U.S. Postal Service: Stronger Mail Acceptance Controls Could Help Prevent
Revenue Losses (Letter Report, 06/25/96, GAO/GGD-96-126).

Pursuant to a congressional request, GAO reviewed the Postal Service's
(USPS) controls over postage paid on presorted and barcoded mail,
focusing on whether USPS controls ensure that mailer-claimed discounts
are earned.

GAO found that during fiscal year 1994: (1) 40 percent of required bulk
mail verifications were not performed and postal supervisors did less
than 50 percent of required follow-up verifications; (2) rejected
mailings were resubmitted and accepted into the mailstream without
proper corrections or postage; (3) mail acceptance clerks were not given
adequate tools to determine whether increasing volumes of mailer-applied
barcodes met USPS standards; and (4) postal management was unable to
make informed decisions concerning the adequacy of bulk mail acceptance
controls or determine the amount of revenue lost through improperly
prepared mailings. GAO also found that: (1) USPS needs to determine the
management strategy and financial investment necessary to minimize
revenue loss; (2) the random method of selecting bulk business mailings
for in-depth verification may not result in the best use of available
staff; (3) USPS could better target its verification efforts based on
risk by considering such factors as mailer histories and the postage
value of mailings; (4) postal managers are developing a bulk business
mail acceptance system, updating acceptance handbooks, acquiring
barcoding verification equipment, and requesting field units to submit
verification reports; and (5) USPS is also exploring a new risk-based
approach for in-depth verification, improving revenue controls, and
planning to install its new bulk mail system by 1997.

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

 REPORTNUM:  GGD-96-126
     TITLE:  U.S. Postal Service: Stronger Mail Acceptance Controls 
             Could Help Prevent Revenue Losses
      DATE:  06/25/96
   SUBJECT:  Human resources utilization
             Postal rates
             Government sponsored enterprises
             Losses
             Internal controls
             Risk management
IDENTIFIER:  USPS Coding Accuracy Support System
             USPS National Bulk-Rate Mail Verification Program
             USPS National Bulk Mail System
             
******************************************************************
** 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 the Chairman, Subcommittee on the Postal Service, Committee
on Government Reform and Oversight
House of Representatives

June 1996

U.S.  POSTAL SERVICE - STRONGER
MAIL ACCEPTANCE CONTROLS COULD
HELP PREVENT REVENUE LOSSES

GAO/GGD-96-126

Mail Acceptance Controls

(240184)


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

  ABE - Automated Barcode Evaluator
  BARQUEST - Barcoding, Addressing, Readability Quality Utilizing
     Electronic Systems Technology
  BMEU - Business Mail Entry Unit
  CASS - Coding Accuracy Support System
  DIF - Discriminate Function Analysis
  DMU - Detached Mail Unit
  IRS - Internal Revenue Service
  QPV - Quality Presort Verification

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


B-262118

June 25, 1996

The Honorable John M.  McHugh
Chairman, Subcommittee on the Postal Service
Committee on Government Reform and Oversight
House of Representatives

Dear Mr.  Chairman: 

This report responds to your request that we review the Postal
Service's controls over postage paid on presorted/barcoded mailings
submitted by business customers.  As agreed with the Subcommittee, we
evaluated the current system of controls for such mailings to assess
whether those controls reasonably ensure that mailer-claimed
discounts are granted only when earned. 

Mailers compute the postage they owe on each bulk business mailing. 
Sometimes, the mailing does not qualify for the postage discount
claimed.  To guard against accepting improperly prepared mailings,
Postal Service bulk mail acceptance clerks are supposed to verify the
accuracy of bulk business mailings.  When mail with preparation
errors is accepted into the mailstream at a presort and/or barcode
discount rate, the Postal Service incurs a cost.  This cost is in the
form of higher processing expenses because the Postal Service must
then do work it has already, in effect, paid mailers to do through
discounted postage rates.  In fiscal year 1994, according to Postal
Service data, almost one-half of the Service's total mail revenue of
$47.7 billion was from bulk business mail. 

The Postal Service recognizes the need to have reasonable assurance
that its bulk business mail operation adequately protects the Service
from postage revenue losses.  Toward that end, in 1989 and 1990, the
Postal Service established a control system designed to give it such
assurance.  That assurance is dependent on the control system
operating as designed, including the production of certain
information that management determined was needed for monitoring and
evaluating the effectiveness of the control system and making changes
as appropriate. 

As described in the Postal Service's manuals, the control system
requires that

  -- presort and barcode mail verifications, including supervisory
     reviews, be done and documented;

  -- mailings resubmitted following a failed verification be
     reverified to ensure that the errors are corrected; and

  -- the results of verifications, including supervisory reviews, be
     regularly reported to specific management levels, including
     Postal Service headquarters. 


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

In 1989 and 1990, to address what the Postal Service said was a "seat
of the pants" approach to bulk mail acceptance, postal management put
in place a system of controls designed to help ensure that all
revenue from bulk business mail was being received.  Although the
control system reported catching errors totaling $168 million in
fiscal year 1994, the system did not provide the Service with
reasonable assurance that all significant amounts of revenue due from
bulk business mailings were correctly identified and received because
of the weaknesses discussed below. 

First, during fiscal year 1994, as much as 40 percent of required
bulk mail verifications were not performed.  Additionally,
supervisors did less than 50 percent of required follow-up
verifications to determine the accuracy of clerks' work and identify
training needs.  Second, rejected mailings could be resubmitted and
accepted into the mailstream without errors having been corrected or
additional postage paid.  Third, acceptance clerks were not given the
tools they needed to determine, in an efficient and objective manner,
whether the increasing volumes of mailer-applied barcodes met Postal
Service technical standards designed to ensure that the barcodes
could be "read" by the Service's automated equipment.  Fourth, postal
management was not provided, nor did it seek, information necessary
to make informed decisions concerning the adequacy of bulk mail
acceptance controls, staffing, and training at acceptance units. 

The Postal Service's bulk business mail control system also lacked a
mechanism for identifying how much revenue was being lost, overall,
by the Service accepting improperly prepared mailings.  Unless the
Service has an accurate indication of the magnitude of the risk, it
will be hindered in (1) making informed decisions about the
management attention and dollar investment necessary to manage and
minimize revenue losses and (2) assessing the effectiveness of
initiatives to improve revenue protection controls. 

The Postal Service's primary method of selecting bulk business
mailings for in-depth verification results in more mailings selected
for verification than the Service can realistically do.  Further, its
random selection of mailings, without regard to the risk of potential
revenue losses, may not result in the best use of available
acceptance staff resources.  Other federal agencies that collect
revenues and have large verification operations, such as the Internal
Revenue Service (IRS) and the Customs Service, have adopted more
selective verification plans.  We believe these plans could offer
advantages over the Service's random selection because the plans
target those cases for audit/inspection that historically have had a
higher potential for error and monetary pay-off. 

Although the Postal Inspection Service has long considered bulk
business mail acceptance to be a high-risk activity, in the past,
postal management has not devoted much attention to the adequacy of
acceptance controls.  Recently, however, managers have begun
developing a new bulk business mail acceptance system and are taking
steps that have the potential to improve bulk mail acceptance
controls.  The more significant steps include updating bulk mail
acceptance handbooks, acquiring equipment to help acceptance clerks
verify barcodes, and requesting field units to submit required
reports on bulk mail verifications.  In addition, the Service is
exploring a risk-based approach for selecting mailings for in-depth
verification and has created a new unit to help improve controls over
revenue.  The Service expects to have almost all of its new bulk mail
acceptance system in place by later this year or early calendar year
1997. 

