&c jg
BEFORE THE
POSTAL RATE COMMISSION
POSTAL RATE AND FEE CHANGES,
1997
:
3 L’Ij; ;i ‘CJ
,,,,~,,~
wt..,,,: :ii ::, r :
DOCKET NO. R97-1
DIRECT TESTIMONY OF
STEPHEN E. SELLICK
ON BEHALF OF
UNITED PARCEL SERVICE
,:~/
TABLE OF CONTENTS
Paae
INTRODUCTION
PURPOSE
OF TESTIMONY
MODS-BASED
A.
B.
C.
.
ALLOCATION
.
.
. .
.. .
.
.
OF MAIL PROCESSING
Mr. Degen’s MODS-Based Approach
Improvement Over Past Practice .
.
... . 1
.
2
COSTS
3
Is An
.
The Criticisms of MODS Piece Handling Data
Do Not Apply to Mr. Degen’s Use of MODS
Workhours Data
.
.
4
.
. 11
Mr. Degen’s Distribution Method Should Be
Used With Minor Modifications
. .
.
BASE YEAR AND TEST YEAR COST CALCULATIONS
PRIORITY
MAIL PROCESSING
COST DIFFERENCES
RECALCULATION
OF DBMC NON-TRANSPORTATION
AVOIDED IN OUTGOING OPERATIONS
.
.
SUMMARY
OF CONCLUSIONS
. . .
.
12
15
BY SHAPE!
.
18
COSTS
.
19
.
21
LIST OF TABLES
Table 1 --
Table 2 --
Comparison of Key Elements: LIOCATT versus
Postal Service Proposal
. . .
.
7
BY 1996 Volume Variable Cost Segment 3.1
CostsbySubclass.....................................
14
Table 3 --
BY 1996 Volume Variable Costs by Subclass
16
Table 4 --
Test Year 1998 Volume Variable Costs by Subclass
Table 5 --
Mail Processing Costs by Shape for Priority Mail
(TY 1998) Mail Processing Labor Costs 100%
Attributable
.
.
Table 6 --
Parcel Post Costs Excluded from DBMC Avoided
Cost Calculation
. .
.
.
17
.
.
19
21
BEFORE THE
POSTAL RATE COMMISSION
POSTAL RATE AND FEE CHANGES,
1997
:
DOCKET NO. R97-1
DIRECT TESTIMONY OF
STEPHEN E. SELLICK
ON BEHALF OF
UNITED PARCEL SERVICE
INTRODUCTION
1
My name is Stephen E. Sellick.
2
I am an Associate
3
8 Bartlett, Inc. (“PHB”), an economic and management
4
in Washington,
5
California;
a New Zealand subsidiary with offices in Auckland
6
Australian
subsidiary
7
affiliate, Putnam, Hayes & Bartlett Ltd., with offices in London.
0
PHB’s Washington,
D.C.; Cambridge,
Massachusetts;
with offices in Melbourne
consulting firm with offices
Los Angeles and Palo Alto,
and ‘Wellington;
I arn located in
D.C., office, 1776 Eye Street, N.W., Washington,
10
wide range of assignments
11
analysis of regulated
12
environmental
13
costs.
in regulatory economics,
industries.
an
and Sydney; and a United Kingdom
I have more than eight years of consulting
9
at Putnam, Hayes
experience,
cost accounting,
D.C. 20006.
including a
and financial
In addition, I have extensive exps!rience in
litigation, including projects dealing with the allocation
of common
I have worked on PHB’s analytic investigations
1
of United States
2
Postal Service (“Postal Service”) costing issues since 1990. In Docket No. R90-1
3
and again in Docket No. R94-1, I assisted Dr. George R. Hall in the preparation
4
testimony regarding
5
Mail. In Docket No. R94-1, I assisted Dr. Colin C. Blaydon in the preparation
6
analyses and testimony concerning
7
Office Cost System (“IOCS”).
