Automated Export System: Prospects for Improving Data Collection and
Enforcement Are Uncertain (Letter Report, 11/14/97, GAO/NSIAD-98-5).

Pursuant to a congressional request, GAO reviewed the potential impact
of the the Customs Service's Automated Export System (AES) and the views
of the export community regarding AES, focusing on whether AES is likely
to achieve its objectives of improving export data, enhancing
enforcement efforts, and streamlining export data collection.

GAO noted that: (1) it is not yet clear what benefits will result from
the use of AES because many critical implementation issues remain
unsolved; (2) although AES has the potential to improve export reporting
and enhance enforcement efforts, it is unlikely to achieve these
objectives unless more exporters are willing to participate and
limitations that prevent other agencies from fully using the system are
resolved; (3) concerning the trade community's limited participation,
GAO found that: (a) only a small fraction of the export community is
using AES; (b) most exporting companies responding to GAO's survey are
not likely to use AES over the next 3 years; and (c) twenty-five percent
of all U.S. ocean freight forwarders had not heard of AES; (4) benefits
cited by companies using AES include automated filing, reduced
paperwork, personnel, and administrative costs, participating in the
initial development of AES, and filing all data at a central filing
point; (5) some segments of the trade community contend that the
predeparture filing requirement is inconsistent with their business
practices and costly; (6) AES is designed to help target illegal
shipments, identify high-risk shipments, and compile exporter histories;
(7) the system's usefulness as an enforcement tool is limited because:
(a) it is not linked with the databases of other law enforcement
agencies; (b) a proposal to allow exporters to file data after shipment
could undermine efforts to detect export violations; (c) AES allows
export data to be transmitted only hours before a shipment departure,
which may not provide sufficient time to target possible illegal
shipments; and (d) many Customs inspectors anticipate that illegal
exporters are unlikely to use AES to file their export data; (8) AES
faces limitations in achieving its goal to create a single information
collection and processing center for the electronic filing of required
export documentation; (9) many export-related agencies are subject to
existing regulations requiring them to retain their own licensing
procedures and have requirements that will not be satisfied through AES;
(10) Customs is attempting to resolve these issues through several
means; and (11) a cost-benefit analysis is needed to determine how or
whether to proceed with implementation of AES.

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

 REPORTNUM:  NSIAD-98-5
     TITLE:  Automated Export System: Prospects for Improving Data 
             Collection and Enforcement Are Uncertain
      DATE:  11/14/97
   SUBJECT:  International trade
             Statistical data
             Data collection
             Interagency relations
             Export regulation
             Customs administration
             Data integrity
             ADP
             Law enforcement information systems
IDENTIFIER:  Customs Service Automated Export 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, Committee on the Judiciary, U.S.  Senate

November 1997

AUTOMATED EXPORT SYSTEM -
PROSPECTS FOR IMPROVING DATA
COLLECTION AND ENFORCEMENT ARE
UNCERTAIN

GAO/NSIAD-98-5

Automated Export System

(711238)


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

  ABI - Automated Broker Interface
  AERP - Automated Export Reporting Program
  AES - Automated Export System
  AES-PASS - Automated Export System Post-Departure Authorized
     Special System
  DEA - Drug Enforcement Administration
  FOB - free on board
  IOU - "I owe you"
  MOU - memorandum of understanding
  NVOCC - Non-Vessel Operating Common Carrier
  SED - Shipper's Export Declaration

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


B-278136

November 14, 1997

The Honorable Orrin Hatch
Chairman, Committee on the Judiciary
United States Senate

Dear Mr.  Chairman: 

Since the 1980s, federal agencies responsible for compiling U.S. 
trade statistics and enforcing U.S.  export laws have experienced
serious problems in obtaining accurate and timely data on exports. 
To improve reporting and enhance enforcement efforts, the U.S. 
Customs Service and the Census Bureau in 1991 initiated the Automated
Export System (AES).  AES allows exporting companies to
electronically enter data on shipments, automatically checks for
errors, and provides data to help detect export violations.  Although
it introduced the system as voluntary, Customs recognized that
achieving a high level of participation in AES would be difficult and
wanted its use to be mandatory.  In 1996, Customs sought to make AES
mandatory for some users.  However, AES was not made mandatory, in
part because companies raised concerns about the impact of AES on
their business practices, especially if they are required to enter
data before shipments depart. 