In response to the control issues raised in this report, we have
included a number of specific recommendations.  We believe these
recommendations, incorporated into the Postal Service's bulk mail
acceptance system, would help the Service better ensure that
significant amounts of revenue are not being lost. 


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

According to Postal Service figures, of the 177 billion pieces of
mail it processed in 1994, over 118 billion pieces, or 67 percent,
were categorized as bulk business mail.  In fiscal year 1994, the
Service recorded revenue from bulk business mail of $23.1
billion--48.4 percent of its total mail revenue. 

The Postal Service began offering postage discounts to mailers who
presorted their mail in 1976, and in 1988 it began offering discounts
for barcoding.\1 Presort and barcode discounts are to compensate
mailers for performing work that otherwise would have to be done by
the Postal Service.  In fiscal year 1994, about 34 percent of all
First-Class mail and 92 percent of all third-class mail was
discounted.  According to Postal Service studies, the value of these
discounts, during that year, totaled about $8 billion. 

Most bulk business mail is entered at Business Mail Entry Units
(BMEU) and Detached Mail Units (DMU), located throughout the Postal
Service's 85 districts.  DMUs are postal acceptance units located at
mailers' mail preparation facilities.  BMEUs are often located in or
adjacent to mail processing plants, which are postal facilities that
process mail for distribution to both local and national
destinations.  Bulk mail is also entered at many of the 40,000 post
offices located throughout the country. 

The Postal Service's mail acceptance clerks are the gatekeepers for
accepting bulk business mail into the mailstream.  Their job is to
ensure, before mail enters the Postal Service's processing and
distribution facilities, that mailers have prepared their mail in
accordance with postal requirements and that discounts given for
presorting and barcoding have, in fact, been earned.  This is a
difficult task given the time constraints and the wide variation in
the way bulk business mail can be prepared and still meet Postal
Service standards.  If mail barcoded by a mailer is accepted by
clerks but later fails to run on postal barcode sorters, the Postal
Service incurs additional costs.  This is because the Postal Service
must rework the mail at its own expense even though it gave the
mailer the barcoded rate to perform that work. 

Bulk mail acceptance clerks are to perform cursory verifications on
all mailings and in-depth verifications on randomly selected
mailings.  For every in-depth verification completed, mail acceptance
clerks are required to prepare a written verification report (Form
2866).  Postal facilities that receive 100 or more bulk mailings
during a 4-week accounting period are to prepare a consolidated bulk
mail acceptance report (Form 2867) documenting the results of their
in-depth verifications.  Summary reports of Forms 2867 are to be used
by postal managers at various times to monitor, among other things,
mail volume and revenue generated through the bulk mail acceptance
system.  Mail acceptance supervisors are to regularly verify the work
of the clerks and report the results to postal management. 

At Postal Service headquarters, management responsibility for the
bulk business mail program resides with the Vice President of
Marketing Systems, who reports to the Chief Marketing Officer and
Senior Vice President.  Area Vice Presidents and district managers
are responsible for ensuring that bulk mail acceptance activities
conform to prescribed standards within their geographic spans of
control. 

Appendix I contains additional background information on the
Service's bulk mail acceptance system. 


--------------------
\1 The Postal Service has postal business centers throughout its 85
districts that help mailers correctly presort and barcode their mail
to qualify for discounts. 


   OBJECTIVE, SCOPE, AND
   METHODOLOGY
------------------------------------------------------------ Letter :3

Our objective in this report was to determine whether the current
system of controls for accepting bulk business mailings reasonably
assures the Postal Service that mailer-claimed discounts are granted
only when earned.  The scope of our review was limited primarily to
the Service's BMEUs and DMUs, which account for the majority of the
bulk mail accepted by the Service.  We did not review controls at
other acceptance units, such as post offices and branches. 

To evaluate bulk mail acceptance controls, we (1) obtained and
analyzed policies and procedures affecting bulk business mail
acceptance; (2) visited 7 district offices located in 6 of 10 Postal
Service area offices, and interviewed postal staff assigned to 17
business mail acceptance units in those districts; (3) collected and
analyzed bulk mail acceptance reports that were available from 77 of
85 district offices for fiscal year 1994; and (4) interviewed various
Postal Service managers and operations personnel at Postal Service
headquarters and selected field locations.  We selected field
locations judgmentally primarily on the basis of management reports
submitted by acceptance units. 

We also interviewed officials from eight commercial bulk business
mailers at the field locations visited.  Additionally, we interviewed
a Postal Service contractor who is studying the feasibility of
utilizing risk assessment as a means of targeting high-risk mailings,
and we interviewed and obtained written information from IRS and
Customs Service officials about verification methods employed by
their respective agencies.  We interviewed the Executive Director of
the National Association of Presort Mailers to obtain information on
the presort industry's views regarding the Service's bulk mail
acceptance system. 

We obtained and analyzed documentation on proposed and ongoing Postal
Service initiatives to improve bulk mail acceptance
practices--although we did not evaluate the effectiveness of those
initiatives because they are not yet fully implemented.  Finally, we
reviewed recent Postal Inspection Service audits on bulk mail
operations and discussed ongoing work with cognizant postal
inspectors. 

The work done for this report was part of our broader revenue
protection survey that began in November 1993.  In May 1994, as part
of our revenue protection work, we reported on postage meter fraud.\2
For the most part, our review of the Service's bulk mail acceptance
controls was done at Postal Service headquarters and selected field
locations\3 between February 1995 and February 1996. 

We did all of our work in accordance with generally accepted
government auditing standards.  We obtained written comments on a
draft of this report from the Postal Service.  Its comments are
discussed at the end of this letter and are reprinted as appendix
III.  The Postal Service also provided additional technical comments
on the draft, which were incorporated where appropriate. 


--------------------
\2 Postage Meters:  Risk of Significant Financial Loss but Controls
Are Being Strengthened (GAO/GGD-94-148, May 26, 1994). 

\3 We visited field locations in Connecticut, Kentucky, Minnesota,
New Jersey, Pennsylvania, and Tennessee.  We also did limited work at
postal facilities in Colorado, Virginia, and Washington, D.C. 


   MAIL ACCEPTANCE CONTROLS DO NOT
   REASONABLY PROTECT AGAINST
   SIGNIFICANT REVENUE LOSSES
------------------------------------------------------------ Letter :4

It is inevitable that some revenue losses will occur in a program of
this magnitude, and, as with any business enterprise, the risk of
revenue losses must be weighed against the cost of establishing
controls to prevent and detect such losses.  The Postal Service,
however, is hindered in its ability to make data-driven decisions
about the adequacy of bulk mail acceptance controls.  For example,
the Service does not know the full extent of losses resulting from
mailer preparation errors, and, furthermore, it has not sought to
develop a means for identifying such losses.  Rather, the Postal
Service operates under the premise that since the Inspection Service
and managers in charge of bulk mail acceptance have not reported
large dollar losses, then such losses must not have occurred. 

We did not attempt to estimate the extent to which revenue losses
have occurred.  However, we believe that sufficient evidence exists
for the Postal Service to be concerned that substantial revenue
losses may have occurred and gone undetected in the bulk business
mail program. 

In 1989 and 1990, to address what it acknowledged to be a "seat of
the pants" approach to bulk mail acceptance, the Postal Service
developed and implemented new management guidelines and verification
requirements designed to give it reasonable assurance that
significant amounts of bulk mail revenue were not being lost.  Those
guidelines contained specific procedures and approaches for bulk mail
acceptance and provided guidance to supervisors and managers for more
analytical and effective management of acceptance employees. 

