, thus dismissing
was
witness
and stated that it
Ellard’s assertion
PRC Op. R97-1, Vol. 1 at 324, 75189.
Ellard
research
constituted
proof
did not want to be offered a discount to mail bills and other courtesy
that
reply
pieces.
The plain truth is that the Postal Service does not want to give consumers
choice.
No one (except the Postal Service) has suggested
57
Tr. 23110738. See also Tr. 23110782
that CEM be mandatory.
a
Docket No. R2000-1
Consumers
- 60 -
are capable of making choices;
if they do not want to use CEM postage
stamps, they will not use them.58
3.
Education funds have alreadv been set aside
The Postal Service contends
be coupled with an education
VII-39.
In Docket
requirement
that any “CEM enforcement
program designed
No. R97-1, the Commission
to educate consumers
program would need to
to educate the public.”
included
in the use of CEM.
$33 million in the revenue
Over the two years rates from
Docket No. R97-1 will be in effect, that amounts to $66 million.
already
in possession
of the funds it seeks.
USPS Brief at
The Postal Service is
Since the Postal Service has been given
these funds as a result of Docket No. R97-1, there is no need to allocate them again.
4.
The confusion
The Postal Service would
fragile creatures,
easily confused
issue is a red herring
have the Commission
by postage
as they say in Texas, is hogwash!
175.
Consumers
every day.
understand
that consumers
USPS Brief at VII-36-39.
The OCA has addressed
this issue.
choose among a large number of alternative
It is called competition.
products
and one of which does not), then choosing
are
That,
OCA Brief at
and services
If the Postal Service thinks consumers
two postage stamps for two different envelopes
Postage Allowed”
53
stamps.
believe
cannot
(one of which says “CEM
a long distance
carrier or
The Postal Service relies on witness Wlllette’s discussion of some aspects of CEM at a
conference to claim that witness Gillette agrees with the Postal Service on consumer interest. USPS Brief
at VII-37. To the contrary, the major point made by witness Gillette at the Vancouver Postal and Delivery
Economics Meeting in June 2000 was that since the Postal Service had resisted CEM since Docket No.
R67-1, the opportunity to use it effectively to stem diversion to electronic media may have been lost by
now. Electronic bill paying and ordering have become less costly and more convenient during the period
of more than a decade since CEM was first proposed.
Docket No. R2000-1
a cellular telephone
-61 -
service provider must be impossible.
as a monopolist-tailoring
rather than proactively
and restricting
offering consumers
attractive and relevant to them.
The Postal Service is acting
its service offerings
to suit its convenience,
choices that would make mail service more
The Postal Service (which claims that it must compete)
should begin to act like a competitor
and provide consumers
with choices
instead
of
assuming that they lack intelligence.
5.
Enforcement
Enforcement
40-42.
Enforcement
can be manacled
is again trotted out as a reason to reject CEM.
is an issue throughout
USPS Brief at VII-
the Postal Service and it has been an issue
since the Postal Service started to charge rates. So long as the Postal Service charges
rates, enforcement
will be an issue.
apply the wrong amount of postage.
of enforcement
be tolerated.”
exceeds
There are some mailers who, knowingly
Witness Willette testified that in general “if the cost
the underpayment
Tr. 23110766.
million to collect $10 million in underpaid
Operations
postage,
point of view.
would
Spending
$70
as the Postal Service suggests,
show that customers
OCA Brief at 179-60.
is
The
overpaid postage in 1999 by
Tr. 21/9112-13.
6.
The
then the underpayment
would allow it to happen.
data on the record of this proceeding
$204.6 million.
of postage,
This is a rational economic
irrational and no sane manager
or not,
Postal
Administrative
reasonina
Service
imagines
costs
are
the worst
are at risk and a costly administrative
exaggerated
possible
nightmare
and
world
based
if CEM
will ensue.
on
faulty
is adopted.
USPS Brief at
Docket No. R2000-1
VII-43.
176.
-62-
These objections
There
Consumers
are the result of seriously flawed
is no reason
are capable
to assume
Fortunately
e.g., by mail or over the internet.
stamps
unattractive
Whenever
habits at a much faster
Certainly most consumers
standing in line to purchase stamps.
probably
In addition,
would
this happens-that
consumers
not bother
to purchase
addition,
and E-stamp
the American
a discount
Bankers
Mailers have jointly proposed
Association
CEM stamps-the
proposal
for First-Class
a discount
Mail in this proceeding.
meters.
Tr. 29/13893.
for users of PC postage.59
and the National
the establishment
denominations.
it would change nothing.
for users of postage
have proposed
both
If the Postal Service is correct
The CEM proposal is the most researched
Pitney Bowes has proposed
Stamps.com
who find the notion of two
do not purchase
about the lack of interest in CEM, then implementing
discounts
pace than the Postal
there are other ways to obtain stamps,
is, when consumers
Several parties have proposed
will double.
would rather not spend more time
Postal Service will receive mail. with overpaid postage.
7.
OCA Brief at
that trips to post office windows
of adjusting
Service is willing to admit.
reasoning.
Association
of a new category
26112436-42.
research
None of these
and development.
and the Commission
59
discounts
has had the benefit
CEM has been thoroughly
is satisfied.
examined
It has been demonstrated
See Tr. 29/13650-51 and Tr. 23/10478-80.
of Presort
of First-Class
that would be collected by private firms and presorted to lower its processing
In
Mail
costs.