At your request, we reviewed the potential impact of AES and the
views of the export community regarding AES.  Specifically, we sought
to determine whether AES is likely to achieve its objectives of
improving export data, enhancing enforcement efforts, and
streamlining export data collection and, in so doing, we obtained the
export community's views on AES.  As you requested, we are also
providing information on the export procedures of other selected
countries.  (See app.  I for information on selected countries.)


   BACKGROUND
------------------------------------------------------------ Letter :1

Merchandise trade, the exchange of goods with other nations, is an
increasingly important component of the U.S.  economy.  The U.S. 
Customs Service collects data on imports and exports that the U.S. 
Census Bureau uses to produce statistics on U.S.  trade.  While
Customs has numerous import responsibilities, its export functions
include guarding against the exportation of illegal goods, such as
protected technologies, stolen vehicles, and illegal currency. 

Customs has broad authority to enforce export laws and regulations. 
However, it has historically placed more emphasis on imports than on
exports.  While U.S.  import data is recognized as generally
reliable, export data is viewed as less accurate.  A 1997 Census
report notes that the value of U.S.  exports has probably been
underreported by between 3 and 7 percent but could be understated by
as much as 10 percent.\1 Underreporting of exports can significantly
affect the accuracy of statistics on the nation's trade balance. 
Inaccurate trade statistics can be an impediment in negotiations for
bilateral and multilateral trade agreements.  Export statistics also
are relied on by the government to calculate the gross domestic
product (which is used to assess the performance of the U.S. 
economy) and to determine appropriate promotional programs for
expanding exports.  Export data is also used to establish controls on
sensitive exports. 

A primary source of export statistics is information that is recorded
on a form called the Shipper's Export Declaration (SED).  The SED
contains information such as the nature, value, quantity, and
destination of the goods to be exported.  Generally, exporters or
their agents are required by regulation to file the SED for each
export transaction having a value over a certain amount, now set at
$2,500 for all shipments without a license.\2 Under current
regulations, the SED must be delivered to the exporting carrier prior
to exportation.\3 Ocean and air carriers, with a bond on file,\4 are
permitted to file the complete manifest (a carriers manifest lists
all the cargo it is transporting) with Customs within 4 working days
after departure.  For overland shipments, the SED must be presented
to Customs at the time of export. 

The major sources of error in merchandise trade statistics include
missing SEDs and incomplete or inaccurate reporting.  Since the
1980s, Customs, Census, and other government agencies have conducted
numerous studies, which found serious problems with companies
properly completing the document and filing it at the required time
and place.  For example, Customs completed an audit of selected ocean
vessel manifests in 1996 that found about 29 percent of shipments
listed on the manifest did not have the required SED.  In 1997,
Customs conducted an audit of airline manifests that determined 40
percent of SEDs were incorrectly completed.  Without a properly
completed SED, an export is either not recorded or recorded
incorrectly. 

Currently, about one-third of all export transactions are recorded on
paper SEDs.  Census collects another third\5 of the export data on a
monthly basis directly from exporting businesses through an
electronic filing system known as the Automated Export Reporting
Program (AERP).  Census is terminating AERP in December 1999 because
it believes the system is outdated and that AES will provide more
accurate data.  (Twenty-five percent of all AERP transmissions
contain errors that must be corrected.) Census officials stated that
AERP has systems limitations related to the amount of data it can
accept.  They stated that the system would require a complete
redesign in methodology and computer technology in order to be able
to accept more participants and improve data quality, resulting in a
system similar to AES.  Census officials also noted that about
one-third of all AERP participants submit their data late via AERP. 


--------------------
\1 See "Understatement of Export Merchandise Trade Data," U.S. 
Bureau of the Census, Foreign Trade Division (Washington, D.C.:  Jan. 
1997). 

\2 Mail shipments must be reported if they have a value over $500. 

\3 15 C.F.R.  ï¿½ 30.12. 

\4 15 C.F.R.  ï¿½ 30.24. 

\5 The remaining one-third of export data is captured by the
U.S.-Canada data exchange, whereby the United States and Canada
exchange import records and use them to determine each country's
exports to one another.  For more information on the data exchange,
see Measuring U.S.-Canada Trade:  Shifting Trade Winds May Threaten
Recent Progress (GAO/GGD-94-4, Jan.  19, 1994). 