a
the preparation
9
parcels handled by the Postal Service in First Class and Standard (A) Mail and in
10
supplemental
the attributable
I assisted Ralph L. Luciani in
of analyses and testimony regarding the costs associated
12
Postal Service facilities,
13
occasions,
14
two Associate
15
an Airport Mail Center.
of a number of
D.C., BMC on iwo different
a Sectional Center Facility, two Processing
and Distribution
Centers,
Units, a HASP (Hub and Spoke Project) facility, and
I hold a B.S. in Economics from the University of Penlnsylvania’s
17
Wharton School of Business and an M.A. in Public Policy Studies ,from the
10
University of Chicago.
PURPOSE
19
20
with
rate design for Standard Mail (A) parcels.
including the Washington,
Offices/Delivery
of
the treatment of mixed mail costs in the In-
Since 1995 I have visited and observed the operations
11
16
costs of Parcel Post, Priority Mail, and Express
In Docket No. MC951,
testimony regarding
of
OF TFSTIMONY
I have been asked to review those aspects of the costing proposals
21
the Postal Service which are discussed
22
testimony and workpapers
below.
In so doing, I reviewed the
of Postal Service witnesses
-2-
Degen (US;PS-T-12)
of
1
Alexandrovich
(USPS-T-5),
Moden (USPS-T-4)
2
(USPS-T- 28) Bradley (USPS-T-14)
My testimony provides the following:
4
1.
5
System-based
6
revisions.
A recalculation
8
percent mail processing
9
Kevin Neels (UPS-T-l).
3.
10
11
between
12
(UPS-T-3)
Crum
of Mr. Degen’s Management
Operating
Data
(“MODS”) costing changes to Cost Segment 3, and suggested
2.
7
(USPST-15)
and Daniel (USPS-T-29).
3
An examination
Patelunas
of base year and test year costs under 100
labor cost variability as recommended
A calculation
of the mail processing
by UPS witness
unit cost differences
Priority Mail flats and Priority Mail parcels. UPS witness Ralph L. Luciani
uses this cost differential
4.
13
The identification
to develop a Priority Mail parcel surcharge.
of the costs of certain Parcel Post operations
14
which are then used by Mr. Luciani to calculate a more appropriate
15
discount.
DBMC
MODS-BASED ALLOCATION
OF MAIL PROCFSSING COSTS
16
17
The Postal Service presents two witnesses
10
who address mail
19
processing
labor costs in Cost Segment 3: Mr. Degen (USPS-T-l
20
Bradley (USPS-T-14).
21
this subject; Mr. Degen’s testimony deals with how to distribute
These two witnesses
-3-
2) and Dr.
address entirely separate aspects of
mail processing
1
labor costs among the subclasses
2
degree to which mail processing
In my testimony,
3
of mail, while Dr. Bradley testifies about the
labor costs are variable and therefore
I address only the subject covered bly Mr. Degen, the
4
distribution
5
approach
6
criticisms of MODS piece handlinq data applicable
7
affect Mr. Degen’s methodology,
8
those costs found to be attributable.
9
programming
10
of costs to subclasses
represents
of mail.’
an improvement
modifications,
Specifically,
over past practice.
I discuss why Mr. Degen’s
I also explain why
to Dr. Bradley’s, analysis do not
data to distribute
which uses MODS workhours
Finally, I recommend
that, wii:h minor
Mr. Degen’s approach to distributing
mail processing
labor costs to each mail subclass be adopted by the Commission.
11
12
A.
13
Mr. Degen’s approach to distributing
Mr. Degen’s MODS-Based Approach
Improvement Over Past Practice.
is an improvement
Is An
attributable
mail processing
over past Postal Service and
14
labor costs to subclasses
15
Commission
16
mail and “overhead”
17
mail processing;
18
sacks, bundles, trays, and pallets) and containers
19
distribution
practice in two important respects: (1) it links the distribution
(not handling mail) costs with the operational
and (2) it incorporates
of mixed
characteristics
information on the contents of items (m,
more completely
into the
of mixed mail costs. I discuss each of these improvements
in turn.