Tr.
of over a decade
of
by the Commission,
to the satisfaction
of the
Dock&No.
R2000-1
Commission
-63-
that CEM is a sound proposal that will not harm the financial well-being
the Postal Service.
The same cannot
be said of these new and relatively
of
untested
proposals.
The Commission
The
revenue
threatening.
has acknowledged
consequences
to the
that CEM is a viable consumer
Postal
CEM would offer consumers
Service
a choice.
are
not
large
service.
or financially
CEM also would promote good will
for the Postal Service among consumers.
The arguments
not persuaded
against CEM are uniformly without merit.
by the Postal Service’s
and should not be persuaded
put to rest, one by one.
Consumers
C.
by them now.
In Docket
envelope or package,
No. R2000-1,
to apply the two-pound
user.
lntervenor
to their needs
the One-Pound
Priority Mail Rate
Priority Mail rate for a one-
proposes
a separate
is
one-pound
rate of $3.85 for any envelope
or package,
rate to the special flat rate envelope.
This would
high rate increase
of 20.3 percent for the flat rate envelope
Douglas Carlson argues that such an increase is not justified, and that
the flat rate envelope
The OCA agrees.
CEM.
and for the special flat rate envelope,
the Postal Service
Priority Mail rate of $3.45 and a two-pound
result in an unacceptably
have been
The time has come to move ahead and implement
in Docket No. R97-1, the current
was
in Docket No. R97-1
The Postal Service arguments
The Commission Should Recommend
for the Flat Rate Envelope
pound or a two-pound
and proposes
arguments
need to see a Postal Service that is responsive
As established
$3.20.
disingenuous
The Commission
should be charged the new one-pound
rate. Carlson Brief at 1-9.
Docket No. RZOOO-1
64 -
If the Commission
recommends
popular flat rate envelope
rate envelopes
Service-a
mailed
and a uniform,
requested
approach,
the rate for the
will no longer be the lowest Priority Mail rate available.
are convenient
Postal
the Postal Service
to the public while offering two significant
high postage
payment
machinable
by the Postal Service,
relative
mailpiece
however,
As such, it disturbs the balance
to the average
undermines
between
benefits to the
weight
that is easy to handle.
the convenience
Flat
of pieces
The change
of a flat rate.
the benefits to the public and to the Postal
Service from use of the flat rate envelope.
Until recently,
free specialized
(currently
and conveniently
succeeds
convenience
with
and security
its two-pound
is a free envelope.
flat rate envelope is significantly
is such that postal customers
rate request,
As Mr. Carlson
diminished.
In support of his proposal
Mr. Carlson notes that 77 percent
the average
the only
box.
must
customer
the intrinsic value of a
Carlson Brief at 6.
to use the one-pound
rate for the flat rate envelope,
of all flat rate envelopes
weight
remaining
points out in his Initial Brief, the
Consequently,
and, in fact, 29 percent weigh three ounces or less.
calculates
into any USPS pick-up
more than one pound to a USPS window clerk. If the Postal
envelope alone is not worth $0.40 to consumers.
Service
as would fit into a
apply the lowest Priority Mail postage
drop the envelope
times have changed,
now present mail weighing
Service
could place as much mail material
flat rate Priority Mail envelope,
$3.20)
Unfortunately,
customers
weigh one pound or less
Carlson
Brief at 1. The Postal
of all flat rate envelopes
to be only
10.3
-65.
Docket No. RZOOO-1
ounces.60
On average, then, the Postal Service currently collects $3.20 for mail that
could be sent by First-Class
rate should adequately
Mail for $2.53.6’
compensate
The average weight is significant:
a flat
the Postal Service for the average piece mailed for
the rate, not just the relatively few pieces at the higher end of the weight range of flat
rate envelopes.
Flat rate envelopes
10.3 percent
have been widely accepted
of all Priority Mail volume and an astonishing
Mail volume for pieces weighing
pieces weighing
four ounces
Business
by the public.
four ounces or less..”
They comprise
56.2 percent
of all Priority
Just within flat rate envelopes,
four ounces or less comprise 35 percent of the total.63 Pieces weighing
or less, however,
can be sent by First-Class
Mail for $0.99
or less.
mailers are unlikely to pay the current Priority Mail $3.20 rate to mail only a
few ounces (when First-Class rates are much cheaper), which indicates that consumers
are using flat rate envelope
perception
average
for light mailpieces,
of premium service.
weight
(two-pound
would
unfairly decrease
60
USPS-LR-I-165,
By overpricing
either for convenience
the flat rate envelope
rate vs. 10.3 ounce average weight),
the value of the offering to consumers.
or for the
compared
to its
the Postal Service
The Postal Service
filename USPST34A, Worksheet “Input Data”, cell D155
61
Because of the 13 ounce limit on First-Class
Class is currently $2.97.
62
Mail, the maximum charge for a piece sent First-
USPS-LR-I-165, filename tJSPST34A. Worksheet “Input Data” indicates that in FY 1998 there
were 75,728,918 Priority Mail pieces weighing between O-4 ounces, of which 42.552,090 were flat rate
envelopes (56.2%).
For all weights, there were 119,934,151 flat rate envelopes out of a total of
1.167,814,341 Priority Mail pieces, or 10.3%.
63
Id. (42,552,090 + 119,934,151.)
Docket No. R2000-1
- 66 -
should not overcharge
consumers
for their acceptance
of the convenience
of the flat
rate envelope.