      CONTEXT FOR AES' DEVELOPMENT
---------------------------------------------------------- Letter :1.1

Customs and Census initiated AES in 1991 to improve (1) the
collection and reporting of export statistics and (2) the enforcement
of export regulations.  Initially, the system was designed to replace
the manual process of handling paper SEDs with a more efficient and
less costly automated process that would increase the accuracy,
completeness, and timeliness of SED data. 

AES is an interactive system that allows exporters or their agents to
electronically transmit SED information directly to Customs before a
carrier's departure.  In order to improve the quality of export data,
data transmitted via AES is subjected to a series of automatic edits. 
The system in turn sends back to the exporter a message to check the
data if it does not fall within statistical parameters developed by
Customs and Census.  (See app.  II for information on how AES works.)

According to Customs, AES was also designed to improve the
enforcement of export controls by evaluating the risk of export
shipments based on certain criteria, such as the country of
destination and the type of cargo;\6 compiling exporter histories;
allowing for trend analysis; and providing inspectors with detailed
commodity data prior to departure.  Customs officials believe that
the more export information they have, the more focused their efforts
to target illegal shipments will be. 

Finally, in 1994, in response to several initiatives including the
Vice President's National Performance Review, Customs decided that
AES should be expanded to provide a centralized database for
collecting and processing export documentation required by the U.S. 
government.  Customs planned to work with other U.S.  government
agencies that have export-related responsibilities to help these
agencies meet their export information requirements through AES. 
(See p.  18 for a discussion of the present status of the single
electronic filing center.)

Customs installed AES in all U.S.  vessel ports in October 1996, and
currently it is operational in all ports, including air, rail, and
truck transit ports.\7 Customs and Census officials estimate that
they spent approximately $12.9 million to develop and implement AES
from fiscal
year 1992 to 1997.  These costs include, among other things, expenses
for contractors, travel, and training.  According to Customs' and
Census' figures, both agencies estimate that together they will spend
an additional $32.2 million through fiscal year 2002 on AES
implementation and maintenance. 

This new system would require companies to make various changes in
how they submit their export data to Customs.  Companies that submit
their export data via AERP will have to undergo modifications in
programming the processing of their export data or return to
submitting paper SEDs.  Companies that submit their data via paper
SEDs but want to participate in AES will have to automate their
export processing, and/or purchase AES software, or use the
facilities of a port authority or service center to transmit their
data. 

In addition, some segments of the trade community have alleged that
AES will require major changes in their current business practices. 
Because Customs has not strictly enforced the legal requirement that
companies submit their SEDs to the carrier prior to departure, many
companies have grown accustomed to turning in their SEDs to the
carriers late.  AES will require that companies file their export
data directly to Customs, rather than the carrier, prior to
departure. 

The trade community has varying views on AES.  To obtain these views,
we conducted two surveys of potential AES users:  (1) a nationally
representative sample of 400 U.S.  ocean freight forwarders and
Non-Vessel Operating Common Carriers (NVOCC)\8 and (2) 80 U.S. 
exporting companies that file paper SEDs.  We also interviewed
officials from 12 of the largest U.S.  sea and air carrier companies
and several trade groups representing various segments of the export
community.  (See app.  III for complete survey results.) (We did not
independently verify information provided by U.S.  companies and
trade groups.) We also interviewed Customs officials at 13 sea, air,
and land ports.  As of June 1997, we completed the surveys and
interviews. 


--------------------
\6 Customs officials estimate that they inspect less than 1 percent
of export shipments.  Customs has not historically tracked the number
of seizures that have resulted from its inspections.  However, as of
June 1997, all ports have been required to collect data on all
outbound cargo inspections, including all commercial or personal
shipments exported by any mode of transportation, and the resulting
number of seizures of illegal exports. 

\7 AES is not yet accepting transportation data for air, rail, and
truck carriers.  According to Customs officials, AES will be
accepting air and rail transportation data by the end of 1998 and
truck transportation data in 1999. 

\8 Freight forwarders and NVOCCs make the arrangements for the
shipment of goods for most exporters, including the filing of
information required by Customs, Census, and other federal agencies. 
NVOCCs differ from freight forwarders in that they may also lease a
vessel to transport the goods.  For ease of presentation, hereafter
we refer to both groups as "freight forwarders." Although our freight
forwarder study population consisted of active licensed ocean freight
forwarders and NVOCCs listed by the Federal Maritime Commission, and
is not representative of air freight forwarders, 85 percent of the
company officials we spoke with told us that their firms provide
services to clients exporting by air. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :2

It is not yet clear what benefits will result from the use of AES
because many critical implementation issues remain unsolved. 
Although AES has the potential to improve export reporting and
enhance enforcement efforts, it is unlikely to achieve these
objectives unless more exporters are willing to participate and
limitations that prevent other agencies from fully using the system
are resolved.  Concerning the trade community's limited
participation, we identified the following: 

  -- Only a small fraction of the export community--37 companies--is
     currently using AES, and the data entered represents less than 1
     percent of all exports. 