In previous cases, the Postal Service has relied on IOCS and
20
21
attributable.
LIOCATT (a series of Postal Service computer programs) to distribute attributable
1.
Dr. Neels addresses
Dr. Bradley’s testimony.
of
1
mail processing
costs for clerks and mailhandlers
2
sampling system which estimates the proportion
3
spend on different activities associated
4
and providing each type of special service.
5
distribute attributable
lOlCS is a work
of time clerks and mailhandlers
with the processing
of each type of mail
The time proportions
in-office costs to subclasses
IOCS observations
6
by subclass.
alre then used to
of mail and special services.
can be “direct” or “mixed.”
Direct observations
are
7
recorded when the IOCS data collector observes an employee handling (a) a single
8
piece of mail; (b) an item or container that contains only one subclass of mail
9
(“identical”
items and containers);
or (c ) a sufficiently
random non-identical
item by
IO
recording
the subclass of the top piece using the “top piece rule.” Mixed tallies are
11
those observations
12
mixture of different classes or shapes of mail. Mixed mail tallies include uncounted
13
items and containers
14
IOCS also records “overhead”
15
on break, clocking in or clocking out, or moving empty equipment.
in which the employee is engaged in an activity involving a
as well as “working but not handling mail” observations.
tallies, which are observations
when the employee
is
The LIOCATT procedure formerly used by the Postal Service
16
17
distributed
the costs associated
18
proportion
to the class and shape distribution
19
accomplished
20
Function.’
Overhead costs were then distributed
21
proportion
to the final distribution
2.
with mixed mail to the subclasses
of direct mail tallies.
of mail in
LIOCATT
this process through cost pools (“strata”) grouped by CAG and Basic
to subclasses
of mail in
of direct and mixed mail costs.
CAG stands for Cost Ascertainment
on revenue.
Group, a classification
-5
of facilities based
Mr. Degen’s revised methodology
methodology
in four ways:
and mailhandlers’
processing
differs from the previous
(1) hours data from MODS are used to partition clerk
compensation
costs into “cost pools” based on certain mail
activities and machinery types; (2) the distribution
of mixed mail costs is
stratified by these cost pools rather than by CAG and Basic Punctieon; (3) the mixed
mail distribution
estimates,
incorporates
developed
IOCS data on container contents; an’d (4) variability
by Dr. Bradley, are then applied to each of the cost pools.
Table 1 compares the Postal Service’s current appro’ach in this case
with the previous methodology
for the key elements involved:
-6-
Table 1
1
10
11
12
13
14
15
16
17
Comparison
of Key Elements: ILIO CATT versus Postal SC
srvice Proposal
Issue
I?94-1 (LIOCATT)
H\‘)7-1 (MODS/lOCS)
Division of Cost Segment 3 Labor
Costs Among Mail Processing,
IOCS Based
MODS Based
Window Service, and
Administrative Costs
Cost Pools for Distributing Mixed
TIODS operation, BMC
CAG and Basic
Mail Tallies
operation type, or Basic
Function Only
Function
Uncounted Items Distribution Key
Mail subclasses
All Direct Mail and
observed for the same
Counted Mixed Mail
type of item within the
within Cost Pool
same Cost Pool
Mail subclasses
Uncounted Container: Items
All Direct Mail and
Distribution Key
observed for the same
Counted Mixed Mail
type of item within the
within Cost Pool
same Cost Pool
Mail subclasses
Uncounted Container: Loose Mail
All Direct Mail and
observed for the same
Distribution Key
Counted Mixed Mail
mail shape within the
within Cost Pool
same Cost Pool
Mail subclasses
Not Recorded and Empty Container
All Direct Mail and
observed for the same
Distribution Key
Counted Mixed Mail
container type within the
within Cost Pool
same Cost Pool
Mail subclass in the cost
Overhead Distribution Key
Final Cost
pool where overhead is
Distribution
incurred
-
The Postal Service’s new approach
18
is a significant
improvement
19
previous practice.