Available documentation shows that during fiscal year 1994, the bulk
business mail control system identified mailer preparation errors
totaling $168 million.\4 However, the control system fell short of
providing the Postal Service with the assurance it needs that
significant amounts of revenue are not being lost in the bulk
business mail program, as discussed below. 


--------------------
\4 Documentation on verification results was available for 77 of the
Postal Service's 85 districts.  Some documentation, however, included
only a portion of a district's total activity.  Eight districts did
not compile reports on verification results. 


      BULK BUSINESS MAIL
      VERIFICATIONS NOT BEING DONE
      IN ACCORDANCE WITH POSTAL
      SERVICE REQUIREMENTS
---------------------------------------------------------- Letter :4.1

Available Service documentation and our interviews with Service
officials indicated that a large amount of bulk mail was accepted
without proper verification.  This occurred because clerks often
skipped required in-depth verifications of bulk mail.  Additionally,
supervisors frequently failed to do required follow-up verifications
of acceptance clerks' work.  The Postal Service's failure to ensure
that required verifications were done, and done properly, left it
vulnerable to revenue losses. 

Postal Service figures show that during fiscal year 1994, the Service
accepted\5 over 16.2 million bulk business mailings of various sizes,
classes, and levels of preparation nationwide.\6 Typically, over
50,000 mailings were accepted daily, and the mailings averaged about
6,900 mail pieces.  According to Postal Service requirements, all of
the mailings should have received a cursory review, and between 2.3
and 2.9 million should have received an in-depth verification.  The
Postal Service estimated that given the criteria for selecting
mailings for in-depth verification, each acceptance location should
have done in- depth verifications on 14 to 18 percent of the mailings
received.  However, available documentation shows that only about 1.7
million in-depth verifications were done--about 60 to 75 percent of
the required verifications.  The remaining verifications were either
not done or not documented.\7

Available documentation for fiscal year 1994 shows that about 30
percent (23 of 77) of the postal districts reporting the results of
their in-depth verifications did less than the estimated
minimum-required 14 percent.  Among the 77 districts, the percentage
of mailings verified in-depth ranged from less than 2 percent to more
than 30 percent. 

Because acceptance procedures are not implemented uniformly
throughout the United States, Postal Service managers and acceptance
employees, as well as individuals in the business mail industry, said
that some mailers "shop around" for the "best" acceptance unit.  The
Executive Director of the National Association of Presort Mailers
cited inconsistencies among acceptance units as a concern of the
mailing industry. 

At almost half of the locations we visited, officials said that heavy
workloads and unscheduled leave were frequently the reasons that
required in-depth verifications were not being performed.  They also
said that balancing the goals of doing required mail verifications
and improving customer service further complicated the situation. 
Some acceptance unit managers we spoke with said that the
verification function is secondary to the Postal Service's goal of
increasing the level of customer satisfaction.  One said that this
conflict makes it difficult to do all required verifications because
mailings that fail verification are more likely to miss dispatch
times and delivery schedules and are, therefore, likely to decrease
customer satisfaction. 

Another reason why some of the required verifications were not done
is that the Postal Service allows acceptance clerks to skip
verifications without higher level approval.\8 For example, certain
mailings are designated by computer program software as requiring an
in-depth verification.  However, clerks can override the system and
enter mail directly into the mailstream without performing the
required verification.  Two acceptance unit managers told us that
such overrides frequently occur but that they do not keep records on
the extent of this practice.  They said the overrides can generally
be attributed to time pressures to "keep the mail moving."

Test mailings initiated by headquarters program officials also raised
questions about the adequacy of the verifications.  To develop some
baseline information on the quality of bulk mail presented at entry
units, the Inspection Service, at the request of headquarters bulk
mail acceptance program officials, agreed to submit 36 test mailings
at selected bulk mail acceptance locations.  Each test mailing was to
be submitted as a first-time mailing and therefore required to
undergo an in-depth verification.  Each test mailing was to consist
of about 11,000 to 12,000 pieces of third-class mail--about 25
sacks--and each sack was to contain mail preparation errors that the
inspectors believed should have easily been caught by acceptance
clerks. 

The Inspection Service completed only three test mailings before the
project was discontinued at the direction of the headquarters bulk
mail acceptance program officials.  For the first test mailing,
acceptance clerks at that location did not identify any of the
errors.  Moreover, although the erroneously prepared test pieces were
presented as lower, third-class bulk rate mail, they were processed
as First-Class mail--giving them priority over other third-class bulk
business mailings.  Test results were not any better for the next two
mailings--the "planted" errors were not detected in the verification
process. 

Bulk mail acceptance program officials explained that they
discontinued the test mailings because they provided little useful
information for improving bulk mail acceptance controls.  They
believed audits of mailings deposited by mailers would provide better
data to assess the types of preparation errors that are slipping
through the acceptance process.  Accordingly, program officials
replaced the test mailings with audits of mailer-deposited mailings. 
These audits were led by bulk mail acceptance program officials.  In
February 1996, bulk mail acceptance program officials said that they
were still reviewing data from the audits.  They said about 930
mailings were audited at 8 locations in November and December 1995. 
The results of those audits were not available at the time of our
review. 


--------------------
\5 Acceptance functions, as of October 1995, were performed by 1,202
bulk mail acceptance clerks, 1,793 bulk mail acceptance technicians,
and 470 part-time flexibles (clerks or mail handlers).  According to
a postal official, there has not been much fluctuation in the size of
this workforce over the past several years.  Additionally, some bulk
mail was accepted at small post offices and branches that did not
have acceptance clerks or technicians. 

\6 A minimum of 500 pieces is required for a First-Class bulk
business mailing, and 200 pieces for a third-class mailing. 

\7 All postal units accepting 100 or more bulk mailings during a
4-week accounting period are required to report the results of their
verifications to district offices. 

\8 In May 1996, Postal Service officials stated they planned to
notify acceptance units that override use would be restricted to
supervisory personnel. 


      REQUIRED SUPERVISORY REVIEWS
      FREQUENTLY NOT DONE
---------------------------------------------------------- Letter :4.2

To help ensure that the required verifications are done, and done
properly, the Postal Service requires that supervisors do at least
four Quality Presort Verifications (QPV) every 4-week postal
accounting period.  The QPV entails a supervisor rechecking an
in-depth verification performed by a clerk to assess performance and
also identify training needs.  Analysis of Postal Service data
showed, however, that such verifications are frequently either not
done or not reported.  For example, 67 of 74\9 postal districts
reported doing fewer than the required number of
verifications--including 4 that reported doing none.  The 74
districts should have done at least 111,000 QPVs but reported doing
only 44,000--about 40 percent. 

The manager of one of the acceptance units we visited said QPVs are
not being done because of a lack of supervisory staff and inadequate
supervisory training on verification of mailing statements.  The
Service does not require supervisors directly responsible for BMEU
and DMU activities to have any training relating to verification
activities.  In contrast, the Service requires that BMEU and DMU
acceptance clerks receive 120 hours of classroom training.  Our
interviews at selected acceptance units showed that clerks had
generally received the required training. 


--------------------
\9 Data from 3 of the 77 reporting districts were insufficient for us
to make a determination as to whether or not the required number of
QPVs had been done. 


      FAILED MAILINGS CANNOT BE
      TRACKED TO ENSURE THEY ARE
      NOT RESUBMITTED AND ACCEPTED
      WITHOUT CORRECTION
---------------------------------------------------------- Letter :4.3

According to Service officials, under the current system of controls,
previously failed mailings can enter the mailstream at a later time
or at a different BMEU or some other Service acceptance unit without
the errors being corrected.  Acceptance clerks do not have a reliable
way of tracking failed mailings to ensure that when those mailings
are resubmitted for entry into the mailstream, they can be identified
and rechecked. 

