D.C. Area Mail Delivery Service: Resolving Labor-Relations and
Operational Problems Key to Service Improvement (Letter Report, 02/23/95,
GAO/GGD-95-77).

Since 1990, mail service in the Washington, D.C., metropolitan area has
been among the worst in the nation, and customer satisfaction has
generally been below the national average.  Both mail service and
customer satisfaction in the D.C. area declined dramatically in 1994.
Postal Service officials attribute the deterioration in service to an
unexpected growth in mail volume in early 1994.  Local units were unable
to handle this growth due to employee shortages, recent organizational
change, mail-handling process problems, and poor labor-management
relations. The Postal Service has made progress toward restoring service
to 1993 levels, but breakthrough improvement will require postal
management and labor unions to work together to resolve long-standing
employee relations problems that are reported to be more severe in the
D.C. area than in most other locations.  GAO summarized this report in
testimony before Congress; see: Postal Service: Mail Delivery Service in
the Washington Metropolitan Area, by J. William Gadsby, Director of
Government Business Operations Issues, before the Subcommittee on
Treasury, Postal Service, and General Government, House Committee on
Appropriations.  GAO/T-GGD-95-94, Feb. 28, 1995 (15 pages).

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  GGD-95-77
     TITLE:  D.C. Area Mail Delivery Service: Resolving Labor-Relations 
             and Operational Problems Key to Service Improvement
      DATE:  02/23/95
   SUBJECT:  Postal service employees
             Mail delivery problems
             Postal facilities
             Mail transportation operations
             Labor-management relations
             Postal service
             Personnel management
             Working conditions
             Job satisfaction surveys
IDENTIFIER:  USPS Customer Satisfaction Index
             USPS External First-Class Measurement System
             District of Columbia
             USPS Employee Opinion Survey
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to Congressional Requesters

February 1995

D.C.  AREA MAIL DELIVERY SERVICE -
RESOLVING LABOR-RELATIONS AND
OPERATIONAL PROBLEMS KEY TO
SERVICE IMPROVEMENT

GAO/GGD-95-77

D.C.  Area Mail Delivery Service

(240163)


Abbreviations
=============================================================== ABBREV

  CDVF - City delivery volume feet
  CSI - Customer Satisfaction Index
  EOS - Employee Opinion Survey
  EXFC - External First-Class
  P&DC - Processing and distribution center
  ULTRA - Unit-Load Tracking Architecture

Letter
=============================================================== LETTER


B-260033

February 23, 1995

The Honorable Jim Lightfoot
Chairman, Subcommittee on Treasury, Postal
 Service, and General Government
Committee on Appropriations
House of Representatives

The Honorable Steny H.  Hoyer
Ranking Minority Member, Subcommittee on Treasury,
 Postal Service, and General Government
Committee on Appropriations
House of Representatives

The Honorable Paul S.  Sarbanes
United States Senate

The Honorable Frank R.  Wolf
House of Representatives

This report responds to your requests that we review mail delivery
service in Washington, D.C.; Northern Virginia; and Southern
Maryland.  Your request followed public disclosure of mail delivery
problems in the Washington, D.C., metropolitan area and third
quarter\1 fiscal year 1994 delivery scores that showed service in the
Washington metropolitan area to be among the worst in the nation. 

Specifically, you asked that we (1) document the recent history of
on-time mail delivery service problems for overnight, First-Class
Mail in the Washington, D.C., metropolitan area; (2) determine the
reasons why mail service was below the desired level; and (3)
identify any Postal Service actions to improve service. 


--------------------
\1 Throughout this report, references to quarterly data refer to
postal quarters, which make up the postal fiscal year.  The postal
fiscal year starts and ends during September of each year and is made
up of 13 four-week accounting periods.  Postal quarters 1 through 3
include 3 accounting periods each.  Postal quarter 4 includes the
last 4 accounting periods. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

Since 1990, mail service in the Washington, D.C., metropolitan area
has been among the worst in the nation.  Also, customer satisfaction
has generally been below the national average.  Both mail service and
customer satisfaction in the Washington, D.C., metropolitan area
declined dramatically in 1994. 

According to Postal Service officials, mail delivery in the
Washington area deteriorated in early 1994 due in part to unexpected
growth in mail volume.  Local units were unable to handle this growth
and maintain mail service and customer satisfaction at previous
levels because of a complex set of factors that included employee
shortages, a recent organizational change, mail handling process
problems, and poor labor-management relations. 

The Postal Service has taken steps designed to solve the immediate
problems that resulted from the unexpected growth in mail volume as
well as the mail handling process problems.  The Postal Service has
made progress toward restoring service to 1993 levels, but
breakthrough improvement will require that postal management and
labor unions work together to address long-standing employee
relations problems that are reported to be more severe in the
Washington, D.C., metropolitan area than in most other locations. 


   BACKGROUND
------------------------------------------------------------ Letter :2

The Postal Service's goal is to deliver at least 95 percent of local
First-Class Mail overnight and to achieve 100-percent customer
satisfaction.  Delivery performance is measured in 96 metropolitan
areas across the nation and results are published quarterly.  This
measurement system, known as the External First-Class Measurement
System (EXFC), is based on test mailings done by Price Waterhouse. 
For the Washington, D.C., metropolitan area,\2 EXFC results are
available separately for Washington, D.C.; Northern Virginia; and
Southern Maryland.  Nationwide averages are also available for
comparison purposes. 

Customer satisfaction is measured in 170 metropolitan areas across
the nation, and results are also published quarterly.  This
measurement system, known as the Customer Satisfaction Index (CSI),
is administered by the Opinion Research Corporation.  Each quarter it
mails a questionnaire to thousands of households asking them how they
would rate their overall satisfaction with the Postal Service's mail
service.  For the Washington, D.C., metropolitan area, CSI results
are available separately for Washington, D.C.; Northern Virginia;
Southern Maryland; and Suburban Maryland.\3 The processing and
distribution facility for Southern Maryland is located in Prince
George's County.  The facility for Suburban Maryland is located in
Montgomery County.  Nationwide averages are also available for
comparison purposes. 

The Postal Service said that its delivery service and customer
satisfaction goals--nationwide and locally--are ambitious, and
attaining those goals will require a high level of employee
commitment.  For example, the quarter 4, 1994, EXFC nationwide
average was 12 percentage points below the established goal. 

To gauge employee attitudes and satisfaction levels, the Service has
administered a questionnaire to all employees in each of the last 3
years.  This questionnaire is commonly known as the Employee Opinion
Survey (EOS), and survey results are available for the nation, broken
down by local postal facility. 

In conducting our review, we (1) obtained and analyzed numerous
Postal Service reports containing data on factors affecting mail
processing and delivery; (2) obtained and analyzed numerous types of
performance data for both the local Washington, D.C., area and the
nation, as well as for other selected locations; (3) interviewed
various postal and union officials; (4) observed mail processing
operations at local processing and distribution centers and local
postal stations; and (5) examined recent reports on mail service
issued by the Postal Service's Inspection Service and the Surveys and
Investigations Staff of the House Committee on Appropriations. 
(Additional background information and more details on our
objectives, scope, and methodology are presented in appendix I.)


--------------------
\2 Mail service for the Washington, D.C., metropolitan area includes
Arlington, Fairfax, Loudoun, Prince William, Fauquier, Culpeper,
Rappahannock, Frederick, Warren, Clarke, Madison, and parts of
Shenandoah, Stafford, Orange, and Page counties in Virginia.  It also
includes Montgomery, Prince George's, Calvert, Charles, St.  Mary's,
and parts of Anne Arundel and Howard counties in Maryland, as well as
the District of Columbia. 

\3 While CSI reports a separate score for Suburban Maryland, EXFC
does not. 


   THE PROBLEM
------------------------------------------------------------ Letter :3

Mail service and customer satisfaction in the Washington, D.C.,
metropolitan area have consistently been below stated goals;
generally below the national average; and, in 1994, substantially
below the levels attained in 1993.  Specifically, service in the
Washington metropolitan area, as measured quarterly by EXFC,\4

has been below the national average in 16 of the 17 quarters since
EXFC was first established in 1990.  The national average ranged
between 79 and 84 percent in that time period but has always been
below the 95-percent on-time delivery goal.  Figure 1 compares mail
delivery service in the Washington metropolitan area, over time, with
the national average and delivery service goal. 

   Figure 1:  EXFC First-Class
   Overnight Delivery, Nationwide
   and in the Washington, D.C.,
   Area

   (See figure in printed
   edition.)

Source:  Postal Service data. 

Further analysis of EXFC data showed that delivery scores in the
Washington, D.C., metropolitan area have been among the worst in the
nation.  For example, 88 percent of the time, service in Northern
Virginia and Southern Maryland was in the bottom 25 percent of all
locations where service was measured; 76 percent of the time, service
in Washington, D.C., was in the bottom 25 percent.  Additionally,
delivery service scores in the Washington, D.C., metropolitan area
for quarter 4, 1994, were significantly below the scores attained for
quarter 4 the previous year.\5 Southern Maryland's score, for
example, dropped 8 percentage points. 

Residential customer satisfaction in much of the Washington, D.C.,
metropolitan area, as measured by CSI, has generally been below the
national average.  (See figure 2.\6 ) Since 1991, the Opinion
Research Corporation has sent CSI questionnaires to postal customers
on a quarterly basis asking them how satisfied they were with mail
service.  Information collected during these 16 quarters show that in
each quarter between 85 and 89 percent of customers nationwide rated
their satisfaction with the Service's overall performance as
excellent, very good, or good.  In 12 of 16 quarters, Northern
Virginia customers reported being as satisfied, or more satisfied,
than the nation as a whole.  Customer satisfaction in the other
locations that make up the metropolitan area--Southern Maryland;
Washington, D.C.; and Suburban Maryland--was lower.\7 For example,
Washington, D.C., customers rated the Postal Service lower than the
national average in all 16 quarters. 