20
techniques
21
notes, the previous method (LIOCATT) for distributing
22
costs into “pools” based on (1) CAG, which relates to the amount of revenue
23
generated
24
the type of processing
over
The primary point of difference between the new and the old
is to refine the mixed mail distribution
methodology.
A:s the table above
mixed mail costs grouped
by a postal facility, and (2) the Basic Function involved, which relates to
operation
-- Incoming, Outgoing, Transit, and Other. The
-7-
1
new method also uses cost “pools,” but these cost pools represent
2
level of distribution
3
characteristics
4
mail, instead of CAG and Basic Function, which do not drive mail /processing labor
5
COStS.3
than LIOCAlT.
a much finer
The new pools relate to operational
and machine type, which affect the costs incurred in processing
The new method treats mixed mail observed in OCR operations,
6
for
7
example, as likely to be similar to direct mail at OCR operations.
The old method
8
was much less refined; it assumed that mixed mail observed in OCR operations
9
was similar to &l direct mail at postal facilities of a similar size and Basic Function.
10
The old method ignored the fact that mixed mail at OCR operations
11
resemble direct mail at OCR operations
12
operations.
13
which recognize
than direct mail at OCR and non-OCR
In fact, the old method completely
ignored available clperational
the different character of various mail processing
In adopting this refinement,
14
operations.
the Postal Service has addressed
and the Commission
data
standing concerns
16
costs associated
17
that the costs of “not handling mail” are allocated to the subclasses
18
found on the same machine type or in the same processing
19
employees
20
Priority Mail processing
21
handling mail, the costs of the time they spend while not actually handling mail will
with “not handling mail” IOCS tallies.
are handling
have expressed
long-
15
3.
that intervenors
is more likely to
The new method assures
operation
mail. If, for example, postal employees
operation
are more frequently
For non-MODS offices, the new approach
define the cost pools.
-a-
about the
of mail that are
when
in the manual
observed working but not
continues to use Basic Function to
1
be allocated only to the subclasses
2
they are handling mail.
of mail with which those employees
work when
Postal Service reports as far back as 1992 have recommended
3
4
essentially
this approach.
5
Foster Associates
For example, a report prepared for the Postal Service by
states:
“the present undifferentiated allocation of equipment
handling costs as ‘overhead’ needs review because,
with automation (and, for that matter, mechanization) as
distinct from manual processing, some mail classes alre
apparently more dependent on containerization
and
related handling equipment than others.“4
6
7
8
9
10
11
12
13
This observation
14
case, the costs of moving empty equipment)
15
machine-specific
16
Service in this case) results in a more accurate measurement
17
between “not handling
18
those costs.
clearly indicates that distributing
and operation-specific
“not handling mail” costs (in this
to subclasses
of mail on the basis of
cost pools (as proposed
mail” costs and the subclasses
by the Postal
of the relationship
of mail whic’h give rise to
This logic is not limited to the cost of moving empty equipment.
19
20
same report made a similar observation
21
component
22
of break time can be expected as automation
4.
The
for break time, another significant
of “not handling mail” costs; because “continuing
negotiated
is extended to previously
increases
non-
Overhead and Subclass Cost Study, prepared for the United States Postal
Service under Contract No, 104230-90-B-0505
by Foster Associates Inc.,
November 1992 (“Foster Associates Report”), page 5.
-9.
1
automated
situations,”
2
machine-specific
the cost of breaks should be distributed
and
cost pools, as proposed by the Postal Service.’
An additional
3
within operation
improvement
in the new cost methodology
4
mail distributions
5
Previous Postal Service practice allocated the costs of containers
6
shapes of mail in proportion
7
that different types of containers
8
On the other hand, Mr. Degen “exploits the association
9
shapes and/or subclasses
is that mixed
now reflect actual data on the contents of items alnd containers.
to the set of all direct mail tallies.
with mixed
This ignores the fact
are used for different types (subclasses)
of mail.
of item types within certain
of mail.“6 He does so by “using the corresponding
10
piece- or item-handling
11
for which the contents were identified as (a) items or (b) loose mail shapes.’