To the extent some consumers
premium service, the customer’s
cross-examination,
Commissioner
USPS
Goldway
claims that customer
choose
perception
witness
Priority Mail because
of the perceived
of Priority Mail may be changing.
Mayes
responded
to
a question
regarding the hierarchy of USPS mail services.
perception
During
posed
Witness Mayes
of Priority Mail service relative to competitive
has altered.
Tr. 1 l/4600.
Certainly,
performance
of Priority Mail lags that of First-Class
by
services
Postal Service data show that the actual service
MaiLB4 Under the circumstances,
there is no basis to charge a high rate based on premium service
As Mr. Carlson points out, customers
(under one pound) pieces will not generally
mailing lighter weight Priority Mail pieces
be inclined to pay the additional
for the privilege of using the Postal Service’s
Customers
can purchase
$0.40 just
“special” flat rate Priority Mail envelope.
an envelope for much less than $0.40, and will likely do so to
save on Priority Mail postage when their mailpiece
weighs less than one pound.
the
uniform,
customer
abandons
envelopes,
the Postal
packaging.
Machinable
should be encouraging
64
the
Service
Postal
machinable
loses one of the advantages
mail, however,
customers
Service’s
When
Priority
of providing
Mail
uniform
is exactly the type of mail the Postal Service
to provide them.
In FY 1999, 93.3 percent of First-Class Mail met a one-day standard versus 90.4 percent of
Priority Mail; for a two-day standard, it was 86.5 percent for First-Class Mail but only 79.3 percent for
Priority Mail. Tr. 2118857: see also Tr. 1 l/4602-06.
-67-
Docket No. R2000-1
The Postal Service’s pricing strategy lacks foresight.
Mail envelope
higher than the lowest
representative
of the predominant
rate available
Pricing the flat rate Priority
and at a weight
use of flat rate envelopes,
increment
not
will doom the Priority Mail
flat rate envelope to obsolescence.65
The OCA supports
Mr. Carlson’s
proposal to set the flat rate at the one-pound
rate. This would be a more modest increase of 7.8% (at the Postal Service’s proposed
rate levels).
Using the one-pound
rate for flat rate Priority
everyone.
The Postal Service would get more machinable
payments.
Customers,
use and needs.
meanwhile,
Mail envelopes
benefits
mail and higher postage
would get a product more in line with their actual
Given the lower service performance
Class letter mail, a more modest increase
of Priority Mail in relation to First-
in the flat rate Priority Mail envelope fee is
certainly warranted.
D.
The
The Commission Should Rely on the “As-Filed”
the Number of Additional Ounces in First-Class
Postal
methodology
methodology-the
Service
struggles
by refuting the analysis
methodology
mightily
attempting
of witness
Callow
originally
proposed
the number of additional ounces of single-piece
65
Methodology
to justify
in support
to Forecast
the
“revised”
of the “as-filed”
by the Postal Service to forecast
First-Class
Mail. USPS Brief at VII-15
Mr. Carlson argues that the Postal Service is likely to create a new Priority Mail envelope similar
to the current flat rate envelope, but which would be weighed and rated. Carlson Brief at 7. citing Tr.
7/2872. This is an unnecessary complication.
If the one-pound rate is applied to the flat rate envelope,
the Postal Service will need only one envelope. The only change required would be to state that the onepound rather than two-pound Priority Mail postage is to be applied. To the extent that customers use
older envelopes after the new rates are in effect, and mistakenly apply the two-pound postage, the Postal
Service will benefit. Certainly, however, the fact that existing envelopes specify the two-pound rate is not
a substantive reason to apply that rate once a one-pound rate is in existence.
Docket No. R2000-1
19. However,
- 68 -
Postal Service efforts cannot shake what is clear from the evidence
this proceeding:
the “as-filed”
Commission
the long-term trend in the number of additional
methodology
The
of the overpayment
Service
hopes
methodology
by referencing
methodology
previously.
ounces.
that trend.
The
efforts to offset
postage by adopting
the “revised”
trend
the
Commission
USPS Brief at VII-17.
will choose
the
“revised”
has used the “revised”
This is irrelevant.
The Commission
that best reflects the rising long-term trend in the number
by the “as-filed”
line projected
contrast, the “revised”
that
In this regard, witness Callow shows that the number of additional
per piece forecast
long-term
of First-Class
the fact that the Commission
should adopt the methodology
ounces
that best reflects
is rising, and
to forecast the number of additional ounces.
Postal
of additional
a forecast
should not be swayed by the Postal Service’s self-serving
its error correction
methodology
produces
ounces
in
through
methodology
methodology
tracks “almost
to the test year.
exactly”
the
OCA Brief at 162-63.
By
results in zero growth through the test year.
Id. at
167.
The Postal Service chastises
witness
Callow for not “taking into account
historical events might have affected the data” presented
number
of additional
ounces;
specifically,
the sharp
between
1994 and 1995, and 1997 and 1998.
on the long-term
increase
Witness
explains that his analysis shows that even if the years with large increases
(and only the other
additional
at 163.
years
are considered),
the long-term
trend in the
in additional
USPS Brief at VII-67.
trend
ounces
Callow
are ignored
in the number
ounces is still rising, although the rate of increase is not as great.
how
of
OCA Brief
Docket No. R2000-1
-69.
Moreover, the Postal Service places great emphasis
increases,
or alternatively,
at VII-68.