   Figure 2:  National and
   Washington, D.C., Area
   Quarterly CSI Results Since the
   Inception of the CSI Program

   (See figure in printed
   edition.)

Source:  Postal Service data. 

Further analysis of CSI scores showed that customer satisfaction was
lower in all parts of the Washington, D.C., metropolitan area during
quarter 4 of fiscal year 1994 than during comparable periods in 1991,
1992, and 1993.  (A detailed discussion of mail service conditions in
the Washington, D.C., metropolitan area is presented in appendix II.)


--------------------
\4 EXFC measures delivery time between the scheduled pickup of mail
at collection boxes or post offices and the receipt of that mail in
the home or business. 

\5 Recently released EXFC scores for quarter 1, 1995, show some
improvement in delivery service.  The national EXFC score increased
from 83 in quarter 4, 1994, to 84.  Scores for Washington, D.C.;
Southern Maryland; and Northern Virginia increased from 70, 73, and
76 to 73, 74, and 77, respectively. 

\6 The recently released national CSI score for quarter 1, 1995, was
85 percent.  This score represents no change from the previous
quarter.  CSI scores for Washington, D.C.; Southern Maryland;
Suburban Maryland; and Northern Virginia decreased from 67, 77, 80,
and 82 to 55, 71, 72, and 74, respectively. 

\7 CSI was at least 3 percentage points below the national average in
all 16 quarters for Washington, D.C.; in 6 quarters for Southern
Maryland; in 3 quarters for Suburban Maryland; and in 1 quarter for
Northern Virginia.  In the fourth quarter of 1994, Washington, D.C.,
was 18 percentage points below the national CSI average; Southern
Maryland was 8 points below average; Suburban Maryland was 5 points
below average; and Northern Virginia was 3 points below average. 


   THE CAUSES OF THE PROBLEM AND
   POSTAL SERVICE ACTIONS
------------------------------------------------------------ Letter :4

Mail service in the Washington, D.C., metropolitan area is poor for a
number of reasons, including (1) the Postal Service's inability to
effectively deal with the unexpected growth in mail volume, (2) mail
handling process problems, and (3) labor-management problems.  Over
the past few months, the Postal Service has initiated additional
actions in each of these areas in an effort to improve mail service. 


      INABILITY TO RESPOND
      EFFECTIVELY TO UNEXPECTED
      MAIL VOLUME GROWTH
---------------------------------------------------------- Letter :4.1

In 1994, the percentage increase in the amount of mail delivered in
the Washington, D.C., metropolitan area was twice the national
average.  According to Postal Service officials, the Postal Service
had not anticipated this growth and was unprepared to process and
deliver the increased volume of mail. 

Complicating the situation were several factors that worked against
the Postal Service.  First, according to Postal officials, local
processing and delivery units experienced staffing problems because
more craft people than expected accepted a retirement incentive
(buyout) of up to 6 months' salary and left the Service during the
1992 restructuring.  Also, staffing ceilings were put into place in
anticipation of more automation equipment.  These events, according
to Postal officials, left the delivery units with too few people to
handle the increased volume of mail.  Additionally, the processing
units were operating with too many unskilled, temporary employees who
had been hired to replace more costly career employees who retired in
1992.  Training also became an issue when some new supervisors were
placed in jobs where they were not familiar with the work of the
employees they were supervising.  After considerable attention was
focused on these problems in the spring of 1994, the Postal Service
took steps to hire new, permanent employees and strengthen training
for supervisors and craft personnel. 

Second, to focus additional attention on customer service, separate
lines of reporting authority were established for mail processing and
mail delivery functions under the Executive Vice President/Chief
Operating Officer during the 1992 restructuring.  This realignment of
responsibilities was done as part of the Postmaster General's broad
strategy to make the Postal Service more competitive, accountable,
and credible.  This action left no single individual with the
responsibility and authority to coordinate and integrate the mail
processing and delivery functions at the operating levels of the
organization.  The primary focus of each of the function managers was
to fulfill the responsibilities of his or her function.  Working with
the other function managers became a secondary concern. 
Consequently, because critical decisions affecting both mail
processing and customer services could not be made by one individual
at the operating level of the organization, coordination problems
developed. 

In June 1994, the Postmaster General moved responsibility for
processing and delivery down to the Area Vice President level, and on
January 10, 1995, postal officials announced plans for establishing a
position under the Mid-Atlantic Area Vice President that would be
responsible for overseeing all processing and delivery functions in
the Washington, D.C., metropolitan area and Baltimore area. 

Time slippages in the automation program was another factor that
affected the Postal Service's ability to handle the increased volume
of mail.  More mail than planned had to be processed manually or on
mechanical letter-sorting machines.  The Postal Service had expected
that by 1995 almost all letter mail would be barcoded by either the
Postal Service or mailers and be processed on automated equipment. 
However, automation fell behind schedule in 1993-1994.  The new
projected date for barcoding all letter mail has slipped to the end
of 1997.  (A detailed discussion of the Postal Service's inability to
respond effectively to the unexpected mail volume growth in the
Washington, D.C., metropolitan area is presented in appendix III.)


      MAIL HANDLING PROCESS
      PROBLEMS
---------------------------------------------------------- Letter :4.2

Delivery service in the Washington metropolitan area was also
adversely influenced by various mail handling process problems,
including (1) the unnecessary duplicative handling of much mail
addressed to Northern Virginia, (2) overnight service areas that
managers believed were geographically too large, (3) mail arriving
too late for normal processing, (4) the absence of a control system
for routinely pinpointing the specific causes of delays in specific
pieces or batches of mail, and (5) failure of employees to follow
prescribed processing procedures. 

The Postal Service has taken action to address, at least in part,
each of these problems.  Some of the more significant actions taken
include (1) reducing the amount of mail handled by more than one
processing facility in Northern Virginia, (2) processing more mail at
local facilities rather than transporting it to distant processing
and distribution centers, (3) working with the large mailers to get
them to mail earlier in the day and give advance notice when mailing
unusually large volumes, (4) taking the first steps to develop a
system that can pinpoint causes of delayed mail, and (5) requiring
greater adherence to established operating procedures.  Additionally,
a number of service improvement teams are continuing to examine mail
handling processes in an effort to identify other areas needing
improvement. 

Examples provided by local postal officials that most clearly
illustrate problems affecting the local area are discussed below. 

Duplicative mail handling:  Much mail sent to the Northern Virginia
area was delayed because it was processed by both the Dulles and
Merrifield facilities.  Further delays also occurred because of the
time lost transporting mail between the two facilities.  Duplicative
mail handling occurred because the Dulles and Merrifield facilities
are jointly responsible for certain ZIP Code service areas and most
facilities sending mail to Northern Virginia did not separate the
mail between the two facilities.  There is no easy way to split up
the service areas between the two facilities geographically--it would
require realigning and changing some ZIP Codes.  That option had not
been vigorously pursued because of the adverse reaction from
customers anticipated by the Service.  However, the Postal Service
recently began working with major feeders of overnight mail to work
out an interim solution--i.e., the feeder facilities are to sort mail
more completely before sending it to the Merrifield and Dulles
facilities.  Additionally, the Postal Service, in commenting on a
draft of this report, said that it will be installing a Remote Bar
Coding System site at the Dulles processing and distribution center
(P&DC) that, along with other processing changes, will virtually
eliminate the need for duplicative handling of mail for some Northern
Virginia ZIP Codes. 

Overnight service areas that are too large:  Consistent overnight
delivery service in some parts of the Washington, D.C., metropolitan
area is difficult to achieve because some service areas may be too
large for the current collection, transportation, and delivery
network.  For example, mail from some of the outlying areas in the
service area--e.g., Leonardtown and California, Maryland--does not
arrive at the Southern Maryland processing facility until 10:00 or
11:00 p.m.  This severely compresses the amount of time available for
processing the mail and getting it back out to the post offices in
time for delivery the next day.  To address this problem, the Postal
Service plans to process mail from Leonardtown and California, in
addition to other Southern Maryland areas, at a closer facility in
Waldorf (Charles County), Maryland.  Additionally, the Postal Service
is installing more "local only" collection boxes, which should reduce
the amount of mail that has to be transported to distant processing
and distribution centers. 

Mail arriving too late for timely processing:  Large quantities of
mail are frequently entered into the mail stream significantly past
the times established for normal processing.  This would not be a
problem, however, were it not for the expectation that deliveries
would be made the next day.  Managers told us they have few options
other than to accept late-arriving mail and then rush to meet
dispatch times.  They said that to do otherwise would upset the
delicate balance between providing customer service and meeting
established time schedules.  To help establish a more orderly
workflow, the Postal Service has been actively working with large
mailers in the area to get them to mail earlier in the day and also
to notify the Postal Service ahead of time when large mailings are
expected to arrive.  (A detailed discussion of all five mail handling
process problems and corrective actions taken is presented in
appendix IV.)


      LONG-TERM SOLUTIONS INVOLVE
      ADDRESSING POOR
      LABOR-MANAGEMENT RELATIONS
---------------------------------------------------------- Letter :4.3

In addition to academic studies, EOS, EXFC, and CSI survey results
indicated that a relationship exists between employee attitudes and
service performance.  Employee attitudes about postal management in
most of the facilities in the Washington, D.C., area, like employee
attitudes in many other big cities, were in the bottom 25 percent of
units nationwide.  Similarly, EXFC and CSI scores for Washington,
D.C., and other big cities were also relatively low compared to other
areas of the country. 

Disruptive workforce management problems were more prevalent in the
Washington, D.C., metropolitan area than in most other parts of the
country.  Postal Service data showed that employees in the
Washington, D.C., metropolitan area experienced greater than average
use of sick leave and a higher-than-normal use of work assignments
with limited/light duties for employees who, due to physical
restrictions, are unable to perform normal duties.  Managers told us
that excessive use of sick leave and limited/light duty assignments
indicate possible abuse and result in lower productivity. 