  -- Our survey of exporting companies suggests that most of them are
     not likely to use AES over the next 3 years. 

  -- Twenty-five percent of all U.S.  ocean freight forwarders had
     not heard of AES. 

The primary benefit cited by companies using AES and planning to use
AES was automated filing.  Other benefits cited included the
potential for reducing paperwork and personnel and associated
administrative costs, participating in the initial development of
AES, and filing all export data at a central filing point.  However,
for some segments of the trade community, these benefits are
outweighed by their concerns regarding the predeparture filing
requirement, which they contend is inconsistent with their business
practices and will be costly and burdensome to their business. 

AES is designed to improve Customs' enforcement efforts by providing
detailed commodity data to help the inspector target illegal
shipments.  It also is designed to provide a database that can be run
against specific criteria to identify high-risk shipments and compile
exporter histories.  However, the system's usefulness as an
enforcement tool is limited for several reasons.  First, as currently
designed, it is not linked with the databases of other law
enforcement agencies.  Second, under a program to facilitate the use
of AES, Customs plans to permit approved exporters to file export
data after shipment, which could undermine efforts to detect export
violations before illegal goods are transported.  Third, AES allows
SED information to be transmitted only hours before a shipment's
departure, which in most cases will not provide inspectors sufficient
time for targeting possible illegal shipments.  Finally, many Customs
inspectors anticipate that illegal exporters are unlikely to use AES
to file their export data, further limiting AES' potential as an
enforcement tool. 

Also, AES faces limitations in achieving its goal to create a single
information collection and processing center for the electronic
filing of export documentation required by the U.S.  government.  The
system was also designed to reduce related export documents.  Many
export-related agencies are subject to existing regulations requiring
them to retain their own licensing procedures.  Ten of the 13
agencies with major export-related responsibilities say they have
export requirements that will not be fully satisfied through AES. 

Customs is attempting to resolve these issues through several means. 
For example, the Commissioner of Customs has proposed that Customs
enter into formal negotiations with the trade community to resolve
outstanding issues of concern.  In addition, Customs is continuing to
work with other agencies to identify any of their export requirements
that can be potentially incorporated into AES. 

Because Customs and Census plan to spend $32 million beyond the $13
million expended to date, we are concerned that Customs and Census
will allow the system to proceed without fully resolving these
issues.  In our opinion, AES is at a critical juncture in its
implementation.  Senior management attention at both Customs and
Census is needed to determine the extent to which concerns can be
addressed that affect participation by the export trade community,
that deal with the improvement of export enforcement, and that
respond to the requirements of other agencies with export-related
functions.  After addressing these concerns, a cost-benefit analysis
of AES is needed to determine how or whether to proceed with
implementation. 


   FACTORS HINDERING AES' ABILITY
   TO MEET ITS OBJECTIVES
------------------------------------------------------------ Letter :3

Although AES has features to improve export data collection and
enforcement efforts and to reduce paperwork, the system's
effectiveness is hindered by low participation of the export
community.  Unless AES participation increases significantly, AES
will not enhance the quality of export data or the enforcement of
export regulations.  In addition, other factors may limit AES'
ability to achieve its objective of enhancing export control
enforcement.  For example, Customs' plans to introduce a
post-departure filing program may impede the system's effectiveness
as a tool for targeting illegal exports.  Finally, AES will not
likely serve as a central point for collecting and processing all
export documents because other export-related agencies have
information needs that they say cannot be fulfilled through AES. 


      TRADE COMMUNITY
      PARTICIPATION IS LOW AND
      LIKELY TO BE LIMITED IN THE
      NEAR FUTURE
---------------------------------------------------------- Letter :3.1

Trade community participation in AES is currently very limited, and
our work showed most companies do not have immediate plans to
participate in AES.  As of September 1997, AES participants included
8 exporters, 27 freight forwarders, and 2 sea carriers out of tens of
thousands of export-related businesses.  Currently, less than 1
percent of all export data is being submitted via AES. 