Use of the long-term
on the fact that these sharp
“one-time events,” will not occur in the test year. USPS Brief
trend takes into account the fact that new, “one-time”
events that drive-up the number of additional ounces are likely to happen again. As the
data show, there were two “one-time”
increase in the additional
events in the past decade.
The Commission’s
ounce weigh limit to 13 ounces from 11 ounces in Docket No.
R97-1 will likely have the same effect in the future.
Finally, the Postal Service takes issues with witness Callow concerning
that the lower rate of growth in additional
“sampling
error.”
USPS Brief at VII-69.
fact that sampling
1998.
problematic
36/16905.
ounces in recent years may be a function of
The Postal Service appears
error also affects the historical
Id. As explained
for estimates
Consequently,
by witness
based
his view
on small samples
of sampling
(or limited periods
estimates
error is most
of time).
Tr.
of the number of additional
ounces based upon the two most recent years of data may not be representative
long-term trend.
by the
data over the period 1990 through
Callow, the existence
the Postal Service’s
comforted
of the
Id.
In summary, what is apparent from this episode is that the “as-filed” methodology
was thoroughly
analyzed
and determined
to be proper, with the full backing of postal
management,
when it was originally filed as part of the Postal Service’s
proceeding.
Postal Service
hollow as the time elapsed
methodology
efforts to disavow
between
did not appreciably
Service now claims its decision
the “as-filed”
methodology
the original filing and introduction
increase
is based.
the amount
request in this
now ring
of the “revised”
of data on which the Postal
The only conclusion
to be drawn is that the
-7o-
Docket No. R2000-1
“revised”
methodology
revenue occasioned
First-Class
was rehabilitated
in order to offset the $192.3 million in total net
by the Postal Service’s necessary
correction
of the overpayment
of
postage.
E.
The Commission
Aspect Ratio
Should
Eliminate
the Nonstandard
In its initial brief, the Postal Service expresses
surcharge
acts as an incentive to use standard-size
objective
of the surcharge
incentive
to design
is to provide
mail pieces
Unfortunately,
the lack of information
on the response
any, nonstandard
Accordingly,
“the primary
who are able to do so sufficient
so that they contributed
USPS Brief at VII-21.
for Low
the belief that the nonstandard
letters.
to those
Surcharge
to operational
efficiency.”
the Postal Service’s belief cannot compensate
for
of mailers, or reliable data on what problems,
low aspect ratio letters present
in the current automated
if
processing
environment.
The Commission
should
eliminate
the nonstandard
ratio letter mail. It is apparent from the evidence
not create problems for postal mail processing
the contrary
lack support.
mail simply perpetuates
Retaining
an unfairness
surcharge
in this proceeding
equipment,
the nonstandard
for low aspect
that such mail does
and Postal Service claims to
surcharge
for low aspect
ratio
for which the Postal Service cannot document
a
problem.
In this proceeding,
nonstandard
surcharge
in Docket No R97-1.”
contributed
the Postal Service claims that “witness Miller has updated the
cost estimates
and addressed
USPS Brief at VII-82.
more verbiage,
revealing
several issues that were raised
In reality, the Postal Service has simply
Commission
criticisms from Docket No. R97-1 to
Docket No. R2000-1
-71-
be as valid today as when its opinion was issued.
PRC Op. R97-1, ~ 52265228.
As
was the case in Docket No. R97-1, the Postal Service is again hobbled in its efforts to
justify the surcharge
result.
Witness
the processing
by the lack of basic data on costs and processing-with
Miller again relies on proxies for estimating
of standard
and nonstandard
one-ounce
Witness Miller again assumes
all manual processing
“the absence
data which indicated
of any empirical
pieces [footnote omitted] can be processed
according
to the Postal
Service,
mail.
ld. at VII-20.
nonstandard
pieces
(or generation
primary “objective.”
Witness
Id. at VII-85
Callow
shows
processing,
that are lower than the surcharge.
ratio letters is minimal-approximately
should
precise
of costs
in pursuit
of
of
of the
ratio letter mail are
OCA Brief at 195-96.
of low aspect
Using a
ratio letters
$6.9 million.
the surcharge
USPS Brief at VII-85.
for low aspect
are
unit costs
that
on low aspect
Id. at 199. The Postal Service also
Consequently,
ratio mail as a matter of fairness,
mailers pay extra for mail that receives little (if any) additional
200.
measure
Id. at 197. Moreover, witness Callow calculates
loss is “not the issue.”
be eliminated
However,
id. at 191-93, witness Callow develops
the revenue loss to the Postal Service from eliminating
surcharge
ld. at VII-19-20.
that the unit costs for low aspect
that at least 50 percent
believes the revenue
given
see a/so Id. at VII-21.
more realistic
for automated
mailpieces,
is not necessary
of all manual processing.
presented
USPS Brief at VII-19.
has little effect on total costs
the
of revenues)
between
the degree to which nonstandard
affected by the assumption
assumption
pieces.
on automation.”
Moreover,
cost differences
of nonstandard
this assumption
nonstandard
the same
processing.
the
as
OCA Brief at
Docket No. R2000-1
72 -
In an effort to challenge
problem
for automated
processing
Callow fails “to appreciate
equipment
the notion that low aspect
the distinction
to read addresses
between
barcodes
primary operation.
the Postal Service fails to appreciate
form of “machinability”
in that low aspect
without
they
witness
processing
Haldi’s
Docket
network.
small-scale
No. R97-l-the
manual processing.
Moreover,
individual
the absence
purposes.
to increase
collection
discouraged
mail.