The ability to identify and recheck previously failed mailings is
necessary for clerks to verify that errors have been corrected. 
However, following a failed verification, mailings can lose their
identity and be entered into the mailstream without the problems
being identified, corrected, or additional postage being paid.  To
help guard against this, some acceptance locations were keeping
informal records of failed mailings.  Several bulk mail acceptance
managers, however, believe that the effectiveness of such records,
while better than nothing, is limited because the records are
informal and not shared with other acceptance units. 

Officials from the bulk mail acceptance program office and Inspection
Service provided us the following examples, which demonstrate several
ways that failed mailings can be entered into the mailstream without
problems being corrected or without additional postage being paid by
the mailer. 

  -- A mailing that failed verification at one location can enter the
     mailstream at another location.  Mailers sometimes have permits
     to enter bulk mail at more than one location, which can work to
     their advantage since there is no exchange of information
     between locations concerning failed mailings.  Overall,
     according to the bulk mail acceptance managers and postal
     inspectors we spoke with, the chance of a failed mailing being
     subjected to an in-depth verification at a second location is
     heavily weighted in the mailer's favor. 

  -- A mailing that failed verification at one location during one
     shift can enter the mailstream at the same location during a
     different shift.  Informal records of failed mailings may help
     prevent some of this, but not all acceptance locations we
     visited kept informal records of failed mailings. 

  -- A failed mailing may be combined with another mailing, thus
     losing its original identity.  It could then enter the
     mailstream without further verification. 


      ACCEPTANCE CLERKS LACK THE
      TOOLS TO DO EFFECTIVE AND
      OBJECTIVE VERIFICATIONS
---------------------------------------------------------- Letter :4.4

Since 1988, the Postal Service has granted postage discounts for
mailer-barcoded mail.  However, it has been slow to provide the tools
necessary to ensure that when accepted, barcoded mail meets the
Service's standards for claimed discounts.  Generally, the Service's
approach to ensuring accurate, machine-readable barcodes has been to
work with bulk mailers to ensure that when the mail is prepared, it
meets the Service's standards and requirements.  Nevertheless,
acceptance clerks are responsible for verifying that barcoded mail
meets Postal Service standards. 

With the volume of mailer-barcoded mail increasing yearly, the Postal
Service recognized the need to try to ensure more standardization of
mailer-applied barcodes.  In the mid-1980s, the Postal Service
developed the Coding Accuracy Support System (CASS) as a quality
control measure that, among other things, is intended to help ensure
that mailer-applied barcodes (1) are produced using current address
information and (2) match the address printed on the mail piece.  To
encourage mailers to have their software CASS certified, in 1991 the
Postal Service began allowing barcode rates only on mailings produced
using CASS-certified software. 

While the purpose of CASS is to ensure that mailers apply barcodes
that reflect the right addresses, it does not ensure that the
barcodes meet the Postal Service's technical standards for height,
width, spacing, placement, and clarity and thus can be processed on
the Service's automated barcode sorters.  Bulk mail acceptance clerks
are to help ensure that mailer-applied barcodes meet the Postal
Service's technical standards and can be read by its sorters. 
However, because of the precision required of machine-readable
barcodes, acceptance clerks need special equipment, such as
electronic scanners that can read barcodes, so that they can
objectively verify the readability of barcodes.  Postal management
recognized the need for such equipment 5 years ago.  For example, in
a memo to regional managers in 1990, a senior Postal Service
headquarters management official acknowledged that the Postal Service
had a problem because it was accepting discounted, barcoded mail even
though it did
".  .  .  not have the mechanisms or capability in the Bulk Mail
Acceptance Units or Detached Mail Units to properly verify the
accuracy and readability of customer applied barcodes.  .  ."

Although the Postal Service has recognized the need for special
equipment to verify barcodes, at the BMEUs and DMUs we visited,
clerks and managers did not have such equipment.  Officials at many
of the BMEUs and DMUs we visited said they check barcode readability
by visual inspection, which they sometimes referred to as
"eyeballing." Many said they supplement visual inspections with such
equipment as eyepieces, templates, and gauges.  However, a cognizant
official at Postal Service headquarters told us that such procedures
are very time consuming. 

Available data suggest that significant losses may be occurring
because of unreadable barcodes.  Through fiscal year 1992, the Postal
Service systematically reported some data on the amount of barcoded
mail that could not be read by its automated barcode sorters.  The
last report produced, which covered fiscal year 1992, showed that 7.4
percent of barcoded mail sent to its sorters could not be read.\10 In
fiscal year 1992, the Service accepted 25.9 billion pieces of
First-Class and third-class mailer-barcoded letter mail.  If the
rejection percentage for fiscal year 1992 were applicable to the mail
pieces, the Service could have lost revenue ranging from $30.4 to
$74.1 million on lower rate First-Class and third-class barcoded mail
that could not be sorted on the Service's sorters--depending on the
method (mechanized or manual) used for processing the rejected mail
pieces.\11

During fiscal year 1994, the Service processed about 47.6 billion
pieces of First-Class and third-class letter mail with mailer-applied
barcodes, compared to 25.9 billion pieces just 2 years earlier--an
84-percent increase.  The volume of all classes of barcoded mail
processed by the Service had increased to about 70 billion by fiscal
year 1995 and is expected to increase to more than 100 billion
letters by fiscal year 1997 as the Postal Service offers greater
incentives for barcoded mail under its mail classification reform
initiative.\12


--------------------
\10 This percentage includes mail with Postal Service and
mailer-applied barcodes--the report did not separate the two
individually. 

\11 Under the current rate structure, as much as 15 percent of a
mailing may not be barcoded and still qualify for an automation
discount.  Under the classification reform initiative, scheduled to
become effective July 1996, all mail pieces in a mailing must be
barcoded to qualify for an automation discount. 

\12 On January 26, 1996, the Postal Rate Commission issued an opinion
and recommendation (Docket No.  MC95-1) on a Postal Service proposal
to reclassify certain postage rates, which would, among other things,
increase discounts for barcoded mail.  The Postal Service Board of
Governors accepted most of the Commission's recommendations, and the
new rates are to become effective in July 1996. 


      RELIABLE PROGRAM DATA NEEDED
      TO ASSESS ADEQUACY OF
      CONTROLS AND RELATED RISKS
      DO NOT EXIST
---------------------------------------------------------- Letter :4.5

Some of the key data needed by Postal Service management to assess
the adequacy of controls and related risks do not exist.  The current
acceptance system does not produce information on (1) the extent to
which improperly prepared mailings are entering the mail stream and
the related revenue losses associated with improperly prepared
mailings--including mailer-applied barcodes that do not meet the
Postal Service's standards; and (2) the amount of rework required for
the Postal Service to correct improperly prepared mailings that enter
the mailstream.  Postal managers told us they had no way of producing
historical estimates of mailer errors and related revenue losses or
the rework time associated with such errors. 

Additionally, our work showed that reports that were to be prepared
by bulk mail acceptance units and summarized for management were not
always prepared or were missing key data, such as verification
results.  Managers at Postal Service headquarters and two district
offices questioned the usefulness of the reports because of concerns
about the completeness and accuracy of the data they contain. 
Information required in verification and acceptance reports, if
properly gathered and used, could provide management at each level
some measure of the effectiveness of bulk mail acceptance controls. 