Customs expects participation in AES to increase for several reasons. 
For example, Customs is planning to introduce the Automated Export
System Post-Departure Authorized Special System (AES-PASS), which is
a program designed to encourage participation in AES by allowing
qualified exporters\9 to submit a minimal amount of information prior
to export--generally an exporter identification number and a
reference number for the shipment.  Further, Census is terminating
AERP and hopes current users will switch to AES.  Customs also
anticipates an increase in participation since AES first came online
in July 1997 for exports via air, truck, and rail. 


--------------------
\9 The exporter or the freight forwarder (on behalf of the exporter)
will be allowed to apply for the program, and government agencies
with export requirements will be responsible for authorizing an
applicant's participation.  These agencies will review all
applications according to criteria and standards established by their
respective agencies.  Approval to participate in AES-PASS must be
unanimous. 


         MOST COMPANIES SURVEYED
         DO NOT HAVE IMMEDIATE
         PLANS TO GET ON AES
-------------------------------------------------------- Letter :3.1.1

Despite Customs' expectations of increased participation, most
companies we surveyed do not have immediate plans to use AES.  We
surveyed 400 randomly selected freight forwarder companies and 80
exporters that file paper SEDs.  As shown in figures 1 and 2, only
about 36 percent of freight forwarders and about 32 percent of
exporters we spoke with currently plan to use AES to submit their SED
information; only 50 percent and 42 percent of those companies,
respectively, reported that they plan to get on AES within the next 3
years.  Of the companies that plan to use AES, only 4 percent of the
freight forwarders and 5 percent of the exporters have filed a notice
with Customs\10 that they plan to participate in AES or are testing
AES.  (See fig.  3.) In addition, more than half the companies we
surveyed that plan to use AES do not know when they will use it. 
Most companies did not know how much it will cost their company to
implement AES and were not familiar with the AES-PASS program (see
fig.  4). 

   Figure 1:  Freight Forwarders'
   Plans Regarding AES

   (See figure in printed
   edition.)

Note:  To present a complete composite of ocean freight forwarders'
posture toward AES, figure 1 includes the responses from two survey
questions:  question 9, which asked whether or not companies had
heard of AES, and question 12, which asked those companies who had
heard of AES about their plans regarding its use. 

\a Companies whose "plans are not known" include companies that were
undecided about whether to use AES and company officials we
interviewed who did not know their companies' plans regarding AES. 

   Figure 2:  Exporters' Plans
   Regarding AES

   (See figure in printed
   edition.)

Note:  To present a complete composite of the total exporter survey
population's posture toward AES, figure 1 includes the responses from
two survey questions:  question 9, which asked whether or not
companies had heard of AES, and question 12, which asked those
companies who had heard of AES about their plans regarding its use. 

\a Companies whose "plans are not known" include companies that were
undecided about whether to use AES and company officials we
interviewed who did not know their companies' plans regarding AES. 

   Figure 3:  Freight Forwarders'
   and Exporters' Status of
   Involvement With AES (Those
   Planning to Use AES)

   (See figure in printed
   edition.)

Note:  Figure applies to the 117 freight forwarders and 20 exporters
that plan to use AES. 

\a Includes companies that filed a notice or letter of intent with
Customs that they intend to get on AES, and companies that are
testing AES. 

\b Comments from companies that responded that their status of
involvement in AES was "other" included those that were in the
process of developing AES software or not at that time actively
pursuing participating in AES because of a lack of information on the
system. 

   Figure 4:  Freight Forwarders'
   and Exporters' Knowledge of AES
   (Those Planning to Use AES)

   (See figure in printed
   edition.)

Note:  Figure applies to the 117 freight forwarders and 20 exporters
that plan to use AES. 


--------------------
\10 In order to participate in AES, companies must file a notice, or
letter of intent, with Customs. 


         BENEFITS CITED BY THE
         TRADE COMMUNITY DO NOT
         OUTWEIGH CONCERNS
-------------------------------------------------------- Letter :3.1.2

Companies and industry groups we spoke with cited certain benefits to
getting on AES.  The primary incentive mentioned was automating their
export system.  About 50 percent of the companies we surveyed that
plan to use AES said that automation was an incentive to use AES. 
The other benefit voiced by over 15 percent of respondents was the
potential for a single filing point for all export data, referred to
as "one-stop filing" by Customs.  In addition, those companies we
interviewed that are already using AES said that they were doing so
to reduce paperwork and personnel and associated administrative
costs, take advantage of new automated initiatives, and participate
in the development of AES. 