Id.
with seasonal
of any Postage
and reading of barcodes
is a
at least
processed
through
the entire
(not sentimental)
greeting
Due markings
cards
value of
presented
in
means there was no
Id. at 199.
Similarly,
due marking
means
that neither
Consequently,
individual
of low aspect ratio nonstandard
the Postal
Service
claims
mailers
have no
letters entered
that mailers
to business
as they lack any support.
mailers, whose
as
need to be
low aspect ratio letters by retaining the surcharge.
is referring
the
know that low aspect ratio letters are “nonstandard”
Such claims are unpersuasive
the Postal Service
at 21. What
To the extent such letters are delivered
have been
of a postage
the volume
from designing
Brief at VII-85.
processed
OCA Brief at 194.
mailer nor the recipient
for processing
incentive
absence
Id. There is no dispute
have been successfully
That is the evidentiary
experiment
of
The Postal
ratio letters are being processed,
network.
automated
USPS Brief at VII-84.
is that the application
Due” marking,
and the capability
Tr. 713225; see also USPS-T-24
partially, through the mail processing
a “Postage
machinability
test results” from the mid-70’s.
ratio letters exhibiting
through the “outgoing”
the Postal Service states that witness
and apply barcodes.”
Service relies on “machinability
that low aspect
equipment,
ratio letter mail is not a
USPS
Presumably,
mail matter includes
bills,
Docket No. R2000-1
statements
- 73.
of accounts,
mailers are economically
the entire
motivated
mail processing
documents.
possibility
and important
to design
network
In effect, business
correspondence,
mailpieces
to ensure
timely
mailers have no incentive
that such letters may not be processed
among
others.
Business
that move efficiently
delivery
of these
to do otherwise,
timely would discourage
though
important
since any
the volume
the Postal Service fears.
F.
Criticism of the OCA Rate Stabilitv Proposal Is Misouided
The
circumspect:
Postal
comments
on the
the Postal Service encourages
but acknowledges
VII-88.
Service’s
that the “reserve account
The Postal Service,
related to “the management
however,
OCA
the Commission
considers
of postal finances
also contends
Service moving to a two-year
statements
stability
development
and accounts
Id. at VII-87.
USPS Brief at
and rate implementation
Id.
To counter
upon the Postal
the widespread
of postal officials that the Postal Service expects to implement
again in 2003, the Postal Service cites the belatedly filed Library Reference
492.
This consists of an interview conducted
are
of such a plan to be
that the idea is premised
rate cycle.
proposal
to “shelve” the proposal
idea is not uninteresting.”
policy,” and best left to the discretion of the Postal Service.
The Postal Service
rate
by the Association
new rates
USPS-LR-I-
for Postal Commerce
in late August with Deputy PMG John Nolan, and posted on the PostCom website on or
about September
1, 2000.
In the interview,
implement new rates at any specific time.
Mr. Nolan says that there is no plan to
Docket No. R2000-1
- 74 -
Library Reference
the evidentiary
statement
Service
to be an effort
statements
is unsolicited
and is not sponsored.
It is not in
record of this case, and its use by the Postal Service on brief as a
of Postal
appears
USPS-LR-I-492
policy must be disregarded.
Moreover,
by Mr. Nolan to muddy the waters
about
to large mailers that rates will change again in 2003.
the interview
his prior public
Many of these large
mailers are encouraging
the Postal Service to put rate changes on a more regular basis
so as to have smaller,
more frequent
and predictable
rate changese6
Certainly,
the
present case will change rates after only two years.
However, the effort by the Postal Service to keep everyone
timing of future rate changes
permitting
frequent
It is intended
on the Postal Service committing
to provide greater convenience
the Postal Service to meet the requests
and predictable
be raised-low
inflation,
about the
is largely irrelevant to the OCA stability proposaL6’
OCA proposal does not hinge exclusively
rate cycle.
guessing
rate changes.
to a short
to the mailing public while
of large mailers for smaller, more
But if conditions
high postal productivity
The
are such that rates need not
gains, volume growth, etc.-then
reserve account would not fall short if the rate stability plan were in effect.
the
The plan
works to the benefit of the public both in short rate cycles and when conditions
permit
the Postal Service to delay rate increases.
66
67
See, e.g., Tr. 22/10132-33
(Callow); Tr. 2279829 (Rosenberg).
This appears to be an ongoing problem with the Postal Service. Senior officials make various
statements to mailers that appear to differ from statements made for the benefit of the Commission. The
OCA encourages the Postal Service to develop a more consistent approach to informing mailers, the
Commission, and the public about the Postal Service’s future rate plans.
- 75 -
Docket No. R2000-1
The OCA notes that the Postal Service asks the Commission
and let it be considered
by senior postal management.
the OCA hopes that senior management
OCA suggests that the Commission
The Direct Marketing
“unworkable
prompted
mailers.
USPS Brief at VII-88.
Association
would
primarily
affect First-Class
Mail (such as the Major Mailers Association,
the National Association
proposal
as
DMA Brief at 9. Evidently, the DMA’s concern
Id. at 11. It is notable that none of the parties representing
Association,
The
such consideration.
attacks the OCA’s rate stability
by its belief that the proposal
using First-Class
Naturally,
will give the plan serious consideration.
actively encourage
and probably unlawful.”
to “shelve” the plan
business
the American
is
business
mailers
Bankers
of Presort Mailers, and Keyspan Energy, Inc.) has
taken this position.68
The reason, of course, is that the OCA proposal is designed
single-piece
First-Class
and by businesses
Mail.