Those managers believed, and EOS tended to support the view, that
excessive employee absences and unavailability for regular duties
were often the result of substance abuse and poor employee attitudes. 
EOS data suggested that employees in the Washington, D.C.,
metropolitan area perceived a greater than average level of substance
abuse and had more negative attitudes about postal management than
employees in most other locations nationwide. 

Postal management recognizes that improving employee attitudes and
attendance is critical to improving delivery performance and customer
satisfaction.  However, the Postal Service cannot improve employee
attitudes and attendance unilaterally.  Successful change will
require the support and cooperation of employees and their unions. 
The need for joint cooperation was pointed out in our recent report
on Postal Service labor-management relations.\8

The Postmaster General has initiated a number of actions to improve
this relationship.  For example, he recently invited all the parties
representing postal employees to attend a national summit and commit
to reaching, within 120 days, a framework agreement for addressing
labor-management problems.  The rural carriers union and the three
management associations accepted the invitation.  However, the
leaders of the three largest postal unions had not accepted as of
December 31, 1994.  They said they would wait until the current round
of contract negotiations is completed before making a decision on the
summit.  (A detailed discussion of labor-management relations is
presented in appendix V.)


--------------------
\8 U.S.  Postal Service:  Labor-Management Problems Persist on the
Workroom Floor (GAO/GGD-94-201 A/B, Sept.  1994). 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :5

The Postal Service provided written comments on a draft of this
report.  It recognized the need to improve service and highlighted
its continuing efforts to produce significant improvements in
customers' satisfaction with their mail service.  The Postal Service
said that it was continuing to move ahead with numerous improvements
in the area's mail processing and distribution centers.  For example,
it cited the installation of the Remote Bar Coding System site at the
Dulles P&DC to help resolve the duplicative handling of some mail
addressed to Northern Virginia.  It also cited efforts to begin
processing more mail at the Waldorf (Charles County), Maryland
facility in order to improve service in Southern Maryland. 
Additionally, the Postal Service said that it was looking into
diagnostic technologies as a means of improving its ability to
identify underlying causes of delayed mail. 

The Postal Service said that new supervisors are receiving the
training they need, and that the Service is continuing to hire more
letter carriers and mail handlers and to place them where they are
most needed.  The Postal Service further said that through the
outstanding work of thousands of dedicated employees, it was turning
the corner in providing quality service in the Washington, D.C.,
metropolitan area.  It said that the actions taken are beginning to
produce results and cited, as an example, the improved EXFC scores
attained during the first quarter of 1995. 

The Postal Service agreed with our conclusion that improving
labor-management relations is a key element in any long-term solution
to mail service problems.  It said that efforts in this area must
include correcting problems that arise from a collective bargaining
process that is not working.  Further, it said that postal unions and
postal management must work together to change this process. 

Where appropriate, the Postal Service's comments have been
incorporated into the text of this report.  Its comments, in total,
are included as appendix VI. 


---------------------------------------------------------- Letter :5.1

As arranged with your office, unless you publicly announce the
contents of this report earlier, we plan no further distribution
until 30 days from the date of this letter.  At that time, we will
distribute copies of the report to the Postmaster General, other
House and Senate postal oversight committees, and other interested
parties.  Copies will also be made available to others upon request. 

Major contributors to this report are listed in appendix VII.  If you
have any questions about the report, please call me on (202)
512-8387. 

J.  William Gadsby
Director, Government
 Business Operations Issues


BACKGROUND AND OBJECTIVES, SCOPE,
AND METHODOLOGY
=========================================================== Appendix I


   BACKGROUND
--------------------------------------------------------- Appendix I:1

In fiscal year 1994, the Postal Service delivered about 177 billion
pieces of mail nationwide.  About 94 billion, or 53 percent, was
First-Class Mail.  Revenues for all classes of mail totaled about $50
billion in fiscal year 1994.  Revenue from First-Class Mail totaled
about $29.4 billion--approximately 59 percent of total revenue. 

The Postal Service field organization comprises 10 service areas. 
The Mid-Atlantic Area provides service to the Washington, D.C.,
metropolitan area and surrounding states.  (See figure I.1.)

   Figure I.1:  Map of Geographic
   Area Covered by the
   Mid-Atlantic Service Area

   (See figure in printed
   edition.)

Source:  Postal Service. 

The Mid-Atlantic area is subdivided into nine performance clusters. 
The Northern Virginia and Capital performance clusters provide mail
service for the Washington, D.C., metropolitan area.\9 The Northern
Virginia cluster consists of two mail processing and distribution
centers (P&DC), one of which is at Merrifield, Virginia, and one at
Dulles International Airport; and the Northern Virginia customer
service district.\10 The Capital cluster consists of three P&DCs,
with one each in Capitol Heights, Maryland (Southern Maryland);
Gaithersburg, Maryland (Suburban Maryland); and Brentwood
(Washington, D.C.); as well as the Capital customer service district. 
The Mid-Atlantic Area Vice President is responsible for day-to-day
management of the Mid-Atlantic Area. 

Efficient collection, processing, and transportation of mail are
critical to timely mail delivery and customer satisfaction.  Most
processing is done at P&DCs, which (1) distribute most local mail to
post offices for delivery and (2) dispatch nonlocal mail to other
postal facilities for further sorting and distribution.  The types of
mail processing operations include (1) high-speed processing on
automated equipment, (2) mechanized processing on letter sorting
machines, and (3) manual sorting.  Automated processing is the most
efficient of the three methods, and its use is increasing as more
automated equipment is installed. 

The Postal Service's goal is to deliver at least 95 percent of its
First-Class Mail within the following timeframes:  (1) overnight for
First-Class Mail originating (being sent) and destinating (being
received)\11 within the local delivery area defined by the Postal
Service; (2) 2 days (generally) for First-Class Mail traveling
outside the local area, but within 600 miles; and (3) 3 days for all
other domestic First-Class Mail.  Nationwide, during the fourth
quarter of fiscal year 1994, the Postal Service delivered about 83
percent of its overnight mail, 74 percent of its 2-day mail, and 79
percent of its 3-day mail within established delivery standards. 

The Postal Service has for several years sponsored measurement
systems--the External First-Class Measurement System (EXFC), the
Customer Satisfaction Index (CSI), and the Employee Opinion Survey
(EOS)--that have allowed assessments of its delivery performance, as
well as of customer and employee satisfaction.  The Service uses
information from these systems to identify areas needing improvement
and also publishes summary data that the Service and public can use
to hold management and employees accountable for Postal Service
performance. 


--------------------
\9 The Northern Virginia cluster includes Arlington, Fairfax,
Loudoun, Prince William, Fauquier, Culpeper, Rappahannock, Frederick,
Warren, Clarke, Madison, and parts of Shenandoah, Stafford, Orange,
and Page counties in Virginia.  The Capital cluster includes
Montgomery, Prince George's, Calvert, Charles, St.  Mary's, and parts
of Anne Arundel and Howard counties in Maryland, as well as the
District of Columbia. 

\10 Customer service districts, through an extensive network of local
post offices, stations, and branches, provide collection, delivery,
and retail services. 

\11 "Originating" and "destinating" are terms commonly used by the
Postal Service. 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
--------------------------------------------------------- Appendix I:2

Our objectives were to (1) document the recent history of on-time
mail delivery service problems for overnight First-Class Mail in the
Washington, D.C., metropolitan area; (2) determine the reasons why
mail service was below the desired level; and (3) identify any Postal
Service actions to improve service.  We did not review the Postal
Service's delivery performance for First-Class Mail outside the local
service area or for other mail classes (i.e., Express, second-,
third-, and fourth-class).  The Washington, D.C., metropolitan area,
as used in this report, includes the Northern Virginia and Capital
clusters. 

To accomplish our objectives, we obtained and analyzed numerous
Postal Service reports containing data on factors affecting mail
processing and delivery.  We examined national and local Postal
Service workhour reports, financial reports, and "FLASH" reports. 
FLASH reports provide, among other things, detailed information on
overtime, mail volume, the number of addresses where mail can be
delivered, sick leave usage, limited duty workhours, and the number
of hours spent on training. 

The reports generally covered 4-week accounting periods for fiscal
years 1991 through 1994.  They included information for the nation,
as well as for the Northern Virginia cluster, the Capital cluster,
and the units included in these two clusters.  Because of changes in
accounting and reporting in fiscal year 1993, we did not use 1993
data below the cluster level. 

We also obtained and analyzed numerous types of performance data for
the local Washington, D.C., area and for the nation, as well as for
other judgmentally selected locations.  These data included delivery
service scores as measured by the Postal Service's EXFC measurement
system, customer satisfaction scores as measured by CSI, and employee
opinions as determined by EOS.  These data covered fiscal years 1991
through 1994, except for EOS, which was conducted in 1992, 1993, and
1994.  In 1992, we reported that CSI was a statistically valid survey
of residential customer satisfaction with the quality of service
provided by the Postal Service.\12 We have not evaluated the validity
of the EXFC and EOS survey. 

We interviewed (1) the Chief Operating Officer/Executive Vice
President of the Postal Service; (2) the Vice President of the
Mid-Atlantic Service Area; (3) the customer service managers for the
Northern Virginia and Capital clusters; (4) the plant managers at
Merrifield, Brentwood, and Capitol Heights; (5) Inspection Service
officials responsible for audits of postal operations; and (6)
various other program and operations officials at headquarters, the
Mid-Atlantic area office, local P&DCs, and local delivery units.  We
also discussed the causes of mail delivery problems with
representatives from the National Association of Letter Carriers and
the American Postal Workers Union. 

Additionally, we observed mail processing operations at local P&DCs
and local postal delivery units.  We also obtained and analyzed
documentation on initiatives to improve service in the Washington,
D.C., metropolitan area, although we did not evaluate the
effectiveness of those initiatives.  We also reviewed recent reports
on mail service issued by the Inspection Service and the Surveys and
Investigations Staff of the House Committee on Appropriations. 