A key element of the control system put in place in 1990 was the
requirement for a revised Bulk Mail Acceptance Report (Form 2867),
which was to summarize the bulk mail acceptance and verification
activities of BMEUs and DMUs.  This report was designed to provide
management at local, regional, and Postal Service headquarters levels
with consolidated information that could be used to assess the
adequacy of controls over the bulk business mail acceptance system
and to monitor related risks.  For example, at the Postal Service
headquarters level, a "critical factors report" was to be prepared to
assess whether required verifications were being done, whether
staffing of acceptance units was adequate, and to provide other
necessary management information. 

During our review, management officials at several levels said that
the 1992 Postal Service reorganization significantly altered postal
employees' views about bulk mail acceptance.  Some district managers
said they did not use information from the reports for decisionmaking
purposes because the data had become unreliable.  An area office
official said that after the reorganization, the Postal Service
ceased to regard bulk mail acceptance reports as mandatory.  He
stated that Postal Service headquarters did not drop the reporting
requirements; rather, it never told the newly created district
offices where to send the reports.  Another area official said that
following the 1992 restructuring, Postal Service headquarters
conveyed to area offices that it no longer wanted to receive reports
on bulk mail acceptance.  Some area offices told their district
offices that bulk mail management reports were no longer required. 
Postal Service headquarters program managers said that the
information derived from reports that were received was of so little
value that at one time they had considered eliminating them
altogether. 

When we asked each of the Postal Service's 85 district offices to
provide us with all acceptance reports (Forms 2867) for fiscal year
1994, we found that 7 did not prepare consolidated acceptance reports
for their districts.  When we compared the bulk business mail revenue
and volume reported on the reports with Postal Service headquarters'
estimates of total bulk business mail revenue and volume, we found
that the volume and revenue reported on the acceptance reports
represented only about one-half the revenue and volume estimated by
Postal Service headquarters. 

Management was also not receiving other required information that
would allow it to assess the adequacy of staffing and training at
mail acceptance units.  This missing information was to have been
provided each quarter to management in Quality Presort Verification
reports, which mail acceptance supervisors are required to fill out
for consolidation and use at each successive management level,
including Postal Service headquarters. 


      POSTAL SERVICE MANAGEMENT
      HAS INITIATIVES TO ADDRESS
      WEAKNESSES IN BULK MAIL
      ACCEPTANCE
---------------------------------------------------------- Letter :4.6

Although the Postal Inspection Service has long considered bulk
business mail acceptance to be a high-risk activity and has reported
on a number of control weaknesses,\13 top postal management has not
provided sustained attention to ensuring that adequate controls exist
for accepting bulk business mail.  Required information about bulk
mail acceptance that was to help management oversee the program has
not been received at Postal Service headquarters or some area offices
since the 1992 Postal Service reorganization. 

In the November 1995 issue of the Postal Bulletin, which is widely
distributed to the mailing public and within the Postal Service, the
Postmaster General announced that preventing revenue loss is a top
priority of the Postal Service.  He stated that "no business
[including the Postal Service] can afford to lose thousands of
dollars in uncollected revenue daily and expect to remain fiscally
viable for very long." He announced that "efforts are under way to
make improvements in mail acceptance and revenue collection areas."
The Postmaster General's sentiments, especially as they apply to bulk
mail acceptance, were repeated to us by numerous postal officials,
including inspectors with first-hand knowledge of the weaknesses in
the bulk mail acceptance system. 

At the completion of our review, postal management was taking a
number of actions that have the potential to significantly improve
bulk mail acceptance.  Postal officials told us that in October 1995,
they notified all area and district offices that completing Forms
2867 was mandatory and that the forms were to be completed and
forwarded to the Rates and Classification Center in Northern Virginia
for summarization.  In turn, summary reports are to be forwarded to
Postal Service headquarters for information purposes.  After the
reports are reviewed, irregularities are to be referred back to the
areas responsible for oversight. 

However, officials stated in February 1996 that even with the renewed
emphasis on the Forms 2867, compliance has been spotty.  They noted,
for example, that for accounting period 4 (December 9, 1995, to
January 5, 1996), only 51 of 85 districts submitted Forms 2867 as
required--fewer than the number we obtained for fiscal year 1994. 
The officials suspected that compliance has been incomplete because
many area and district officials came into their jobs following the
1992 reorganization and did not know or understand the significance
of bulk mail reporting.  Postal Service headquarters had not
explained the significance.  Postal officials attributed some of the
problems now occurring with bulk mail acceptance to outdated manuals. 
Officials told us they have been working on a new manual to replace
the old bulk mail acceptance manuals--DM102 and DM108.  As an interim
measure, officials told us that they planned to issue, in March 1996,
laminated cards for bulk mail acceptance clerks to use that would
include instructions on changes to bulk mail acceptance procedures
that the Postal Service is ready to make immediately. 

Additionally, the Postal Service has recently tested, and plans to
soon deploy, what it believes to be a better tool for verifying
barcodes--the Automated Barcode Evaluator (ABE).  According to postal
officials, ABE will assist acceptance clerks in evaluating barcoded
mail pieces and objectively determining whether the barcodes meet
Postal Service technical standards designed to ensure that the mail
piece can be sorted on the Postal Service's automated processing
equipment.  In February 1996, Postal officials said they were in the
process of purchasing about 260 ABEs for deployment to units that
accept the most barcoded mail, and officials said they will later
assess the need for additional ABEs. 

The Postal Service was also testing equipment, called Barcoding,
Addressing, Readability Quality Utilizing Electronic Systems
Technology (BARQUEST), to help its customer service representatives
identify bad barcodes and work with mailers to increase and improve
their barcoding.  BARQUEST is used to read and electronically store
images of mail pieces rejected by the Postal Service's automated
equipment at mail processing centers.  It is also supposed to allow
better monitoring of rejected mail and enable the Postal Service to
know if mailers' barcoding problems have been resolved.  As of
February 1996, the Service had deployed BARQUEST to 55 sites.  It
expects to deploy BARQUEST to 77 more sites by September 1996 and to
55 more sites during fiscal year 1997. 

Postal Service officials stated that to address the problem of failed
mailings being resubmitted and accepted without correction, the
Service is modifying bulk mail control system computer software to
capture information, by mailer, on failed mailings.  They stated this
change should enable the Service to identify mailings that have
failed verification and were not later identified as such when
resubmitted--a situation Service officials believed would suggest
that the mailer could have reentered the mail without correcting the
errors. 

In acknowledging the need for information on the extent of losses
associated with accepting improperly prepared mailings, the Postal
Service said in May 1996 that it would conduct an investigative
review to determine what methodologies might be applied in
identifying such losses.  We recognize there are a number of
methodologies that the Postal Service could use to determine the
extent of revenue losses.  We do not know of any one particular
methodology that would work best.  However, we believe there are a
number of possibilities that could be used, including (1) statistical
sampling, (2) ad-hoc studies, (3) cooperative studies with the
Inspection Service, (4) a systematic method for documenting and
reporting mailings that failed to meet Postal Service standards, and
(5) various combinations of these methods.  Other acceptable
methodologies may also exist.  Nevertheless, regardless of the
methodology the Postal Service employs, emphasis on identifying
losses resulting from accepting barcoded mail that does not meet the
Service's standards for automation compatibility is particularly
important because, with the rate reclassification initiative that
becomes effective in July 1996, the vast majority of discounts
granted are to be for barcoded mail.  Furthermore, producing such
information should not be a daunting task for the Postal Service
since, until the 1992 reorganization, it routinely captured and
reported the amount of barcoded mail that it was unable to process on
its automated equipment. 