This is the mail service used universally
for small volumes
would set new rates for all workshare
based on a calculated
to be limited to only
First-Class rate.
of First-Class
categories
Mail matter.
of First-Class
Only for single-piece
by households
The OCA proposal
Mail in every rate case
First-Class
integer rate then be set to permit the rate to remain in effect through
Mail would the
two rate periods.
OCA Brief at 182-87.
The DMA is flat wrong when it states that OCA witness
that the “vast majority” of mailers affected
68
Callow acknowledged
by the rate stability plan would be business
The DMA, as it stated in its intervention in this case, “expects to be concerned primarily
Standard (A) mail.” “Notice of Intervention of Direct Marketing Association, Inc.,” January 13, 2000.
with
Docket No. R2000-1
mailers.
- 76 -
DMA Brief at 11. The interrogatory
at Tr. 22/10181,
concerned
response
cited to by the DMA, appearing
the OCA’s primary proposal
to retain the existing 33-cent
rate. Witness Callow was asked to estimate the savings to all First-Class
rate remained at 33 cents rather than being increased
those savings accrued to household
asked
about the portion
versus business
of single-piece
First-Class
mailers if the
to 34 cents, and what portion of
mailers.
Witness Callow was not
Mail sent by consumers
versus
business mailers, as the DMA’s brief suggests.
The DMA also complains
Brief at 11. The OCA proposal,
are examined
workshare
incentives
rates are set.
would be distorted.
however, is careful to assure that workshare
in every rate case, as is the calculated
rate for single-piece
First-Class
The only effect is that the relationship
mail and the workshare
the second rate case, because
calculated
that workshare
rate for bulk First-Class
volume, and proposed
to assure that the swings are always moderate.
discounts
rate from which the
between
the integer
Mail will change in
the integer rate will remain the same.
possible swings in workshare
DMA
Witness Callow
a limit on the difference
OCA Brief at 185.
Nor does the DMA succeed in its attempt to cast a shadow on the OCA proposal
by suggesting
that it “entails significant
proposal works creatively
prerogatives
legal problems.”
within the framework
DMA Brief at 12. The OCA
of the Postal Reorganization
Act.
The
of the Postal Service with respect to the timing and size of rate requests
and the timing of rate implementation
any issue in every case are preserved.
are preserved.
The rights of all parties to litigate
The use of the reserve account will assure that
Docket No. R2000-1
single-piece
- 77.
First-Class Mail breaks even over two rate cases, and that other classes of
mail are unaffected
VI.
by providing rate stability to single-piece
THE OCA COST MODEL IS ACCURATE
First-Class
Mail.”
AND RELIABLE
The Postal Service resists the use of any cost model other than its own.
includes
resistance
witness
Thompson.
presentation,
Ewen.
to certain improvements
Witness
Thompson
certain problems
replicated
In the course
encountered
Service’s
Smith and
Thompson
or rebut witness Thompson’s
the Postal Service now attempts
cost model
of witnesses
of doing so, witness
in cost modeling
On brief,
that the OCA cost
USPS Initial Brief at V-159-60.
the OCA cost model is an improvement
corrected
Tr. 23/10410-412.
testimony.
to create the impression
model has “flaws” or that it is “difficult to use.”
contrary,
the Postal
with the Postal Service’s cost model.
The Postal Service did not challenge
however,
in the cost model process offered by OCA
then illustrated the effect of the recommendations
Tr. 23/10407-415.
This
To the
that is accurate and
that should be relied upon by the Commission.
The “flaws” claimed
by the Postal Service are that witness
manually edit certain data to remove Standard
model uses a rounding
simplified
Service
69
certain
program
component
cost model must be discussed
Id.
had to
(A) single piece costs, that the OCA cost
at the last stage before printing,
numbering.
Thompson
The rounding
to understand
and that the OCA
problem
in the Postal
the need to manually
edit the
The Joint Brief of the Association for Postal Commerce and the Mail Advertising Services
Association International discusses the stability proposal in a footnote. APClMASA Brief at 8, n. 1. That
discussion notes that if the Commission accepts the OCA’s primary proposal to retain the existing 33.cent
rate, it is not necessary to recommend the rate stability plan in this docket. The OCA agrees. OCA Brief
at 187.
Docket No. R2000-1
- 78 -
Standard (A) single piece costs. As will be seen, the Postal Service is clearly stretching
to claim fault when there is none.
A.
The OCA Approach
Service’s Cost Model
The Postal
numbers.
Thus,
Service’s
cost model
whenever
Therefore,
costs
for each
potential to calculate
is limited
the program
to storing
performs
to the
integer
a mathematical
has a problem, because
While this may seem innocuous,
Consequently,
it cannot store
result to an
it impacts the USPS rollthe
rounding takes place innumerable
times as the USPS cost model costing and roll-forward
program that must repeatedly
operation
With each cost calculation,
mail class and subclass.
a real occurs.
Postal
values-whole
the USPS program must round each calculated
integer prior to data storage.
forward
Is an Improvement
the USPS program
involving division or multiplication,
decimal values.
to Rounding
process proceeds.
A computer
round costs to integer values prior to electronic
storage is
less accurate than one that maintains costs as reals.