While the cost of automation and lack of knowledge regarding AES were
cited as possible impediments to AES participation by the export
community,\11

predeparture filing emerged as a key concern among some segments of
this group.  Our work indicates that whether or not predeparture
filing posed a problem for businesses was related to the type of
export or mode of transportation used to export. 

According to industry groups and several companies' officials we
interviewed, filing information predeparture is inconsistent with
their business practices.  These officials told us that they often do
not know the precise volume and value of their final shipment until
just before departure, which makes it difficult to file their
paperwork on time.  Predeparture filing was a particular concern for
exporters of bulk goods or grain commodities.  Some of these
companies said that they would have to enter estimates in AES prior
to departure and that the estimates would then have to be revised
later, thereby resulting in rework.  One exporter described this as
having to do "twice the work."

Regarding carriers, all of the eight airlines and air couriers we
spoke with said that meeting the predeparture filing requirement
would present a problem for their current business operations; six
said that they would not participate in AES due to this requirement. 
While the air couriers we interviewed said that they generally have
the SEDs in hand prior to departure, because of the fast-paced nature
of the air courier business they are unable to input SED data into
AES before the aircraft departs.  Representatives from both these
industries told us that they anticipate having to input data into AES
as a service to exporters and freight forwarders. 

Representatives from companies participating in Customs' evaluation
of AES, conducted in two vessel ports in 1996 before AES was expanded
to all ports, indicated problems with predeparture filing. 
Representatives from some companies stated that 80 percent of the
time they have all the information needed to complete the SED prior
to departure of the vessel.  However, for the remaining 20 percent of
the time, they have difficulty in obtaining and providing
predeparture data.  In addition, the evaluation did not include
airlines and air couriers, which have significant concerns regarding
predeparture filing.  It also did not include exporters of bulk
commodities that have similar concerns. 

Export industry groups also have repeatedly expressed concerns about
the AES-PASS program, which allows exporters to file most of their
export information postdeparture, but still requires companies to
file some data prior to departure.  Specifically, they have stated
that AES-PASS will be costly and burdensome to exporters without
providing much benefit to the government.  For example, in a March
1997 letter to the Commissioner of Customs, a group of large
exporters stated that AES-PASS requires exporters to bear a
predeparture reporting burden for all shipments while doing "nothing
to improve data collection or compliance." They stated that AES-PASS
would require two submissions for a single shipment--both pre- and
post-departure--resulting in additional programming of automated
processes.  In June 1997, the Trade Resource Group, the private
advisory group to Customs on AES, expressed similar concerns in a
letter to the Commissioner of Customs. 

In response to the export community's continued dissatisfaction with
AES requirements (particularly filing information predeparture via
AES and/or AES-PASS), in June 1997 the Commissioner of Customs
proposed that industry groups enter into formal negotiations with
Customs to resolve issues of disagreement regarding AES.  Customs has
used such negotiations, which rely on an outside moderator, to
resolve issues with the trade community in the past.  According to
Customs officials, this approach will provide a forum for the trade
community and Customs to discuss, and potentially resolve, the
outstanding issues of concern to both parties.  Customs officials
told us that they are uncertain as to when the negotiations will
begin. 


--------------------
\11 Cost of automation was cited by 28 percent of freight forwarders
not planning to use AES as a reason for not doing so.  None of the
exporters we surveyed cited cost as a reason for not getting on AES. 
(Current AES participants reported that their costs ranged from a $50
setup fee and $2.50 per AES transmission fee for filers using the
Internet to $250,000 for a company undergoing a complete conversion
from manual to automated filing.) Lack of knowledge about AES was
cited by 11 percent of freight forwarders and 36 percent of exporters
we surveyed who did not plan to use AES. 