12
technique
13
shapes of mail by distributing
14
direct mail jn those item tvoes.’
15
identified,
16
types with different classes or subclasses
recognizes
distribution”
the relationship
by cost pool to allocate the costs of containers
This
between item types and certain classes or
the costs of uncounted
items in proportion
to the
For those containers for which the contents are not
Mr. Degen similarly makes use of the association
of different container
of mail and allocates non-identified
5.
Foster Associates
Report, page 5.
6.
USPS-T-l 2, page 10.
7.
USPS-T-12,
8.
For example, the cost of uncounted Blue & Orange sacks (used for Express
Mail) are distributed in proportion to the direct mail in Blue (1 Orange sacks.
LIOCATT would distribute those costs in proportion to 8!j direct mail,
ignoring the fact that Blue & Orange sacks are designated for Express Mail
use. See Tr. 12/6580.
pages 9-10.
-lO-
container
costs in proportion
to direct plus identified container
contents by cost
pool.
In summary, the Postal Service’s new methodology
(using cost pools
based on machine and operation type as well as counted mixed mail) is superior to
5
the old LIOCATT process.
6
information
7
Commission.
The new system takes advantage
for the mixed-mail distribution.“g
8
9
10
B.
11
Postal Inspection
of “more and better
It should be adopted by the
The Criticisms of MODS Piece Handling
Data Do Not Apply to Mr. Degen’s Use of
M.QDS Workhours Data.
Service audits have directed significant
data relied upon by Dr. Bradley.”
criticisms at
12
the MODS piece handlings
13
how crucial this piece handling data is to Dr. Bradley’s analysis and how its lack of
14
reliability
15
to which mail processing
calls into serious question Dr. Bradley’s conclusions
regalrding the degree
labor is other than 100 percent variable with volume.
It is important to understand
16
Dr. Neels discusses
that while Dr. Bradley’s analysis is
17
undermined
by these criticisms, Mr. Degen’s analysis is not affected by them. Mr.
18
Degen does not rely upon the MODS piece handling data in his analysis; he relies
19
only upon the employee workhours
20
processing
data from MODS in order to partition mail
labor costs into cost pools, as described
above. The MODS workhours
9.
Tr. 12/6421.
10.
National Coordination Audit: Mail Volume Measurement and Reporting
Systems, United States Postal inspection Service, December 1996, LR-H220.
-ll-
data are directly linked to the Postal Service’s payroll system, creating additional
accounting
and managerial
controls, and have been measured on ,the same basis
for at least nine years.”
In short, criticisms of the ability of MODS to measure piece handlings
have no bearing on Mr. Degen’s analysis since he does not use that data.”
6
7
C.
8
The improvements
9
Mr. Degen’s Distribution Method Should
Be Used With Minor Modifications.
the Postal Service has implemented
in distributing
the costs in Cost Segment 3 should be adopted whether or not the Commission
10
chooses to continue the long-standing
11
processing
12
increase in overhead/not
13
distribute mixed mail costs -- have no necessary
14
variability
15
can be easily adapted to incorporate
labor costs. Two improvements
100 plercent of mail
made by Mr. Degen -- addressing
the
handling mail tallies and refining the methods used to
of mail processing
Decoupling
16
practice of attributing
relationship
labor costs. The methodology
full attribution
Mr. Degen’s distribution
to the degree of
outlined by Mr. Degen
of mail process,ing labor costs.
key analysis from the Postal
17
Service’s proposal to abandon the historical attribution
level of maiil processing
18
labor costs does, however, require some small modifications.
The Commission
has
11.
Tr. 1115878.
12.