To illustrate the effect of using integers (the USPS cost model) versus reals (the
OCA cost model), consider the following example.
one widget are $1 .OO. Further, assume that producing
material costs of manufacturing
a widget
involves
a manufacturing
facility to house operations,
Assume that in Year One, the actual
etc.
process
that requires
All production
production
is to continue
and the product
production
costs increase widget manufacturing
capital equipment,
labor, a
costs must be borne by the widget
is to break-even.
For illustrative
costs 60 percent.
if
purposes,
In our example,
the
final cost to produce one widget, in Year One, is $1.60 ($1 .OO l ISO%), if one uses the
OCA cost model.
However,
if one uses the Postal Service’s cost model program,
the
- 79 -
Docket No. R2000-1
$1.60 is subsequently
rounded and stored as $2. Further assume that in Year Two, the
cost to produce a widget is “rolled forward” so that the “forecasted”
be determined.
Continuing
with our example,
costs are expected to increase
an additional
assume
cost model, the cost to produce
There are no differences
50 percent.
In the OCA cost model, the
illustrates
However,
using the Postal
one widget in Year Two is $3 ($2 * 150%).
in the costing assumptions.
upon how the data is stored electronically.
cost can
that in Year Two, production
widget in Year Two costs $2.40 ($1.60 *150%) to produce.
Service’s
production
The entire difference
The difference
two points: (1) the Postal Service’s
between
is based
the two cost models
model is less accurate
than the OCA’s,
and (2) the Postal Service’s model can artificially inflate costs.
The Postal Service
embedded
fails to explain the multiple
in its cost model program.
only once-just
prior to printing.
The OCA model certainly
the user desires,
However,
rounding
that is
The OCA’s cost model, however, rounds the data
Even here, the use of rounding
is capable
methodology
is only for convenience.
of printing reals to whatever
degree of accuracy
as a practical matter, the OCA decided to round a real to
“zero” decimal places based upon the number of columns of data to be printed on one
paw.
The OCA’s rounding
convention
third grade math. As explained
0.5 is rounded
23/10436.
computational
is not mysterious
by witness Thompson,
or hard to follow-it
any positive data value less than
down to 0.0; while 0.5 to less than 1.0 is rounded
Accuracy
is improved
stage throughout
is basic
up to 1.0.
Tr.
by doing this only at the end, rather than at each
the cost model, as the Postal Service does.
- 80 -
Docket No. R2000-1
In the postal rate making process,
prices.
In many instances,
Certainly,
the prices
the Postal Service’s
charged
for postage
the Postal Service’s
to First-Class
to the nearest integer, because
First-Class
Subsequently,
The
letters and
the results had to
the only instructions
of costs between
the USPS program
Standard (A) single piece letters
letters and Priority Mail.
The OCA’s cost model replicated
the Postal Service’s commands
not force rounding at each step. Consequently,
i.e. residuals-remaining
problem,
the
model to split (e.g., on a 95-5 percent
(A) single piece letters between
could store were integer movements
Perhaps,
model is limited to storing integers.
Priority Mail prior to rolling forward costs to FY 2000.
be rounded
are reals, not integers.
(A) Sinale Piece Costs Is Clear
Postal Service issued instructions to its computer
basis) the costs for Standard
about “real” stamp
to the OCA’s cost model.
The Manual Editina for Standard
As just discussed,
are concerned
cost model could stand improvement.
Postal Service should consider upgrading
B.
intervenors
therefore,
(not the “numerous
a few components
in the data file for Standard
and segments”
had small fractions-
(A) single piece letter costs.
stems from the Postal Service cost model.
cost components
exactly but did
The
In a few components
claimed by the Postal Service), the
residual values rolled forward to the test year with printouts showing a negative zero or
a one in the Standard
potential
instructions
Postal
(A) single piece letter cost category.
Service
confusion,
OCA witness
to the OCA cost model to automate
This improvement
Thompson
In an attempt to minimize
went
the “zeroing-out”
back and issued
of the few residuals.
was made to resolve the problem in the Postal Service model and to
Docket No. R2000-1
-81 -
assist the Postal Service
in following
the change.
It is hard to credit the Postal
Service’s claim that it found this “very difficult to follow.”
C.
The OCA “Scratch Pad” Component
USPS Brief at V-159.
Numbers Are Not Confusing
The USPS claims that the OCA cost model is confusing
same ten component
forward
years.
numbers
numbers
for the same distribution
The Postal Service’s
to accomplish
replicated the Postal Service’s
Thompson
chose simplicity
complaint
is pure makeweight.
keys spanning
cost model required
the same task.
activity.
The OCA’s
twenty
roll-
distinct component
Once again, the Postal Service’s
of the Postal Service cost model was faithful, and the elimination
VII.
multiple
cost model could easily have
It is apparent that the Postal Service is grasping for straws.
Commission
it used the
However, in the interest of clarity, OCA witness
over exact duplication,
final stage is an improvement
because
that should be adopted
The OCA replication
of rounding until the
by the Postal Service.
The
should rely on the OCA cost model.
NEITHER THE POSTAL SERVICE’S NOR CSA’S FEE FOR BULK PARCEL
RETURN SERVICE SHOULD BE RECOMMENDED
BY THE COMMISSION
The Postal Service proposes
$1.65 per piece.
It also recognizes
both mailers and recipients.
proposes
a cost coverage
why the Postal Service’s
a fee for Bulk Parcel Return Service (“BPRS”)
that BPRS provides a fairly high value of service to
USPS-T-39
(revised
of only 146 percent.
proposed
reflect its high value of service.
of
l/28/00)
at 17.