We requested comments on a draft of this report from the Postal
Service.  Written comments were received and are discussed on page 11
and included as appendix VI. 

We did our work from September 1994 to December 1994 in the
Washington, D.C., metropolitan area in accordance with generally
accepted government auditing standards. 


--------------------
\12 U.S.  Postal Service:  Tracking Customer Satisfaction in a
Competitive Environment (GAO/GGD-93-4, Nov.  1992). 


SEVERAL MEASURES SHOWED A HISTORY
OF BELOW-AVERAGE SERVICE FOR THE
WASHINGTON, D.C., METROPOLITAN
AREA
========================================================== Appendix II

The Postal Service's goal is to deliver 95 percent of the mail on
time as measured by EXFC and to achieve 100-percent customer
satisfaction as measured by CSI.\13 To date, however, the Postal
Service has fallen considerably short of those goals, both nationally
and in the Washington, D.C., metropolitan area. 


--------------------
\13 The Postmaster General, in remarks before the Board of Governors
in November 1993, set a goal of raising customer satisfaction from a
high of 89 percent in 1993 to 91 percent in 1994. 


   AREA EXFC SCORES HAVE
   CONSISTENTLY BEEN BELOW THE
   NATIONAL AVERAGE
-------------------------------------------------------- Appendix II:1

EXFC data show that mail delivery service in the Washington, D.C.,
area has consistently been among the worst in the nation.  EXFC is
administered under contract by Price Waterhouse and measures delivery
time between the scheduled pickup of mail at collection boxes or post
offices and the receipt of that mail in the home or business.  EXFC
test mailings are done in 96 metropolitan areas across the country. 
Results are published quarterly for overnight First-Class Mail. 
Within the Washington metropolitan area, EXFC delivery scores are
available for Northern Virginia, Southern Maryland, and Washington,
D.C. 

Since EXFC was first established in 1990, delivery scores for
overnight First-Class Mail in the Washington, D.C., metropolitan area
have, except for the first quarter reported (fourth quarter of 1990),
been below the national average, and the national average has always
been below the performance goal established by the Postal Service. 
(See figure 1.)

Our further analysis of EXFC scores showed that mail service in the
Washington, D.C., metropolitan area was not only below the national
average, but also was generally among the worst in the nation.  As
shown in table II.1, Northern Virginia, Southern Maryland, and
Washington, D.C., frequently ranked in the bottom 25 percent of the
metropolitan areas where delivery performance was measured.  Often,
these locations were in the bottom 10 percent. 



                               Table II.1
                
                     Relative EXFC Rankings For The
                         Washington, D.C., Area


Location                    Number     Percent      Number     Percent
----------------------  ----------  ----------  ----------  ----------
Washington,                     13         76%          11         65%
 D.C.
Southern                        15          88           8          47
 Maryland
Northern                        15          88           8          47
 Virginia
----------------------------------------------------------------------
Source:  Postal Service data. 

Note:  EXFC data were available, by quarter, from the fourth quarter
of fiscal year 1990 through the fourth quarter of fiscal year
1994--17 quarters. 

\a Out of a total of 17 quarters. 

EXFC data also showed that Washington metropolitan area delivery
service in fiscal year 1994 was generally below the levels of service
provided in fiscal years 1991 through 1993.  (See figure II.1.)
Northern Virginia was the exception.  Delivery service in Northern
Virginia was better in fiscal year 1994 than it was in 1991 and 1992,
but not as good as it was in fiscal year 1993. 

   Figure II.1:  Quarter 4 Trends
   of EXFC Scores for Washington
   Area Locations, 1991-1994

   (See figure in printed
   edition.)

Source:  Postal Service data. 

EXFC scores can be affected by the performance of neighboring P&DCs. 
For example, mail originating in Southern Maryland and going to the
District of Columbia passes through the Southern Maryland P&DC and
the Washington, D.C., P&DC (the destinating facility).  The time
taken is reflected in Washington, D.C.'s EXFC score, even though it
may have been delayed because of a problem at the Southern Maryland
P&DC.  Because of the impact other locations may have on individual
EXFC scores, we obtained and compared the test scores for
"turnaround" mail\14 in Northern Virginia, Southern Maryland, and
Washington, D.C., with the published EXFC scores for each of the
three locations where service is measured in the Washington area. 
Table II.2 shows that delivery scores for turnaround mail were higher
than the published EXFC scores, but still below the 95-percent
delivery performance standard. 



                               Table II.2
                
                  Published Overnight-First-Class Mail
                EXFC Delivery Scores and Turnaround Mail
                 Scores for Washington, D.C.; Southern
                Maryland; and Northern Virginia, Fiscal
                        Years 1992 through 1994


                                 Turn-           Turn-           Turn-
                                around          around          around
Postal quarter/fiscal     EXFC    mail    EXFC    mail    EXFC    mail
year                     score   score   score   score   score   score
----------------------  ------  ------  ------  ------  ------  ------
1-1992                      77      80      79      85      76      81
2-1992                      77      77      79      86      77      84
3-1992                      77      78      78      87      74      80
4-1992                      73      74      75      86      72      77
1-1993                      77      80      82      88      70      78
2-1993                      73      76      80      88      74      80
3-1993                      73      75      79      85      74      81
4-1993                      75      78      81      88      79      85
1-1994                      68      69      75      82      78      83
2-1994                      58      64      63      74      66      74
3-1994                      62      69      73      83      71      78
4-1994                      70      78      73      82      76      82
----------------------------------------------------------------------
Source:  Postal Service data. 


--------------------
\14 Turnaround mail is mail that is collected, processed, and
delivered within a single service area. 


   NORTHERN VIRGINIA, SUBURBAN
   MARYLAND, SOUTHERN MARYLAND,
   AND WASHINGTON, D.C., RESIDENTS
   EXPRESSED VARYING LEVELS OF
   CUSTOMER SATISFACTION
-------------------------------------------------------- Appendix II:2

Customer satisfaction with mail service, as measured by CSI, varied
among residents in Northern Virginia, Suburban Maryland, Southern
Maryland, and Washington, D.C.  In fiscal year 1991, the Postal
Service developed and implemented CSI to track residential customer
satisfaction.  CSI is administered under contract by Opinion Research
Corporation.  Each quarter since it was implemented, the contractor
has mailed a questionnaire to thousands of households throughout the
nation asking them how they would rate their overall satisfaction
with the Postal Service's performance (poor/fair/good/very
good/excellent).\15 The Postal Service publicly discloses quarterly
overall satisfaction ratings for 170 metropolitan areas, as well as
the nationwide average. 

The Postal Service began reporting quarterly CSI scores in the first
quarter of fiscal year 1991 for 40 metropolitan areas.  Since then,
the survey has been expanded to 170 locations.  Results from the
first survey showed that, nationally, 87 percent of customers thought
the Postal Service's overall performance was excellent, very good, or
good.  Since then, quarterly scores have ranged between 85 and 89
percent.  The CSI score for quarter 4, 1994, was 85 percent.\16

Among the 170 locations surveyed, customer satisfaction scores are
reported for four locations in the Washington, D.C., metropolitan
area:  Northern Virginia, Suburban Maryland, Southern Maryland, and
Washington, D.C.  Of these locations, as shown in figure 2, residents
of Northern Virginia gave the highest satisfaction rating on the
overall performance of the Postal Service.  In 12 of the 16 quarters
since the Postal Service began reporting CSI scores, Northern
Virginia's scores equalled or exceeded the national average. 
However, in 3 of the last 4 quarters reported, satisfaction
decreased, with scores falling 1 to 3 percentage points below the
national average. 

Suburban Maryland's postal customers were less satisfied.  In 9 of
the 16 quarters since the Postal Service began reporting CSI scores,
Suburban Maryland's scores fell below the national average.  Customer
satisfaction in Suburban Maryland decreased in the last 4
quarters--dropping from 90 percent in quarter 4, 1993, to 80 percent
in quarter 4, 1994. 

Southern Maryland postal customers have been less satisfied than
Northern Virginia and Suburban Maryland customers.  In fact, Southern
Maryland's score fell below the national average in 13 of the 16
quarters since quarter 1, 1991. 

Of the four local areas with CSI scores comprising the Washington,
D.C., metropolitan area, Washington, D.C., itself has been rated
lowest on overall performance.  In all 16 quarters since the Postal
Service began reporting CSI scores, Washington, D.C.'s scores were
lower than the national average.  In addition, its scores, like most
others, began to drop in quarter 4, 1993. 

Further analysis of CSI data showed that customer satisfaction in
Washington, D.C.; Southern Maryland; Suburban Maryland; and Northern
Virginia was lower in quarter 4, 1994, than it was in quarter 4 of
any of the preceding 3 fiscal years.  (See figure II.2.)

   Figure II.2:  Quarter 4 Trends
   of CSI Scores for Washington
   Area Locations, 1991-1994

   (See figure in printed
   edition.)

Source:  Postal Service data. 


--------------------
\15 The questionnaire also asks customers how they would rate their
satisfaction with 37 additional questions on specific service areas,
from letter carrier and window clerk service to parking space
availability. 

\16 The CSI score for quarter 1, 1995, remained at the 85-percent
level. 


SEVERAL FACTORS CONTRIBUTED TO THE
INABILITY TO RESPOND EFFECTIVELY
TO UNEXPECTED MAIL VOLUME GROWTH
========================================================= Appendix III

Postal officials cited the unexpected growth in mail volume in 1994
as one of the principal causes of the breakdown of delivery service
in the Washington, D.C., metropolitan area.  They said the Postal
Service was unable to respond to the unanticipated growth in volume
because (1) local delivery units had numerous unfilled vacancies and
the workforce at the processing and distribution centers comprised
many unskilled, temporary employees; and (2) an organizational change
had weakened management control over the span of processing and
delivery activities.  Timely processing and delivery of the mail were
further complicated because employee complement ceilings had been put
into place in anticipation of automation.  However, automation fell
behind schedule in 1993 and 1994. 