Some questions have been raised about the degree to which Postal
employees actually clock into the MODS operation in which they are
working, Postal supervisors have a strong incentive for ensuring the
accuracy of the workhours data, since different supervisors are responsible
for different operations. Mr. Degen has adequately responded to these
questions. See, m, Tr. 12/6554-56.
-12-
1
found, in very limited instances, that some mail processing
2
and not attributable.13
3
classified
4
in Cost Segment 3.1 must be reversed to ensure treatment consistent
5
Commission’s
6
Postal Service -- stratifying the mixed mail distribution
7
operational
8
mail -- are maintained
9
proposal with Dr. Neels’ recommended
10
attribution
13.
In addition, the “migration”
as administrative
established
characteristics
labor costs are fixed
of some costs previously
(and assigned to Cost Segment 3.3) but now included
practice.
The essential improvements
with the
introduced
by the
process on the basis of
and more fully utilizing actual data on counted mixed
in this approach.
Table 2 compares the Posital Service’s
treatment of Cost Segment 3, which returns
to 100 percent.
One example is “working, but not handling mail” while working on the
Platform (Activity Code 6210).
-13-
Table 2
1
BY 1996 Volume Variable Cost Segment
Subclass
2
3
4
5
6
7
8
9
IO
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
,--
-..- __-._Cards
xwece
- .__. -
I
Total First Class Mail
nmin
-.(.-.
1_,__-,--.
Priority Mail
- press Mail
~ilnrams
county
.egular rate publications
Jonprofit publications
.
Classroom ptmcanons
l-“+ll.I. Pwinfiicsl~
_..ll.__.l
ISingl.. ..i^^^ ?^,”
3.1 Costs by
I
14o,c--’
?Jr
6,033,074
477,606
83,202
79
15,210
467,201
81,970
--^-
_-- -__
M
JailI
I
Bulk - Regular Other
I
Total Standard (A) Regular
.-... ..-.. .-...-.._. .___..
.
.
-1 Other
I
Total Standard (A) Mail
Parcels (zone rate)
Bound printed matter
Special rate
353/
2,199,(
153,L-,
71,2471
65,485/
.- - .-
---, - - 98.25:
92,035
-- ---
a,%,,,.
10.37L
214,564
9.986.633
116,331
10,102,964
3.144,440
13,247,412
,
,_, ,,,capped
International Mail
IAll Mail
rotai Special Services
f
Total Volume Valuable
Other
Total Costs
F
Sources:
Postal Service Proposal - USPS-T-5, WP A-2, pages 3-4; 100% Athtbution
- UPS-Sellick-WP-I-K?., Mail Processing Adjustments Sheet.
-14-
BASE YEAR AND TEST YEAR
COST
1
2
I have calculated
3
Base Year 1996 (BY1996) and Test Year 1998 After
4
Rates (TYAR) costs with mail processing
5
To estimate the effect that changes in the level of attribution
6
of BY1996 mail processing
7
developed
8
costs change in the same proportion
9
particular, for each BY1996 cost component which changes as a result of
10
.
11
model.
Under this model, BY1996 to TYAR
as in the Postal Service’s proposal.
I make to Cost Segment 3, the following calculation
The TYAWBYI
In
is made:
996 ratio resulting from the Postal Service’s proposal
is calculated for each subclass; and
12
.
13
My revised BY1996 cost by subclass is then multiplied by the Postal
14
Service TYARlBYl996
15
A comparison
16
and in the distribution
labor costs in Cost Segment 3 have on TYAR costs, I
a simplified roll-forward
modifications
labor costs at 100 percent attributionq4
presented
14.
ratio to calculate the new TYAR costs.
of the Postal Service’s proposal with my results is
in Tables 3 (Base Year) and 4 (Test Year).
In so doing, I have used the Postal Service’s treatment of Alaska Air costs,
that is, Alaska Air is essentially 100% attributable to Parcel Post. The result
of using the Commission’s Docket No. R94-1 treatment of TYAR Alaska Air
costs is presented by the Postal Service in LR-H-215 (Rule 54(a)(l)
Alternate Commission Cost Presentation) (Rollforward) (Revised).