      AES' ABILITY TO ENHANCE
      ENFORCEMENT IS CURRENTLY
      LIMITED
---------------------------------------------------------- Letter :3.2

AES is designed to provide a "smart targeting system" that would
allow inspectors to focus their attention on possible illegal
shipments among the thousands of exports leaving the country each
day.  For example, AES is designed to compile exporter histories,
allow for trend analysis, and provide a prioritized list of targets
for selective enforcement actions.\12 However, AES' ability to meet
this objective is limited by four major factors.  First, AES does not
currently link with other law enforcement databases, such as those
maintained by the Treasury; the Federal Bureau of Investigation; and
the National Insurance Crime Bureau (which maintains a database on
stolen vehicles).  Customs inspectors told us that AES would be a
more effective enforcement tool if it linked with these databases,
allowing inspectors to obtain information more quickly on exporters
with prior export violations or on stolen vehicles that may be
exported.  Customs officials told us that they are considering trying
to have AES link with other enforcement databases in the future, but
that at present they have no definitive plans to do so.  They noted
that very few enforcement or administrative databases are directly
linked to each other because of logistics, funding, and security
concerns. 

Second, AES-PASS will not provide adequate information to target
shipments because it only requires a minimal amount of data prior to
departure--the exporter identification number; a reference number for
the shipment; and a few additional data elements, such as the license
code and number if the export requires a license.  It will not
provide the detailed commodity data that inspectors told us they need
for better enforcement.  Because so little predeparture information
is provided on AES-PASS, some inspectors we interviewed were
concerned that AES-PASS would undermine any advantage that AES would
have provided, for example, ready access to more detailed commodity
data predeparture. 

Third, AES allows SED information to be transmitted only hours before
a shipment's departure (as with the current paper system), and
inspectors told us that in most cases this is not sufficient time for
targeting possible illegal shipments.  While some inspectors told us
that they would need SED information about 4 hours in advance of the
carrier's departure in order to target shipments, others said they
would need 24 hours.  (We did not evaluate the feasibility of
companies being able to file data within these time frames.) Finally,
since participation in AES is voluntary, an illegal exporter is
unlikely to use the system for filing export data.  Inspectors at
several ports told us that there is no incentive for exporters to get
on AES, and others stated that they believe AES would need to be
mandatory to be effective. 


--------------------
\12 Currently, inspectors only receive general information on the
exports in paper documents provided by the carrier.  (Because SEDs
are held by the carrier and not Customs inspectors, if an inspector
wants to review an SED, he usually must request that the carrier fax
it to the Customs port.) In addition to this information, inspectors
also target shipments based on information provided from informants
and other law enforcement agencies, and their own "gut instinct"
based on years of experience. 


      MOST OTHER AGENCIES' EXPORT
      REQUIREMENTS CANNOT BE FULLY
      SATISFIED BY AES
---------------------------------------------------------- Letter :3.3

According to both Census and Customs, AES has the potential to
provide exporters with "one-stop shopping" by creating a single
electronic filing center for all U.S.  export data.  AES was not
designed to replace any agency's authority to regulate exports.  The
system was designed to serve as a source of export data for agencies
with export requirements and to reduce redundancies in filing and
paperwork associated with various export control requirements. 
However, AES is unlikely to achieve those objectives because most
agencies' export requirements cannot be fully satisfied through AES. 
For example, 8 of the 13 agencies\13 identified by Customs as having
regulatory authority over exports\14 are not using AES to fulfill
their export licensing or permit requirements because of existing
regulations that require them to retain their own licensing
procedures, including collecting information provided by the
exporter.  As a result, exporters will have to continue to apply to
multiple agencies for approval to export certain commodities.  (About
30 percent of all U.S.  freight forwarders export goods that require
export licenses.) For example, exporters seeking to ship products
that have both civilian and military applications would still have to
apply directly to the Commerce Department's Bureau of Export
Administration for approval.  In addition, exporters of chemicals and
pharmaceutical drugs are required to apply directly to the Drug
Enforcement Administration (DEA) 15 days prior to exportation in
order for DEA to conduct an investigation. 

Although AES is designed to eliminate the paper SED, it will not
substantially reduce or eliminate agency paperwork or the electronic
filing associated with the issuance of export licenses, certificates,
or permits.  For example, the Department of Agriculture's Food Safety
and Inspection Service issues inspection certificates for
agricultural exports that must accompany the shipment abroad,
precluding the possibility of electronic filing through AES.  In
addition, DEA officials noted that they are governed by international
conventions, to which the United States is a signatory, that mandate
their use of internationally standardized paper licenses for exports
of certain chemicals and pharmaceutical products. 

According to Customs officials, there are several obstacles that
prevent them from quickly achieving this goal.  For example, many
agencies lack sufficient staff or budgetary resources, have outdated
regulations that may need to be changed, and/or are reluctant to
share data with other agencies even though they may collect the same
data.  Customs recognizes that these obstacles will need to be
overcome in order to have AES fully interface with other
export-related agencies. 