   GREATER THAN ANTICIPATED GROWTH
   IN MAIL VOLUME
------------------------------------------------------- Appendix III:1

Postal officials cited an unanticipated heavy mail volume in 1994 as
one of the principal causes for the slip in service performance, both
nationally and locally.  Nationally, mail volume grew by about 6
billion pieces between 1993 and 1994--a 3.5-percent increase. 

Mail volume data, in number of pieces, were not available below the
national level.  At the local delivery unit level, mail volume is
measured in feet.  This measure, referred to as city delivery volume
feet (CDVF), reflects the amount of mail delivered by carriers.  The
data showed that the rate of increase in the amount of mail delivered
by carriers in the Northern Virginia and Capital performance clusters
was about twice the rate of increase experienced nationwide.  (See
table III.1)



                              Table III.1
                
                Changes in CDVF Mail Volumes From Fiscal
                     Year 1993 to Fiscal Year 1994

                           (Feet in millions)

                                                  Northern
                                                  Virginia     Capital
                                    Nationwide     cluster     cluster
----------------------------------  ----------  ----------  ----------
Fiscal year 1993 volume
                                         722.5         7.4        13.1
Fiscal year 1994 volume
                                         760.4         8.2        14.7
Percentage increase fiscal year
 1993 to 1994                             5.2%       10.8%       12.2%
----------------------------------------------------------------------
Source:  Postal Service data. 

Postal Service officials said they had not anticipated that much
growth in volume either nationally or locally.  Furthermore, they
believed that any 1994 increase in volume could be handled without
increasing the workforce size because the deployment of additional
automated equipment would make processing and delivery more
efficient.  In retrospect, however, the Postal Service officials said
that staffing was inadequate and that automation was able to handle
only about half of the volume increase.\17


--------------------
\17 Inadequate planning and preparing for volume growth in the
Washington, D.C., metropolitan area have historically been a problem
for the Postal Service.  In 1989, we reported that the Postal
Service's inability to handle volume growth was a factor contributing
to deteriorated mail service in Northern Virginia.  See Postal
Service:  Progress Made in Restoring Deteriorated Northern Virginia
Mail Service (GAO/GGD-89-88, June 1989). 


   LOCAL UNITS EXPERIENCED
   EMPLOYEE SHORTAGES
------------------------------------------------------- Appendix III:2

According to Postal Service officials, a shortage of trained
employees contributed to poor mail service in the Washington, D.C.,
metropolitan area.  The shortage resulted from the loss of skilled
employees during the restructuring and buyout, hiring decisions based
on an unrealistic automation schedule, and some inadequately trained
supervisors. 

The Postal Service lost many skilled craft employees as a result of
the 1992 restructuring and buyout.  Nationally, 16,882 clerks, 11,933
city carriers, and 2,346 mail handlers took the buyout--about 5.8
percent of all employees in this group.  Additionally, more than
16,000 other employees also left the Service. 

In the Washington, D.C., area, 1,165 craft employees took the
buyout--about 6.6 percent of the craft employees in the local area. 
Employees in the Washington, D.C., area who took the buyout had an
average length of service of about 27 years.  In testimony before the
Subcommittee on Treasury, Postal Service, and General Government,
House Committee on Appropriations, the Postmaster General said that
in looking back at the 1992 restructuring, the Postal Service "let a
few too many people go, and .  .  .  cut too deeply in some
functional areas."

In planning the 1992 restructuring, the Postal Service had intended
to eliminate approximately 30,000 overhead positions that were not
involved in mail processing or delivery.  However, the Postmaster
General wanted to avoid a reduction-in-force, so he extended the
buyout offer to clerks, carriers, mail handlers, postmasters, and
others in order to open up vacancies for employees whose overhead
positions were eliminated but who were either not eligible or did not
want to retire.  Consequently, more than 47,000 employees opted for
the special retirement incentives offered in the Fall of 1992. 

This number was greater than the Postal Service had expected. 
However, officials viewed the loss as an opportunity to hire less
costly noncareer employees--who could later be terminated more easily
than career employees as more automation was moved into place. 

As the downsizing/restructuring got under way in the fall of 1992,
Members of Congress, mailers, and employee groups expressed
considerable concern about a possible adverse impact on mail delivery
service.  However, when compared to the same periods the previous
year, service nationwide and in Southern Maryland remained stable and
even showed signs of improvement immediately following the
restructuring.  EXFC scores for Washington, D.C., and Northern
Virginia, on the other hand, fell immediately following the
restructuring in comparison to the scores received during the same
period the previous year. 

By quarter 2, 1994, nationwide scores and scores for Washington,
D.C.; Southern Maryland; and Northern Virginia were below the scores
received for quarter 2, 1993, and in July 1994, the Vice President of
the Mid-Atlantic Area said that staffing had become a significant
problem in the Washington, D.C., area.  He noted that in December
1993, in preparation for additional automated sorting systems, the
Postal Service had put in place employee complement ceilings.  As a
result of this action, he said, delivery units struggled with
unfilled vacancies, and the processing and distribution centers had
to rely on a workforce with many unskilled, temporary employees. 

These problems were confirmed by local Washington, D.C., area postal
officials.  They said that because of the departure of many
experienced carriers, clerks, and supervisors during the
restructuring, the Postal Service's ability to quickly and accurately
sort and deliver mail in the Washington, D.C., area was adversely
affected.  They also agreed with the Vice President of the
Mid-Atlantic area that the shortage of career employees resulting
from the employee complement ceilings put in place in late 1993,
combined with the large number of unskilled, temporary employees,
adversely affected their ability to provide accurate, on-time
delivery service. 

Reacting to the staffing problems, the Vice President for the
Mid-Atlantic Area said that the Postal Service was placing emphasis
on obtaining adequate numbers of employees and making sure they were
in the right places at the right time.  As of July 1994, the Postal
Service had approximately 18,000 craft employees in the Washington,
D.C., metropolitan area.  Between that time and October 1994, 130 new
staff had been hired in Southern Maryland, including 55 letter
carriers, 40 clerks, and 35 mail handlers.  In Suburban Maryland, the
Postal Service had hired 62 new letter carriers and 34 clerks.  In
Northern Virginia, 300 new employees had been hired, half of whom
were letter carriers.  In Washington, D.C., 168 letter carriers, 30
clerks, and 31 mail handlers had been hired. 

Another staffing issue that arose from the restructuring involved a
management decision that placed some employees into supervisory
positions when they were not familiar with the work of the employees
they were supervising.  The Postal Service said it did this to avoid
relocating employees outside the Washington, D.C., metropolitan area. 
However, this action raised additional congressional concerns about
the adequacy of training for new supervisors. 

The Postal Service began making changes to its training program after
the restructuring and believes that its ability to train people
properly, quickly, and economically is being strengthened.  For
example, postal officials said that the supervisory training program
was being revised and a curriculum based on needs assessment was
being developed.  In commenting on a draft of this report, the Postal
Service said that new supervisors are getting the training they need,
and that the Service is continuing to hire more letter carriers and
mail handlers and to place them where they are most needed. 

Compounding staffing problems was the delay in expected benefits from
automation.  The Postal Service had expected that by 1995 virtually
all letter mail would be barcoded by either the Postal Service or the
mailer.  However, in April 1994, it announced that the barcoding goal
date had slipped to the end of 1997. 

Automation increases the efficiency of mail processing by decreasing
the volume that has to be sorted by relatively slower and more costly
mechanized or manual processing--potentially leading to higher EXFC
scores.  Mechanized sorting on letter sorting machines, on the other
hand, requires operators to memorize difficult sort schemes and key
in ZIP Code information.  This human intervention results in higher
potential for mishandling mail, causing delays.  With automated
processing, barcoded letters are sorted in high-speed barcode
sorters, often to the level of the street address, with limited human
intervention.  As automation becomes fully deployed, the Postal
Service expects most mail to be already sorted by the time it gets to
a carrier for delivery. 


   ORGANIZATIONAL CHANGE WEAKENED
   MANAGEMENT CONTROL
------------------------------------------------------- Appendix III:3

Shortly after taking office in 1992, Postmaster General Runyon began
a top-down restructuring of the Postal Service.  This was part of a
broad strategy to make the Service more competitive, accountable, and
credible.  One key component of the restructuring was the separation
of mail processing and mail delivery at all levels of the
organization below the Executive Vice President/Chief Operating
Officer of the Postal Service.  This action resulted in splitting
accountability for processes critical to mail delivery service. 

The value of separating responsibility for the mail processing
function (which takes place primarily at processing and distribution
centers) from the mail delivery function (which takes place primarily
at local post offices) has been controversial.  The separation left
no single manager with the responsibility and authority to coordinate
and integrate the mail processing and delivery functions in the
Washington, D.C., metropolitan area.  Each manager's primary focus
became the fulfillment of his or her own individual responsibilities. 
Working with managers of other functions became secondary. 
Consequently, critical decisions affecting both mail processing and
customer services in the Washington, D.C., area were not being made
by one manager at the operating level of the organization. 

For example, when we visited one post office in Northern Virginia,
local postal officials complained that too much unsorted and
misrouted mail was routinely sent to local post offices in order to
keep the Merrifield P&DC from having a backlog of unprocessed mail. 
On the day of our visit, these officials showed us a container of
misrouted mail from Merrifield that included not only overnight
First-Class Mail but also Priority Mail.  The Postmaster noted that
by the time this mail could be sent back to Merrifield to be
correctly sorted, it would be at least 1 day late.  Since there was
no one manager with jurisdiction over processing and delivery
functions in the Washington, D.C., metropolitan area, resolution of
conflicts between the two functions could be accomplished only
through the direct involvement of the area vice president, who had
responsibility for six states and Washington, D.C. 