Also, in late 1994, the Chief Financial Officer/Senior Vice President
of the Postal Service chartered a new revenue assurance organization
and charged it with ensuring that all revenue due the Postal Service
is collected.  This organization is to take a leadership role in the
coordination and development of effective internal controls over mail
acceptance and revenue collection.  The organization, which includes
a Postal Service headquarters manager, 4 staff, and 1 field
coordinator from each of the Postal Service's 10 areas, was given $10
million to identify and recover $100 million in potentially
uncollected revenue by the end of fiscal year 1996. 


--------------------
\13 See appendix II for a summary of Inspection Service work and
reports relating to control deficiencies in the bulk business mail
program. 


   MORE SELECTIVE VERIFICATION OF
   MAILINGS SHOULD PRODUCE GREATER
   RESULTS THAN THE CURRENT RANDOM
   APPROACH
------------------------------------------------------------ Letter :5

While the Postal Service may be able to gain reasonable assurance
that all revenue from bulk business mail is being received by
modifying and more closely following the requirements in its current
acceptance system, a better long-term solution may lie with the
adoption of a risk-based targeting system.  The Postal Service's
primary procedure for selecting bulk business mailings for in-depth
verification is to randomly sample 1 in 10 of each mailer's
statements.  This selection procedure for in-depth verification
applies to every mailer and does not differentiate the risk
associated with certain types of mailers or mailings and does not
selectively target high-risk mailers or mailings for closer scrutiny. 

As discussed earlier, acceptance clerks often have not done the
in-depth verifications called for by the Service's random sampling
plan.  They often disregarded the sampling plan and entered mail
directly into the mailstream without doing the required in-depth
verification.  Other federal agencies that collect revenue and
require employees to selectively verify financial data, such as IRS
and the U.S.  Customs Service, have dealt with large workloads by
developing more selective, risk-based sampling plans. 


      IRS AND CUSTOMS USE RISK OF
      ERROR IN SAMPLING STATEMENTS
      FOR REVIEW
---------------------------------------------------------- Letter :5.1

IRS and Customs are more selective than the Postal Service in their
sampling approaches.  Both IRS and Customs place more emphasis on
auditing those returns and inspecting imports that offer the highest
potential for yielding the most significant results.\14

IRS officials told us that prior to the early 1960s, income tax
returns were chosen for audit through a costly process that relied on
the agency's most experienced revenue agents to manually "eyeball"
returns to ensure taxpayers paid the correct amount of tax.  Later,
IRS refined this process by computerizing criteria used in the manual
process.  In the late 1960s, IRS began developing the system
currently in use--discriminant function analysis (DIF).  This
multivariate statistical selection technique allows IRS to
differentiate among tax returns on the basis of each return's
probability of containing errors.  Instead of using a system that
selects randomly from the entire universe, as the Postal Service
does, IRS uses DIF to screen all individual income tax returns
received annually and identify those more likely to result in a tax
change. 

According to IRS, its system decreases the number of returns audited
that produce no tax changes and reduces the amount of IRS staff and
computer time needed to screen returns.  IRS believes that the DIF
system has significantly increased its efficiency by allowing it to
concentrate its limited audit resources on those tax returns with a
high probability of error, thereby helping ensure that taxpayers who
might otherwise underpay, in fact, pay their fair share.  Further,
IRS does not have to inconvenience as many taxpayers with audits that
produce no change in the tax due, which is a benefit that the Postal
Service might also achieve because in-depth verifications can
inconvenience mailers. 

Like the Postal Service and IRS, the U.S.  Customs Service must
balance the requirements of its mission with the expectation that
enforcement will not disrupt the normal flow of business.  Customs
must determine whether goods entering the United States are properly
classified and correctly valued.  From 1842 to the early 1980s,
Customs' policy for enforcing import laws was to examine a portion of
all cargo shipments, although most of those examinations were
cursory.  Recognizing in the early 1980s that it had to contend with
increasing levels of imports, numerous demands, and limited
resources, Customs shifted its trade enforcement efforts from a
strategy of checking all imports to one of selecting and inspecting
only high-risk imports.  Customs said that it is continuing to refine
and improve this system to meet present-day challenges. 

The Customs system focuses on compliance measurement, enhanced
targeting, and trend analysis.  According to the Customs Service,
fiscal year 1995 marked the first year that Customs implemented a
national compliance measurement program.  According to Customs, it
now has a compliance baseline across a multitude of importing areas,
such as industry, importer, consignee, and country.  Using this data,
Customs said that it is targeting its fiscal year 1996 trade
enforcement efforts toward the most important areas of noncompliance. 
Customs also is randomly selecting shipments to examine in order to
monitor compliance rates and adjust its targeting of high-risk areas,
as necessary.  As a consequence, Customs said that it expects to
increase its targeting efficiency, which will result in more
productive use of its resources, and to reduce attention to areas of
high compliance, thereby facilitating the flow of imports into the
United States. 


--------------------
\14 IRS and Customs also use penalties to encourage accurate
reporting of amounts due the government.  The Postal Service does not
use penalties because it believes they would be counterproductive to
customer service. 


      POSTAL SERVICE IS STUDYING
      THE FEASIBILITY OF TARGETING
      HIGH-RISK MAILINGS
---------------------------------------------------------- Letter :5.2

In March 1994, the Postal Service awarded a contract to a university
professor to study the feasibility of using a risk assessment
approach to sampling bulk mailer statements.  The professor was to
determine whether the Postal Service could identify and quantify
factors that could be used to select mailings or types of mailings on
the basis of the relative risk of mail preparation errors. 
Additionally, the contract called for the professor to explore other
means of improving verification procedures and is scheduled to be
completed in July 1996.  Postal Service officials also stated that as
part of a benchmarking effort, they had contacted IRS and Customs in
late 1995 regarding their methodology for targeting cases for
audit/inspection. 

In February 1996, postal officials told us they expect to put a
completely redesigned bulk mail acceptance system into place by
December 1996 that incorporates a risk-based targeting system.  In
their comments on this report, they also said they plan to do a
staffing requirements analysis as soon as design decisions are made
on the new acceptance system.  Additionally, they said they plan to
issue a new bulk mail acceptance manual when the new acceptance
system is put in place. 


   CONCLUSIONS
------------------------------------------------------------ Letter :6

In fiscal year 1994, the Postal Service derived 48 percent ($23
billion) of its total mail revenue from bulk business mail.  Yet,
weaknesses in the Postal Service's controls for accepting bulk
business mail prevent it from having reasonable assurance that all
significant amounts of postage revenue due are received when mailers
claim presort/barcode discounts. 

Postal Service headquarters recognized in the late 1980s that it
needed to manage its bulk mail acceptance system more effectively and
took steps to do so in 1989 and 1990.  However, according to
officials we spoke with, the system deteriorated after the 1992
reorganization. 

With an estimated $8 billion in discounts allowed in fiscal year
1994, and larger amounts expected as the Postal Service reclassifies
its postage rates and moves closer to full automation in 1997,
sustained top-level management attention is needed to establish and
maintain adequate controls over bulk business mail acceptance.  This
attention can help ensure that required verifications of bulk
mailings, including barcodes, are done and that any errors noted are
corrected before bulk mail is accepted into the U.  S.  mail system. 

Recently, the Postal Service launched a number of initiatives to
improve the bulk business mail acceptance system.  It is too early to
know whether these initiatives will eventually correct the internal
control problems detailed in this report.  However, because they do
address many of the problems, we believe that if they are implemented
as planned and monitored appropriately, the initiatives can improve
bulk mail acceptance operations. 