Nevertheless,
In its initial brief, the OCA discussed
BPRS fee and cost coverage
OCA Brief at 214-15.
Moreover,
does not adequately
the OCA discussed
it
Docket No. R2000-1
- 82 -
why the proposed
fee should be rejected in favor of a Commission-recommended
that results in a cost coverage at least as high as the proposed
this proceeding.
ld. at 215-18.~ Consequently,
systemwide
the following comments
fee
average in
will address the
mischaracterizations,
factual errors, and twisted logic in the initial brief of the Continuity
Shippers
(“CSA”) with respect to the appropriate
Association
fee and cost coverage for
BPRS.
CSA
claims
“mathematical
Mohammad
that the cost
calculations,”
coverage.”
agreement
[among
systemwide
average.”
is simply
the
of the Postal Service’s
in Docket No. MC97-2-the
proceeding
outcome
in which BPRS
of its view
as to the “original intent of BPRS [ ] for a
Id. This is also the corollary to CSA’s view that “there was no
the parties]
that the cost coverage
should
be at or around
the
Id. at 9.
CSA errs on both accounts
and creatively
interprets the history of this service.
As Postal Service witness Adra made clear in Docket No. MC97-4, the proceeding
established
of
pricing witness
CSA Brief at 10. CSA makes this claim in support
that the Postal Service made “no statement”
systemwide
for BPRS
citing the testimony
Adra (USPS-T-2)
was initially proposed.
coverage
that
BPRS:
The proposed fees and fee levels were selected with the existing cost
coverages and rate structures of related services in mind. At these fees,
the cost coverage of BPRS is 156 percent. This cost coverage level is
close to the systemwide cost coverage and easily recovers the associated
cost (criterion 3). Docket No. MC97-4, USPS-T-2 at 16.
CSA’s
Commission
agreement
interpretation
approved
of history
the service
of the “parties”
belies the facts.
at the request
to a revised
Stipulation
BPRS
exists
of the Postal
and Agreement.
because
Service
the
with the
See PRC Op.
Docket No. R2000-1
-83-
MC97-4, Appendix
A. Clearly, the “parties” cannot claim that they did not know that the
BPRS cost coverage
was approximately
had ample opportunity
to object.
Stipulation and Agreement,
the systemwide
Moreover,
the “parties”
CSA also conjures
up numerous
other
signatures
on the revised
agreement,
by the Postal Service at that time.
arguments
fee and the very low cost coverage
arguments-that
and they certainly
including that of CSA’s, by definition constitutes
if only tacitly, with the BPRS cost coverage proposed
alternative
average,
in an effort to justify
it has proposed
BPRS should have the same cost coverage
for BPRS.
as Standard
the
These
(A) Regular,
that the pricing criteria support a cost coverage for BPRS that is the same as Standard
(A) Regular,
extensively
and that a lower cost coverage
makes
considerably
the following
claim:
Not so. The existence
by comparison.
measured
of considerably
access
inconvenience
70
CSA
to
for BPRS.”
And,
to make
Postal
higher priced alternatives
contrary
collection
are
does explain
to CSA’s view, demand,
as
OCA Brief at 216.
argument
system
is
that
individual
somehow
for the return of BPRS parcels and lowers the value of service.
See OCA Brief at 222-24; 215-I 8; and, 218-21, respectively
priced
cheap (at the current fee of
the convoluted
Service’s
alternatives
CSA Brief at 14 (emphasis
is a measure of value of service.
appears
the
available
which is relatively
Id. at 10723.
by price elasticity,
Moreover,
customer
“[tlhat
higher may explain the demand
the demand for BPRS, Tr. 23/10712,
$1.75)
dealt with
in OCA’s initial brief. ” However, CSA’s creativity must be checked.
CSA
added).
for BPRS is justified-were
To wit:
an
84 -
Docket No. R2000-1
In addition, rather than just going to the mailbox, the original recipient may
have to use a collection box of some other location rather than his own
mailbox to enter the BPRS parcel. CSA Brief at 15.
First, CSA is incorrect.
Under BPRS, the original customer
a BPRS parcels using his own mailbox.
other location”
is an enhancement
parcels compared
Before BPRS commenced,
customers
supported
has a high value of service.
by recommending
systemwide
average.
a BPRS
box or some
to BPRS customers
which existed prior to the establishment
pay for the return of their merchandise.
The only conclusion
Second, access to “a collection
of service available
to the conditions
is permitted to return
of BPRS.
were obligated to visit a post office to weigh and
OCA Brief at 218.
by the evidence
The Commission
fee that
returning
in this proceeding
is that BPRS
should reflect that high value of service
produces
a cost coverage
at or above
the
Docket No. R2000-1
-85-
CONCLUSION
The
consistent
OCA
requests
that the Commission
with the discussion
issue
of the issues presented
its recommended
in this Reply Brief and in the
0CA.s Initial Brief.
Respectfully
submitted,
OFFICE OF THE CONSUMER
Director
Emmett Rand Costich
Shelley S. Dreifuss
Kenneth E. Richardson
Attorneys
1333 H Street, N.W.
Washington, D.C. 20268-0001
202) 789-6830; Fax (202) 789-6819
September
22,200O
decision
ADVOCATE
CERTIFICATE
OF SERVICE
I hereby certify that I have this date served the foregoing
participants
of record in this proceeding
Practice.
Washington, D.C. 20268-0001
September 22,200O
in accordance
with Section
document
upon all
12 of the Rules of