The Inspection Service also identified excessive misrouted mail as a
significant problem in the Washington, D.C., area in its May 1994
report on mail conditions in the Mid-Atlantic Area.  In a December
1994 Inspection Service report, it also cited the split in
responsibilities between processing and delivery as a significant
problem in the Washington, D.C., area.  The report cited the absence
of teamwork and cohesiveness among managers.  The Inspection Service
said that there needs to be a "glue" to hold the managers of the
processing and delivery functions together in the Washington, D.C.,
metropolitan area.  Additionally, representatives from the National
Association of Letter Carriers and the American Postal Workers Union
told us that the split in responsibilities between processing and
delivery was a significant contributing factor to poor mail delivery
service in the Washington, D.C., metropolitan area. 

In June 1994, the Postmaster General changed the management structure
to increase the levels of teamwork and accountability in the Postal
Service.  He took this action in response to feedback from Members of
Congress, postal customers, and employees regarding the separation of
the customer service function and the processing and distribution
function that followed the 1992 restructuring.  The Postmaster
General combined the responsibility for customer service and mail
processing and distribution at a lower level in the
organization--from the Chief Operating Officer/Executive Vice
President to the area office level.  Instead of each of the 10 areas
having a manager for customer service and another for mail processing
and distribution, one overall manager with the rank of Vice President
was put in charge of both customer service and mail processing and
distribution. 

On January 10, 1995, the Postal Service made an additional change
designed to push accountability farther down in the organization.  On
that date, postal officials announced plans for establishing a
position under the Mid-Atlantic Area Vice President that would
oversee all processing and delivery functions in the
Washington/Baltimore area. 


SEVERAL MAIL HANDLING PROCESS
PROBLEMS CONTRIBUTED TO POOR
DELIVERY SERVICE
========================================================== Appendix IV

Several mail handling process problems contributed to the poor
delivery service in the Washington, D.C., metropolitan area.  These
problems included (1) the unnecessary duplicative handling of much
mail addressed to Northern Virginia, (2) the difficulty of meeting
delivery standards in some outlying areas, (3) the arrival of mail
too late for processing and delivery the next day, (4) the lack of a
system for routinely pinpointing the causes of delays in specific
pieces or batches of mail, and (5) the failure to follow established
procedures. 


   DUPLICATIVE HANDLING OF MAIL IN
   NORTHERN VIRGINIA
-------------------------------------------------------- Appendix IV:1

Mail addressed to two of the seven ZIP Code service areas in Northern
Virginia is often processed by both the Merrifield and Dulles
processing and distribution centers and is sometimes delayed by the
unnecessary additional processing.  This duplicative handling occurs
because the Merrifield and Dulles centers are jointly responsible for
processing mail addressed to the 220 and 221 ZIP Code service
areas.\18 This is partly a result of the way ZIP Codes were first
assigned within the 220 and 221 delivery service areas. 

In 1963, when the ZIP Code service areas were first established, the
Dulles facility did not exist; therefore, Merrifield was responsible
for all of 220 and 221.  At that time, postal officials at Merrifield
assigned Zip Codes using an alphabetic listing of all post offices in
these two service areas.  Because the assignments were made
alphabetically, there was no clear geographic distinction between the
220 and 221 service areas.  Subsequently, in 1992, when the Dulles
facility became operational, there was no good way of isolating
either the 220 or 221 service area for processing at Dulles. 
Therefore, both facilities assumed joint responsibility for
processing mail addressed to 220 and 221.  In 1991, however, a plan
was approved at the headquarters staff level to restructure the ZIP
Codes in these two service areas, but top management did not approve
that plan because of concerns over reactions from postal customers
about ZIP Code changes. 

Depending on the originating point and predetermined routing
schedules, mail addressed to 220 or 221 is to go to either the
Merrifield or Dulles centers for processing.  The receiving center is
to sort the mail to identify the mail that is to be delivered within
its service area and then dispatch the remaining mail to the other
center for further processing.  Postal officials said this procedure
results in excessive transportation between the two facilities and
duplicative sorting, which can also translate into delayed mail. 
Postal officials were unable to say precisely how much mail was
subjected to this duplicative processing but said it involved
substantial quantities. 

As a partial solution to the problem of duplicative mail handling in
the Northern Virginia area, the Postal Service has begun asking the
primary feeders of overnight mail to Northern Virginia to sort that
mail to a 5-digit level and transport it to the appropriate center in
Northern Virginia for further processing.  The Postal Service expects
this change to reduce the duplicative handling of mail between the
two centers, but it places more processing work on the other
facilities.  The Postal Service, in commenting on a draft of this
report, said that it will be installing a Remote Bar Coding System
site at the Dulles P&DC that it said will virtually eliminate the
need for duplicative handling of mail for some Northern Virginia ZIP
Codes. 


--------------------
\18 There are five other ZIP Code service areas assigned to Northern
Virginia--201, 222, 223, 226, and 227.  This mail is not subjected to
duplicative processing because the Merrifield center is exclusively
responsible for all mail addressed to the 222 and 223 ZIP code
service areas, and the Dulles center is exclusively responsible for
all mail addressed to 201, 226, and 227. 


   DELIVERY STANDARDS ARE
   DIFFICULT TO MEET IN CERTAIN
   OUTLYING AREAS
-------------------------------------------------------- Appendix IV:2

Plant managers at the Southern Maryland and Northern Virginia P&DCs
believe that consistent overnight delivery is difficult to achieve in
certain outlying areas.  They believe an extensive 1990 effort to
revise delivery standards and establish more realistic overnight
delivery service areas did not go far enough.  The plant manager at
the Southern Maryland P&DC, in particular, believes that he has an
excessively large overnight delivery service area, which he believes
has an adverse impact on his EXFC scores. 

In 1990, in an effort to provide better mail delivery service by
improving the Postal Service's ability to consistently deliver mail
within the standards, the Postal Service changed 6,389 (44 percent)
of its 14,578 overnight delivery areas nationwide to 2-day service
areas.  Although this change relaxed the delivery standards for some
areas, standards for other areas were unchanged. 

Two areas in Southern Maryland that were cited by the plant manager
at the Southern Maryland P&DC as examples of outlying locations where
overnight deliveries were not relaxed and are, at best, challenging
are Leonardtown and California, Maryland.  Mail from both of these
locations is processed at the Southern Maryland processing and
distribution center.  The plant manager at Southern Maryland said
mail from Leonardtown and California often does not arrive at the
Southern Maryland center for processing until 10:00 or 11:00 p.m.  He
said the post offices were unable to get the mail to him earlier in
the day because the carriers were often making deliveries and picking
up mail until late in the evening.  He said that because of the time
required to process the mail through the facility, it is difficult to
get the mail back out to Leonardtown and California in time for
delivery the next day. 

Partly to address the delivery problem to outlying areas, the Postal
Service is planning to process mail from Leonardtown and California,
in addition to other Southern Maryland areas, at a facility in
Waldorf (Charles County), Maryland, which is closer to Leonardtown
and California.  The Postal Service believes that by decentralizing
processing it will be better able to serve the Southern Maryland
mailing public and provide more reliable, consistent service. 

In addition, to improve mail flow, the Postal Service is installing
more "local only" collection boxes in high-traffic locations
throughout the Washington, D.C., area.  The ZIP Codes covered by that
service are to be clearly displayed on the collection boxes. 
Customers using these boxes should receive overnight service because
that mail will not leave the local area for processing. 


   MAIL ARRIVING TOO LATE FOR
   TIMELY PROCESSING
-------------------------------------------------------- Appendix IV:3

Mail also arrived late at area P&DCs for reasons other than the size
of the service area.  Each P&DC has established an operating plan
specifying critical entry times for receipt of mail in order to meet
established clearance and dispatch times at the P&DC.  However, area
plant managers told us that large quantities of mail, from mailers
and other postal facilities, frequently arrived past the critical
entry times.  This compressed the amount of time that P&DCs had
available for processing the mail.  The area managers said they have
few options other than to accept the mail and then rush to meet their
clearance and dispatch times.  They feel that to do otherwise would
upset the delicate balance between providing customer service and
meeting established time schedules. 

The Inspection Service identified mail arriving late at P&DC centers
as one of the major contributors to delayed mail.  The Inspection
Service also reported that other delays occurred because bulk
business mail was sometimes worked out of sequence--i.e., the latest
arriving mail was being worked first instead of last.  Postal
officials at the Southern Maryland P&DC said local mailers routinely
deposited large amounts of bulk business mail on their docks late in
the day and expected deliveries to be made the next day. 

To better plan for and manage its workload, Postal Service officials
said customer service representatives were more actively working with
major mailers in the area to get them to mail earlier in the day and
also notify the Postal Service ahead of time when large mailings are
expected to arrive.  Additionally, some of the mail processing that
was being done at P&DCs is now being shifted to local post offices. 
Postal Service officials believe this will expedite mail distribution
to carriers and improve service to customers. 


   LACK OF AN EFFECTIVE SYSTEM FOR
   ANALYZING DELAYED MAIL TO
   IDENTIFY CAUSES
-------------------------------------------------------- Appendix IV:4

As of December 31, 1994, the Postal Service did not have a system
that could be used to examine delayed mail and pinpoint where, in the
processing and delivery stream, the mail fell behind schedule. 
Without being able to pinpoint problems in the mailstream, the Postal
Service is forced to react to the effects of delivery problems on
customer service instead of taking timely steps to avoid or reduce
late deliveries. 

The Postal Service has nearly 40,000 post offices, stations, and
branches that collect and deliver over 570 million pieces of mail
daily.  Between collection and delivery, mail is transported, sorted,
and delivered by over 700,000 employees working in or out of over 349
mail processing and distribution facilities.  A First-Class letter
traveling from coast to coast passes through a myriad of mail
processing, transportation, and delivery operations. 