   RECOMMENDATIONS
------------------------------------------------------------ Letter :7

Because it is too early for us or the Postal Service to reasonably
predict the outcome of its many initiatives to improve bulk mail
acceptance, we are making several recommendations.  We recognize that
the Service's initiatives offer the promise of correcting many of the
concerns raised in this report.  However, we believe recommendations
are warranted as a means of fostering sustained management attention
until the bulk mail acceptance system is operating effectively and
providing the Postal Service with reasonable assurance that all
significant amounts of bulk mail revenues are being collected. 

Specifically, we recommend that the Postmaster General direct bulk
mail acceptance program supervisors and managers to periodically
report to appropriate Service levels on operation of the bulk mail
acceptance system, initiatives, and the progress and effectiveness of
related improvements so management can be reasonably assured that

  -- required mail verifications, including supervisory reviews, are
     done and that the results are documented as required;

  -- mailings resubmitted following a failed verification are
     reverified and errors are corrected;

  -- acceptance clerks and supervisors are provided with adequate,
     up-to-date procedures, training, and tools necessary to make
     efficient and objective verification determinations;

  -- information on the extent and results of verifications,
     including supervisory reviews, is regularly reported to
     appropriate levels, including Postal Service headquarters, and
     that such information is used regularly to assess the adequacy
     of controls and staffing, training needs, and acceptance
     procedures; and

  -- risk becomes the prominent factor in determining mailings to be
     verified. 

Also, we recommend that the Postmaster General direct bulk mail
acceptance program managers to develop methodologies that can be used
to determine systemwide losses associated with accepting improperly
prepared mailings. 


   AGENCY COMMENTS AND OUR
   EVALUATION
------------------------------------------------------------ Letter :8

In its written comments on a draft of this report, the Service
acknowledged that many long-standing problems exist with bulk mail
acceptance, and it expressed confidence that the initiatives it has
under way, which were cited in our report, will remedy acceptance
weaknesses in the bulk mail program and address the report's
recommendations.  The Postal Service said that almost all of the
remedies will be in place later this year or early 1997.  The
Service's written comments are included as appendix III. 

Only after sufficient time has elapsed can we or the Postal Service
tell if these initiatives will correct the problems.  The initiatives
cited by the Service appear to offer promise, but they can easily
falter unless there is strong and continuing top-down commitment to
improving bulk mail acceptance.  In commenting on a draft of this
report, the Postal Service said it is putting increased emphasis on
management oversight of the bulk mail acceptance function at all
levels of the organization. 


---------------------------------------------------------- Letter :8.1

As arranged with the Subcommittee, 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 and other
interested parties.  Copies will also be made available to others
upon request. 

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

Sincerely yours,

J.  William Gadsby
Director, Government
Business Operations Issues


BACKGROUND INFORMATION ON THE
POSTAL SERVICE'S BULK MAIL
ACCEPTANCE PROGRAM
=========================================================== Appendix I

Of the 177 billion pieces of mail processed in 1994 by the Postal
Service, over 118 billion pieces, or 67 percent, were categorized as
bulk mail.  This mail typically arrived at Postal Service mail entry
units in sacks, trays, or on pallets and was mostly
business-generated.  In fiscal year 1994, the Service recorded
revenue from bulk business mail of $23.1 billion--48.4 percent of its
total mail revenue. 

In 1976, the Postal Service began offering postage discounts to
mailers who presorted their mail, and in 1988 it began offering
discounts for barcoding.  The presort and barcode discounts are to
compensate mailers for performing work that otherwise would have to
be done by the Postal Service.  The amount of discount depends on the
depth of work performed by the mailer, e.g., barcoded mail sorted in
delivery point sequence receives a larger discount than nonbarcoded
mail sorted to a 3-digit ZIP Code level. 

Over the years, the total dollar value of business mailer discounts
for presorting and barcoding has grown, and is expected to continue
growing, as the Postal Service moves closer to achieving its goal of
having about 90 percent of all letter mail barcoded by the end of
1997.  The Postal Service estimates that by 1997, 14,000 pieces of
automated equipment costing about $5 billion will have been deployed
to sort the mail. 

In fiscal year 1994, about 34 percent of all First-Class mail was
discounted, and 92 percent of all third-class mail was discounted. 
According to Postal Service studies, the value of these discounts
totaled about $8 billion. 


      BARCODED MAIL EXPECTED TO
      GROW
------------------------------------------------------- Appendix I:0.1

One of the Postal Service's major long-term strategies is to fully
automate mail processing by barcoding almost all letter mail and
processing it on automated barcode sorting equipment.  Processing
letters using automation is more cost-effective than mechanized or
manual processing.  According to the Service, the comparative costs
of processing letters are $3 per thousand using automation, $19 using
mechanized letter-sorting machines, and $42 when done manually. 
Thus, if the Service receives a barcoded letter that must be sorted
by mechanized or manual methods, its processing cost will be about 6
or 14 times the automated cost. 

Under a mail reclassification initiative, in which the Postal Rate
Commission recommended in January 1996 new postage rates for certain
mail, the discount for automation-compatible mail will increase and
the discount for presort-only will decrease.  For example, as
recommended by the Commission, the discount for a First-Class
barcoded letter sorted to a 5-digit ZIP Code level will increase from
6.2 to 8.2 cents, and the discount for a presorted-only letter will
decrease from 4.6 to 2.5 cents.  Similarly, the discount for a
third-class barcoded letter sorted to a 5-digit ZIP Code level will
increase from 15.4 to 16.5 cents, and the discount for a
presorted-only letter will decrease from 13.2 to 11.1 cents.  The
Postal Service expects that adoption of this change, most of which
was approved by the Board of Governors and will become effective July
1996, will increase the First-Class and third-class barcoded mail
volumes by 7 and 12 percent, respectively. 


      ACCEPTANCE UNITS VERIFY
      MAILER SORTING AND BARCODING
------------------------------------------------------- Appendix I:0.2

Most bulk business mail is entered at Business Mail Entry Units
(BMEUs) and Detached Mail Units (DMUs), located throughout the Postal
Service's 85 districts.  DMUs are postal acceptance units located at
mailers' mail preparation facilities.  BMEUs are often located in or
adjacent to large mail processing plants, which are postal facilities
that process mail for distribution to both local and national
destinations.  Bulk mail is also entered at many of the 40,000 post
offices located throughout the country. 

BMEUs typically include a parking/staging area for large trucks and
other vehicles that transport mail from mailers to the BMEU facility. 
They also include a dock for unloading the mail; an area where
acceptance clerks can inspect the mail; and a counter area where
paperwork, such as mailing statements, can be examined and other
business transactions can be completed.  Once the mail has been
accepted by a BMEU mail acceptance clerk, it moves inside the plant
for processing.\15

The Postal Service's mail acceptance clerks are the gatekeepers for
accepting bulk business mail into the mailstream.  It is their job to
ensure, before mail enters the Postal Service's processing and
distribution facilities, that mailers have prepared their mail in
accordance with postal requirements and that discounts given for
presorting and barcoding have, in fact, been earned.  This is a
difficult task given the time constraints and the wide variation in
the way bulk business mail can be prepared and still meet Postal
Service standards. 

If mail barcoded by a mailer is accepted by acceptance clerks but
later fails to run on postal barcode sorters, then the Postal Service
incurs additional costs.  This is because the Postal Service must
rework the mail at its own expense even though it gave the mailer the
barcoded rate to perform that work.  Additional processing costs
incurred by the Postal Service are ultimately reflected in higher
postage rates, unfairly penalizing those mailers who properly prepare
their bulk business mailings. 


--------------------
\15 For additional information on mail processing see Postal Service: 
Automation Is Taking Longer and Producing Less Than Expected
(GAO/GGD-95-89BR, Feb.  22, 1995). 