-15
Table 3
1
BY 1996 Volume
2
3
4
5
6
7
8
9
IO
11
12
13
14
15
16
17
lClass and Subclass
Variable
of Mail
Costs by Subclass
1 PostalService
1 Recommended
1.584,229
342,623
432
75,056
1,448,QO4
317,766
14,074
I,656600
Priority Mail
Express Mail
Mailgrams
Within county
Regular rate publications
Nonprofit publications
Classroom publications
Total Periodicals Mail
18
:i
21
22
23
24
25
--I ,~ou,9431
I
-._ .-.
7,993,908
I
26
27
28
29
30
694.9971
285,041’
226,--
rate)
Id
printed
matter
I rate
31
32
33
34
35
36
37
38
39
40
Sources:
Postal Service Case - Exhibit USPSdA, pages 7-8; Recommended
- UPS-Sellick-WP-I-Cl,
Base Year Costs Sheet.
-16-
Approach
1
Table 4
Test Year 1998 Volume Variable
2
r
Class and Subclass of Mail or
Soecial Service
&em and sealed parcels
Presort letters and sealed parcels
Single Piece Cards
Presort private post cards
Total First Class Mail
Costs by Subclass
Postal Service
CZSS!
$12,466,9681
4,1
z__-,
15.53721
17.060.0151
I
1
Recommended
Annrnaeh
_-_..
11 A?, 40,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
oarcels (zone rate)
Bound printed matter
Special rate
Library rate
Total Standard (6) Mail
31
32
33
34
35
36
37
30
Penalty - U. S. Postal Servtce
Free Mail for Blind/Handicapped
International Mail
Total All Mail
Total Special Services
Total Volume Variable
39
40
41
Sources:
731,138
328,929
254,900
40,569
1,363.534
I
172,9261
31,4:
1,l’
33,o. -,-- .,
1 “-‘ *-- I
I
Postal Service Case - USPS-T-15, WP-G. Table D. pages 7-6, adjusted
for misallocation of Phase I PMPC contract, Tr. 1317293-96; Recommended
Approach - UPS-Sellick-WP-I-I-Cl,
TYAR Summary Sheet.
-17-
These revised TYAR costs are used by UPS witness .J. Stephen
Henderson
(UPS-T-3)
to develop his pricing proposals for certain subclasses
of
mail.
PRIORITY MAIL PROCESSING COST
DIFFERENCES BY SHAPE
4
5
The Postal Service’s own data show that Priority Mail parcels are, on
6
7
average, more expensive
The SAS program MODSHAPE
8
9
to process than are Priority Mail flats.
in LR-H-146 calculates
shape for selected BASE YEAR rate categories”
“costs by
using the new MODS cost pools
IO
for mail processing
11
not include costs by shape for Priority Mail, the MODSHAPE
12
modified to include Priority Mail costs by shape in its output.‘6
13
Postal Service has made this calculation
14
modification
15
extract from the Postal Service’s data the results for Priority Mail in addition to the
16
results the Postal Service calculates
program is easily
but has not presented
Essentially,
for other subclasses
My
I simply
of mail.
table shows the resulting mail processing
costs by
Shape for Priority Mail (TY 1998):
15.
LR-H-146.
Part Ill, pages Ill-2 through 111-15.
16.
See UPS-Sellick-WP-I-III-C
for the details of the modifications
MODSHAPE needed to make this calculation.
-18-
the
the results.
uses the Postal Service’s data and analytic techniques;
The following
17
18
costs.15 While the output provided by the Postal Service does
to
1
Table 5
2
Mail Processing
Costs by Shape for Priority Mail (TY 1998)
Mail Processing
Labor Costs 100% Attributable
3
Mail Shape
4
5
Flats
6
IPPs & Parcels
7
Difference
BY 1996 Volume
BY 1996 Mail
Processing Cost
$214,628
344,192
$442,427
I
589,192
I
I
$0.751
a
9
10
11
12
13
Adjusted TY Difference
Source:
UPS-Sellick-WP-I-III-A,
page 2.