Mail typically moves between processing steps in a distribution
facility, or among facilities, in batches carried in large mail
containers.  The Postal Service has systems that use barcoding or
other forms of automated identification of containers to assist in
the control and movement of containers.  However, these systems are
not designed to provide operational data on a comprehensive basis
that allow the Postal Service to track each mail container through
the entire processing and distribution cycle.  Consequently, postal
management cannot track First-Class Mail that was delayed and gather
related data to promptly determine when, where, and why it fell
behind schedule. 

One floor supervisor at the Brentwood processing facility in
Washington, D.C., explained the implications of this weakness.  He
said that any postal employee can examine a container of mail at any
point in the processing and delivery cycle and determine whether that
mail is on schedule.  This is possible because each P&DC has an
operating plan establishing "windows" for receiving, processing, and
dispatching mail.  Therefore, a mail handler can examine the postmark
on a mailpiece, compare it to the mail processing timetable
(operating plan), and determine whether or not the mailpiece is
delayed.  However, if the mailpiece is delayed, the critical factors
that cannot be determined are when, where, and why the mailpiece fell
behind schedule.  In other words, there is no "history" of the
mailpiece (or container of mailpieces) that would pinpoint breakdowns
in the mailstream and allow the Service to take corrective actions to
prevent future slowdowns.  For example, at Southern Maryland, we
noticed mail waiting to be processed that should already have been
delivered.  The supervisor in charge was unable to tell us if that
mail was delayed before it arrived at Southern Maryland or became
delayed somewhere within the plant, nor could he tell us why it was
delayed. 

Without a diagnostic tool for tracking delayed mail to the source of
the problem, corrective actions can be made only to the extent that
breakdowns in the mailstream are significant enough to either become
conspicuous to postal managers--such as large volumes of mail being
consistently late from a particular facility--or cause EXFC or CSI
ratings to drop. 

Although the Postal Service has not yet developed a system that can
review the history of delayed mailpieces to identify points and
causes of delays, it has taken steps to try to identify systemwide
problems that could cause mail delays.  For example, Postal
Headquarters has set up a National Operations Management Center that
allows officials to monitor mail flow across the nation and respond
to performance problems and changing customer needs.  Management also
reports that it is identifying "pinch points," which slow mail in the
postal network, and rerouting mail when the need arises. 

Postal officials recognize the need for a capability to track delayed
mail.  They said that since most letters and flats are now barcoded,
a logical next step would be the handling of batches of mail under
some form of computer-assisted tracking and control system. 
According to Postal technicians, since all mail moves between
processing steps in a distribution center, or among centers, in
batches carried in some form of container, it is possible to identify
those containers and their contents with a machine-readable code that
would enable computer-based systems to monitor their movements. 
Accordingly, the Postal Service is developing a program for the
automated identification and tracking of single high-value mailpieces
or batches of mail in containers.  This program, known as the
Unit-Load Tracking Architecture (ULTRA), is still in an early
formative stage and may take years to develop and implement. 

Under the ULTRA system, unique codes would be applied to letters,
parcels, sacks, trays, and containers that would allow the Postal
Service to track the units through the postal system.  This
comprehensive system could allow definitive identification of the
points and causes of processing and delivery delays.  In commenting
on a draft of this report, the Postal Service said that it was also
looking into other diagnostic technologies as means of improving its
ability to identify underlying causes of delayed mail. 


   FAILURE TO FOLLOW ESTABLISHED
   PROCEDURES
-------------------------------------------------------- Appendix IV:5

Over the past few months, the Inspection Service reported many
instances where failure to follow established mail processing
procedures contributed to delays.  Many instances have been
identified where mail was not picked up from collection boxes;
various types of mail were commingled in the same container, causing
double handling and reduced cancelling efficiency; color codes
designating delivery dates were not used or were used improperly; and
inaccurate reports were prepared on mail conditions.  For example,
the Washington, D.C., P&DC was not placing color codes on a large
volume of its mail.  This led to mail being worked out of sequence
and sometimes delayed.  The Inspection Service also identified
improper color coding as a significant problem in the delivery units. 

The Inspection Service reported that significant progress has been
made in following established procedures for collecting, separating,
color coding, and properly reporting on mail conditions.  According
to Postal officials, these actions are being accomplished primarily
through increased training and reminders to employees of the need to
adhere to established procedures. 


   OTHER INITIATIVES TO ADDRESS
   PROCESS PROBLEMS
-------------------------------------------------------- Appendix IV:6

In December 1994, several service improvement teams were in place. 
These teams comprised both craft and management employees from a
variety of functions.  A major part of the teams' work is to examine
mail flow processes and identify other weaknesses that may be
contributing to late mail. 


LONG-TERM SOLUTIONS INVOLVE
ADDRESSING LABOR-MANAGEMENT
RELATIONS
=========================================================== Appendix V

Despite the potential benefits of operational changes, long-term
improvements in delivery service will require labor and management to
work together toward a common goal of continually improving customer
service.  Fundamental changes must occur in labor relations in order
to increase employee commitment and reduce the conflicts between
labor and management that currently exist.  This is particularly true
in the Washington, D.C., metropolitan area.  Workforce management
problems that were disruptive to mail handling operations have
occurred more frequently in the Washington, D.C., metropolitan area
than in most other parts of the country. 

Improving employee commitment is one of the Postmaster General's
corporate goals.  In a recent study of labor relations, we found a
negative labor climate that did not foster employee commitment.  Our
report\19 disclosed that labor-management relations problems persist
on the factory floor of postal facilities.  A negative labor climate
can impair both productivity and product quality. 


--------------------
\19 U.S.  Postal Service:  Labor-Management Problems Persist on the
Workroom Floor (GAO/GGD-94-201 A/B, Sept.  1994). 


   WORKFORCE MANAGEMENT PROBLEMS
   WERE MORE SEVERE IN THE
   WASHINGTON, D.C., METROPOLITAN
   AREA
--------------------------------------------------------- Appendix V:1

A number of studies have documented that there is a relationship
between employees' attitudes and performance.  One of the most
prevalent workforce management problems in the Washington, D.C.,
metropolitan area was running mail handling operations without a full
complement of workers.  Often, employees were unexpectedly absent or
otherwise unavailable to do their normal work assignments. 
Unexpected absences often involved the use of sick leave.  Employees
can also be unavailable for their regular work if they have been
injured or are otherwise considered by their physician to be
medically incapable of performing normal duties.  Some managers said
that unusually high usage of sick leave and limited/light duty
indicated possible abuse.  Managers also said, and the EOS tends to
support, that excessive employee absences and unavailability for
regular duties are often brought about by substance abuse or poor
employee attitudes. 

Postal Service data showed that employees in the Washington, D.C.,
metropolitan area experienced greater than average use of sick leave
and a higher than normal use of limited duty and light duty work
assignments.  The EOS also suggested a greater than average level of
perceived substance abuse.  In addition, the EOS index suggested that
Washington, D.C., area employee attitudes about postal management
ranked among the lowest in the country. 


      SICK LEAVE USAGE
------------------------------------------------------- Appendix V:1.1

Figure V.1 shows that sick leave usage from 1992 through 1994 for the
Northern Virginia and Capital clusters was higher than the national
average.  The Northern Virginia sick leave usage rates, expressed as
a percentage of total workhours, were 3.27, 3.11, and 3.29 during the
period, while the Capital cluster rates were 3.56, 3.31, and 3.62,
respectively.  These usage rates were greater than the national
averages, which were 3.22, 3.01, and 3.13 for the period. 

   Figure V.1:  Sick Leave Usage
   Expressed as a Percent of Total
   Workhours for the Nation and
   for the Northern Virginia and
   Capital Clusters

   (See figure in printed
   edition.)

Source:  Postal Service data. 


      EMPLOYEES DOING LIMITED DUTY
      OR LIGHT DUTY WORK
------------------------------------------------------- Appendix V:1.2

As figure V.2 shows, limited/light duty hours as a percent of total
workhours were about twice the national average in the Capital
cluster and about one and one-quarter times the national average in
the Northern Virginia cluster. 

   Figure V.2:  Limited/Light Duty
   Hours Expressed as a Percent of
   Total Workhours for the Nation
   and for the Northern Virginia
   and Capital Clusters

   (See figure in printed
   edition.)

Source:  Postal Service data. 


      SUBSTANCE ABUSE
------------------------------------------------------- Appendix V:1.3

The EOS responses suggested that many employees believed there were
substance abuse problems (alcohol and drugs) in the Postal Service,
which could have caused attendance problems and poor employee
performance.\20 Locally, as shown in figure V.3, a higher than
average percentage of employees in the Southern Maryland; Washington,
D.C.; Merrifield, Virginia; and Suburban Maryland P&DCs believed
alcohol abuse was a problem where they work. 

   Figure V.3:  Percent of Mail
   Processing Employees Agreeing
   That Alcohol Abuse Among
   Employees Was a Problem Where
   They Work

   (See figure in printed
   edition.)

Source:  Postal Service, 1994 Employee Opinion Survey. 

Postal Service employees also perceived drug abuse as a problem in
the Washington, D.C., area, as shown in figure V.4.  None of the
local P&DCs reported lower than average perceptions of drug abuse. 

   Figure V.4:  Percent of Mail
   Processing Employees Agreeing
   That Drug Abuse Among Employees
   Was a Problem Where They Work

   (See figure in printed
   edition.)

Source:  Postal Service, 1994 Employee Opinion Survey. 

Employees in delivery units generally perceived that substance abuse
was much less of a problem than did employees in the P&DCs. 


--------------------
\20 For our analysis, we used EOS responses from P&DCs.  The
responses did not include bulk mail centers or air mail facilities. 