      VERIFICATIONS ARE TO DETECT
      MAILER ERRORS AND ENSURE
      THAT CORRECT POSTAGE IS PAID
------------------------------------------------------- Appendix I:0.3

Verifications performed by mail acceptance clerks fall into two
categories:  (1) cursory reviews of all mailings, and (2) in-depth
verifications of randomly selected mailings.  In performing a cursory
review, acceptance clerks are to randomly check some sacks, trays, or
pallets to verify that (1) the mail is prepared as stated on the
mailer's mailing statement, (2) the number of mail pieces indicated
on the mailing statement is accurate, and (3) the mailer applied the
appropriate postage rates. 

In-depth reviews are to be performed on at least 1 in every 10
mailings submitted by each mailer.  The mailing chosen for an
in-depth review is to be selected at random, and, in most cases,
three sacks, trays, or pallets are to be rigorously inspected to
ensure that the mail was prepared correctly and that all discount
qualifications were met.  A mailing may fail verification for a
number of reasons.  For example: 

  -- Mail pieces do not meet minimum or maximum size standards. 

  -- Addresses are not in the Optical Character Reader's scan area. 

  -- Fonts cannot be read by the Postal Service's automated
     equipment. 

  -- Barcodes do not meet technical specifications. 

  -- The contrast between paper and ink is insufficient. 

  -- There are less than three lines used for the address block. 

  -- The spacing between city, state, and ZIP Code is improper. 

  -- The barcode/address can shift out of the viewing area in window
     envelopes. 

  -- Presort mail is not labeled correctly. 

When verifying mailings, if the acceptance clerk determines that more
than 5 percent of a mailing is not prepared correctly, then the
mailing is failed.  The mailer then has two options:  (1) rework the
mail so that it meets postal specifications and qualifies for the
bulk postage rate applied for or (2) pay the additional single-piece
postage rate for that percentage of the entire mailing estimated to
be in error. 

For every in-depth inspection completed, mail acceptance clerks are
required to prepare a written verification report (Form 2866).  This
report is used to (1) document the results of the verification, (2)
notify mailers of the types of errors found, and (3) aid supervisors
in performing quality presort verifications (QPV).  A QPV entails a
supervisor rechecking an in-depth verification performed by a clerk. 
Postal facilities that receive 100 or more bulk mailings during a
4-week accounting period are to prepare a consolidated bulk mail
acceptance report (Form 2867) documenting the results of their
in-depth verifications. 


      RESPONSIBILITY FOR BULK
      ACCEPTANCE CONTROLS ASSIGNED
      TO MARKETING OFFICIALS
------------------------------------------------------- Appendix I:0.4

At Postal Service headquarters, management responsibility for the
bulk business mail program resides with the Vice President of
Marketing Systems, who reports to the Chief Marketing Officer and
Senior Vice President.  Area Vice Presidents and district managers
are responsible for ensuring that bulk mail acceptance activities
conform to prescribed standards within their geographic span of
control. 


SUMMARY OF INSPECTION SERVICE WORK
RELATING TO BULK MAIL ACCEPTANCE
========================================================== Appendix II

During the late 1980s and early 1990s, the Postal Inspection Service
reported to postal management, on several occasions, that existing
bulk mail acceptance controls were inadequate for preventing revenue
losses.  In 1986, following a national audit of the Postal Service's
revenue protection program, the Inspection Service reported that
procedures for mail acceptance, verification, and classification were
not being effectively administered.  It noted that few of the
employees it interviewed felt that revenue protection was part of
their job and that this lack of awareness and commitment was
resulting in millions of dollars in postage not being collected. 

In November 1991, following a national operational audit of the bulk
mail acceptance system, the Postal Inspection Service observed that
bulk mailings posed a serious risk to Postal Service revenue.  It
concluded that Postal Service organizational changes in 1986 and 1990
had adversely affected the management oversight necessary to ensure
that bulk mail acceptance programs operated as intended.  The
Inspection Service also concluded that this condition had increased
the risk of revenue loss through noncollection of postage and
unnecessary mail processing costs due to acceptance of improperly
prepared bulk mailings.  The Inspection Service found that internal
controls at plant load operations had been allowed to deteriorate and
become unreliable.\16 It stated that this exposed the Postal Service
to serious risk by allowing situations to exist where large mailings
could enter the mailstream without payment of postage. 

In early 1993, the Inspection Service conducted a nationwide review
of the Plant Verified Drop Shipment Program.\17 The Inspection
Service reported that internal controls were not effectively or
consistently applied and that there was a significant risk that mail
could be entered into the mailstream without payment of postage and
that mailers could claim unearned discounts. 

Although losses were not the primary focus of its audits, the
Inspection Service did document and report to management some losses
during this period.  For example, in fiscal year 1994, the Inspection
Service documented losses totaling about $8 million.  These losses,
however, should not be considered all-inclusive because they were not
identified in any systematic manner.  Rather, they were identified as
the Inspection Service was following through on customer complaints,
anonymous tips, management requests, leads developed during financial
audits, and leads provided by other sources.  The losses resulted
from mailers not paying full postage for reasons varying from
understating the number of pieces being mailed to manipulating the
computer software used for generating mailing statements so that the
mailing statements misrepresented, in the mailer's favor, the make-up
of the mailing. 

In 1993, to gain a better understanding of the magnitude of the
losses resulting from mailer preparation errors, the Inspection
Service established a task force that is taking a more systematic
long-term approach to identifying fraudulent mailings that have
resulted in revenue losses in the bulk business mail program. 
According to Inspection Service officials, this approach is being
taken in order to demonstrate to postal management the need to
improve controls over bulk mail acceptance.  Additionally, as of May
1996, the Inspection Service was conducting a National Coordination
Audit on the topic of bulk business mail.  The objectives of the
audit are to (1) conduct a corporate-level review and evaluation of
the alignment of the goals and objectives of bulk business mail
acceptance with the CustomerPerfect!sm initiatives, and (2) provide
an economic value added assessment of bulk business mail in relation
to the corporate goals of the Postal Service.  According to the
Inspection Service, this audit will include and address the following
issues: 

  -- inconsistencies among acceptance units,

  -- balancing the goals of unit operations and improving customer
     service,

  -- conflicts between dispatch and delivery times with customer
     satisfaction,

  -- inability to do a "good job" due to time pressures,

  -- adequacy of training,

  -- understanding of national instructions at the local level, and

  -- identification of new initiatives affecting bulk business mail. 



(See figure in printed edition.)Appendix III

--------------------
\16 A plant load is an operation in which the Postal Service accepts
mail at a customer's plant and provides transportation from the plant
in order to preclude handling at the local post office. 

\17 Under the Plant Verified Drop Shipment Program, mailings, once
accepted by postal personnel, are released back into the control of
the mailer, who provides transportation to the destination entry
point. 


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



(See figure in printed edition.)



(See figure in printed edition.)


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix IV

GENERAL GOVERNMENT DIVISION,
WASHINGTON, D.C. 

Michael E.  Motley, Associate Director, Government Business
Operations  Issues
James T.  Campbell, Assistant Director
Lawrence R.  Keller, Assignment Manager
Roger L.  Lively, Senior Evaluator
Charles F.  Wicker, Senior Evaluator
Kenneth E.  John, Senior Social Science Analyst

BOSTON/NEW YORK REGIONAL OFFICE

James D.  VanBlarcom, Senior Evaluator
John J.  Schick, Evaluator

DALLAS FIELD OFFICE

Donna B.  Svoboda, Computer Specialist

DENVER FIELD OFFICE

James S.  Crigler, Evaluator-in-Charge
Robert W.  Stewart, Evaluator

*** End of document. ***