This mail processing
14
cost difference
15
Priority Mail parcels is used by Mr. Luciani
16
surcharge
between Priority Mail flats and
in proposing
a Priority Mail parcel
17
18
RECALCULATION
OF DBMC NON-TRANSPORTATION
COSTS AVOIDFD IN OUTGOING OPERATIONS-
19
In his Exhibit C, Postal Service witness Crum (USPS-T-28)
20
to estimate the test year outgoing
21
Parcel Post. He calculates
23
used by the Commission
24
particular,
unit costs avoided by DBMC
avoided costs of 37.7 cents per piece,
In his calculation,
22
mail processing
attempts
Mr. Crum uses a methodology
different from that
and the Postal Service in previous proceedings.
the Commission’s
established
methodology
-lQ-
In
excludes the costs for Mail
I
1
Preparation
(Operation
Code 01) and Platform Acceptance
2
calculating
3
treats these costs as part of the costs avoided by DBMC Parcel Post. Also, a
4
premium pay adjustment
5
made by Mr. Crum. In his testimony, Mr. Luciani recommends
6
Commission’s
Code 07) in
the costs avoided by DBMC Parcel Post. Mr. Crum, on the other hand,
has traditionally
methodology
been made by the Commission,
but is not
that the
should be adopted in this case.
In response to an interrogatory
7
(Operation
asking why he did not adjust his
8
avoided cost calculation
9
costs, Mr. Crum indicated that “it would not have been possible to make the
and platform acceptance
However, the SAS data sets in LR-H-146
contain the data
10
adjustments
11
needed to make these adjustments.
12
table also shows the amount of the premium pay adjustment traditionally
13
the Commission.
17.
as such.“”
to exclude mail preparation
The results are presented
Tr. 5/2285.
-2o-
in Table 6. This
made by
Table 8
Parcel Post Costs Excluded from
DBMC Avoided Cost Calculation
1
2
3
4
Attribution
rcent Attribution
of C’ost Segment 3
of Cost
$5,867
$51,187
$1,295
10
Source: UPS-Sellick-WP-1
-IV-A, page 1.
11
Mr. Luciani uses these calculations
SUMMARY
12
In conclusion,
13
14
to arrive at a revised DBMC discount.
.
OF CONCLUSIONS
I find that:
Mr. Degen’s MODS-based
approach to distributing
15
processing
labor costs to subclasses
16
practice and should be adopted by the Commission.
17
approach
is an improvement
more closely aligns the distribution
overhead costs to mail processing
attributable
mail
over past
Mr. Degen’s
of mixed mail and
operational
characteristics
and
19
more fully utilizes Postal Service data on counted mixed mail. The
20
result is an improved distribution
21
.
MODS-based
of the costs in Cost Segment 3.
costing can be implemented
22
historical practice of attributing
23
costs. Mr. Degen’s MODS-based
-21-
while returning to the
100 percent of mail processing
labor
approach should be adopted by the
1
Commission.
2
analysis are provided in my testimony.
3
.
The Base Year and Test Year results oif such an
Extraction of existing data based on the Postal Service’s own analytic
4
techniques
5
more expensive to process than are Priority Mail flats., This data is
6
presented
in my testimony and is used by Mr. Luciani to develop a
7
surcharge
for Priority Mail parcels.
a
9
.
demonstrates
that Priority Mail parcels are, on average,
The data are available to revise the Postal Service’s computation
the non-transportation
of
costs avoided by DMBC in outgoing operations
10
in accordance
with previous Commission
and Postal !$ervice practice.
11
These data are presented
12
Luciani to calculate a revised DBMC discount.
in my testimony and are used by Mr.
-22-