      EMPLOYEE ATTITUDES
------------------------------------------------------- Appendix V:1.4

Employee attitudes can be a factor in the level of employee
commitment.  One measure of employee attitudes is the EOS Index--the
average favorable response on 20 employee opinion survey questions. 
These questions deal with how managers and supervisors treat
employees; respond to their problems, complaints, and ideas; and deal
with poor performance and recognize good performance.  As table V.1
shows, the postal workforce in the Washington, D.C., metropolitan
area gave local management relatively low marks, placing most of the
units in the area in the bottom 25 percent of all units nationwide. 



                               Table V.1
                
                Capital and Northern Virginia Clusters'
                   1994 EOS Index Scores and Quartile
                                Rankings

                                                       EOS    Quartile
Cluster and unit                                   Index\a   ranking\b
----------------------------------------------  ----------  ----------
Capital Cluster
----------------------------------------------------------------------
Customer Service (post offices)                         36          Q4
Washington, D.C., P&DC                                  35          Q3
Southern Maryland P&DC                                  33          Q4
Suburban Maryland P&DC                                  33          Q4


Northern Virginia Cluster
----------------------------------------------------------------------
Customer Service (post offices)                         39          Q4
Merrifield, Virginia, P&DC                              32          Q4
Dulles, Virginia, P&DC                                  33          Q4
----------------------------------------------------------------------
\a Nationwide, the EOS Index scores for customer service units (post
offices) ranged from a low of 36 percent to a high of 57 percent. 
Processing and distribution center (P&DC) EOS Index scores ranged
from a low of 23 percent to a high of 59 percent, nationwide. 

\b A Q4 ranking places the unit in the bottom 25 percent of all
similar facility-type units nationwide.  A Q3 ranking places the unit
in the bottom half of all similar facility-type units nationwide. 

The Washington, D.C., area was not unlike other large, urban areas
with regard to the relationship between low employee morale and low
service scores.  As table V.2 shows, the EOS Index scores for most
units in nine other large urban areas that we judgmentally selected
for comparison purposes ranked in the bottom half of all units
nationwide. 



                               Table V.2
                
                Other Large Urban Areas' 1994 EOS Index
                      Scores and Quartile Rankings

                                                       EOS    Quartile
City and facility type\a                           Index\b   ranking\b
----------------------------------------------  ----------  ----------
Atlanta CS                                              42          Q3
Atlanta P&DC                                            37          Q3
Boston CS                                               38          Q4
Boston P&DC                                             37          Q3
Chicago CS                                              37          Q4
Chicago P&DC                                            40          Q2
Dallas CS                                               41          Q4
Dallas P&DC                                             34          Q4
North Texas P&DC                                        34          Q4
Los Angeles CS                                          44          Q3
Los Angeles P&DC                                        37          Q3
Miami (South Florida) CS                                36          Q4
Miami P&DC                                              37          Q3
Ft. Lauderdale P&DC                                     33          Q4
South Florida P&DC                                      34          Q4
New York CS                                             41          Q4
NYC General Post Office P&DC                            42          Q2
NYC Morgan P&DC                                         40          Q3
NYC Church St P&DC                                      39          Q3
NYC Bronx P&DC                                          36          Q3
Philadelphia CS                                         37          Q4
Philadelphia P&DC                                       37          Q3
San Francisco CS                                        41          Q4
San Francisco P&DC                                      40          Q2
----------------------------------------------------------------------
\a CS (customer service) is post offices within the district.  P&DC
is a processing and distribution center located in the customer
service cluster. 

\b See footnotes in table V.1. 

Source:  Postal Service data. 

Like the EOS Index scores, the EXFC and CSI scores for these nine big
cities also were relatively low compared to scores in other areas of
the country.  Figures V.5 through V.7 show that EXFC scores for most
of the nine cities have usually fallen below the national average. 

   Figure V.5:  History of EXFC
   Scores for New York,
   Philadelphia, and Boston
   Compared to the National
   Average

   (See figure in printed
   edition.)

Source:  Postal Service data. 

   Figure V.6:  History of EXFC
   Scores for Miami, Atlanta, and
   Dallas Compared to the National
   Average


   (See figure in printed
   edition.)

Source:  Postal Service data. 

   Figure V.7:  History of EXFC
   Scores for Chicago, Los
   Angeles, and San Francisco
   Compared to the National
   Average


   (See figure in printed
   edition.)

Source:  Postal Service data. 

Figures V.8 through V.10 show that CSI scores for eight of the nine
cities have also usually fallen below the national average. 

   Figure V.8:  History of CSI
   Scores for New York,
   Philadelphia, and Boston
   Compared to the National
   Average

   (See figure in printed
   edition.)

Source:  Postal Service data. 

   Figure V.9:  History of CSI
   Scores for Miami, Atlanta, and
   Dallas Compared to the National
   Average


   (See figure in printed
   edition.)

Source:  Postal Service data. 

   Figure V.10:  History of CSI
   Scores for Chicago, Los
   Angeles, and San Francisco
   Compared to the National
   Average


   (See figure in printed
   edition.)

Source:  Postal Service data. 

The relationship between employee attitudes and service performance
is an important factor in improving labor-management relations. 


   IMPROVING LABOR-MANAGEMENT
   RELATIONS IS A LONG-TERM
   PROPOSITION
--------------------------------------------------------- Appendix V:2

We recently reported, and the Postal Service has acknowledged, that
improving labor-management relations is a long-term proposition.  In
our recently issued report on labor-management relations,\21 we
recommended that the Postal Service, the unions, and management
associations develop a long-term agreement (at least 10 years) for
changing the workroom climate for both processing and delivery
functions. 

Postal Service efforts to address problems in Chicago illustrate that
breakthrough improvements require a long-term effort.  Responding to
our 1990 letter\22 highlighting our observations on the need for mail
delivery service improvements in Chicago, the Postmaster General
developed a plan for improving service.  Four years later, service in
Chicago remained poor. 

Chicago has a long history of low EXFC scores, and in early 1994
attention was again focused on its mail delivery service problems. 
About 40,000 pieces of undelivered mail were found in a letter
carrier's truck parked outside a post office in Chicago.  The oldest
envelopes bore postmarks from December 1993.  A month later the
Chicago police discovered more than 100 pounds of burning mail
beneath a viaduct on the Chicago South Side.  That same day, another
20,000 pieces of undelivered mail--some up to 15 years old--were
found behind the home of a retired carrier in southwest Chicago. 
When CSI quantified the level of customer dissatisfaction, Chicago
ranked last 15 of the 16 times the survey has been conducted. 

The Postmaster General reacted by creating a 27-member Chicago
Improvement Task Force to identify and correct service problems.  The
Postal Service reported a number of corrective actions instituted by
the task force that were designed to improve mail delivery service. 

Similar to the situation in Washington, D.C., the task force found
operations problems as well as problems with the attitudes of
employees.  Despite the task force's corrective actions, Chicago has
not made breakthrough improvement.  Although there has been greater
on-time performance, reduced delayed mail, fewer complaints, and less
waiting time in line, Chicago's EXFC performance for quarter 4, 1994,
remained 6 points below its score in the same quarter in the prior
year and 12 points below the national average.  Customer satisfaction
also remained poor at 51 percent. 

Operations improvements are vital, but they will not solve all
delivery service problems.  Short-term gains through operational
improvements may eventually succumb to the obstacle to permanent
improvement--namely, a negative labor climate.  Long-term
improvements require substantive improvements in labor-management
relations. 

Since taking office in July 1992, the Postmaster General has been
working to forge a labor-management partnership to change the culture
in the Postal Service.  His goal is to shift the Postal Service
culture from one that is "operation driven, cost driven,
authoritarian, and risk averse" to one that is "success-oriented,
people oriented, and customer driven." We previously reported\23 that
the Postmaster General developed a labor-management partnership
through the National Leadership Team structure, held regular
leadership meetings that included all Postal Service officers and the
national presidents of the unions and management associations, and
changed the management reward systems to encourage teamwork and
organizational success. 

However, as we also previously reported, there is no overall
agreement among the unions and management for change at the field
operations level.  They have been unable to come to terms on a clear
framework or long-term strategy for ensuring that first-line
supervisors and employees at processing plants and post offices buy
into renewed organizational values and principles. 

In his November 30, 1994, statement before the Subcommittee on
Federal Service, Post Office, and Civil Service, Senate Committee on
Governmental Affairs, the Postmaster General testified that the
Postal Service supports our September 1994 report recommendations
calling for the Service, unions, and management associations to
develop a long-term agreement on objectives and approaches for
demonstrating improvements in the work climate of both processing and
delivery operations.  At the hearing, he proposed that the Leadership
Team form a task force made up of leaders of the unions and
management associations and key postal vice presidents.  Mr.  Runyon
said the task force should have a 120-day agenda "to explore [GAO's]
recommendations, set up pilot projects, and move forward now to
accelerate change in our corporate attitudes and culture." While his
labor-management summit proposal received the support of the rural
carriers and the three management associations, the leaders of the
three largest postal unions have not yet agreed to the summit.  They
said they are waiting until the current round of contract
negotiations is completed before making a decision on the summit. 



(See figure in printed edition.)Appendix VI

--------------------
\21 GAO/GGD-94-201 A/B. 

\22 Letter to Postmaster General Anthony M.  Frank, February 26,
1990. 

\23 GAO/GGD-94-201 A/B. 


COMMENTS FROM THE U.S.  POSTAL
SERVICE
=========================================================== Appendix V



(See figure in printed edition.)


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================= Appendix VII

GENERAL GOVERNMENT DIVISION,
WASHINGTON, D.C. 

Michael E.  Motley, Associate Director
James T.  Campbell, Assistant Director
Lawrence R.  Keller, Evaluator-in-Charge
Roger L.  Lively, Senior Evaluator
Charles F.  Wicker, Senior Evaluator
Lillie J.  Collins, Evaluator
Kenneth E.  John, Senior Social Science Analyst

*** End of document. ***