Despite these limitations, officials at three agencies with export
reporting (rather than licensing) requirements--Census, the Maritime
Administration, and the Energy Information Administration--indicate
that AES has the potential to satisfy their needs.  Specifically, it
is expected to eliminate their paperwork processing and help them
fulfill their reporting requirements.  Several other agencies,
including the Bureau of Export Administration, the Office of Foreign
Assets Control, and the Office of Arms Control and Nonproliferation
indicated that as designed, AES would provide a more efficient means
to track and monitor cargo shipments against approved licenses. 
Currently, AES validates cargo against State Department and Bureau of
Export Administration licenses. 


--------------------
\13 For two other agencies, other factors preclude their full use of
AES:  the Department of Energy's Office of Nonproliferation and
National Security has export license review functions that will not
be fully satisfied though AES; and the Environmental Protection
Agency cannot use AES for other reasons (see p.  21). 

\14 In 1994, Customs established a group, consisting of 13 U.S. 
government agencies, to determine each agency's procedures for
regulating exports and the information each agency collects from the
export community to fulfill its statutory and regulatory
requirements. 


         SLOW PROGRESS IN
         DEVELOPING INTERFACES
         WITH OTHER AGENCIES
-------------------------------------------------------- Letter :3.3.1

Since 1994, Customs has tried to develop an automated interface with
other government agencies to maximize opportunities to share export
information through AES and streamline data collection.  After
Customs determines the feasibility of working with a particular
agency, Customs seeks to (1) reach commitments to collaborate on
their use of AES, (2) define and incorporate the informational
requirements of these prospective users, and (3) conclude the process
with a memorandum of understanding (MOU) that guides the
implementation of the final interface. 

Currently, only Census has signed an MOU with Customs legally
stipulating each agency's responsibility for collecting,
transmitting, and securing data captured in AES, in addition to
cost-sharing arrangements.  Customs has obtained written commitments
from five other agencies to collaborate on AES.  Some of the areas
being discussed are data to be included in AES, information sharing
and access, and the development of compatible information systems.\15

Of these six agencies, Customs has completed and incorporated into
AES the user requirements of Census, the Bureau of Export
Administration, and the Office of Defense Trade Controls, specifying
each agency's requirements for collecting and processing data.  One
reason that progress has been slow is that Customs has assigned only
one full-time person to develop interfaces with other government
agencies. 

Other agencies have not committed to use AES for a variety of
reasons.  For example, according to officials at the Bureau of
Alcohol, Tobacco, and Firearms, the agency lacks sufficient resources
to develop a compatible automated system, and DEA has regulations
that preclude its use of AES.  Officials at the Environmental
Protection Agency told us that they cannot use AES, as they do not
currently have an agreement with Census to access SED data. 
Furthermore, although AES does include Nuclear Regulatory Commission
license codes in order to validate licensed shipments, agency
officials indicated that the agency already has an automated
information system that meets its needs.  (See fig.  5.)

   Figure 5:  Status of Customs'
   Agreements With Export-Related
   Agencies

   (See figure in printed
   edition.)

It has been well documented that successful information systems
require the continuing involvement and commitment of senior
executives.\16 In this case, where the concept of AES entails
integrating the export reporting functions of 14 separate federal
agencies, extensive high-level coordination and exchange are not
presently in place to explicitly define what export reporting and/or
licensing requirements can or cannot be accommodated by AES and what
distinct licensing and/or reporting requirements must remain. 


--------------------
\15 None of the agencies that issue licenses, certificates, or
permits have agreed to transfer their license application data
through AES. 

\16 See Executive Guide:  Improving Mission Performance Through
Strategic Information Management and Technology (GAO/AIMD-94-115, May
1994). 


   CONCLUSIONS
------------------------------------------------------------ Letter :4

The quality of export data has been a long-term problem.  AES
represents a major initiative to improve the quality of export data
that is used to negotiate trade agreements and enforce export laws
and regulations.  While the trade community believes export data
needs to be automated, the reluctance of U.S.  companies to
participate and the uncertainty that other agencies will be able to
interface with AES raise serious questions about the system's
viability.  In addition, Customs' planned use of AES as an
enforcement tool is limited because AES is not currently linked to
other law enforcement databases, and AES-PASS allows approved
exporters to file almost all of their export data post-departure.  We
question whether AES will be able to meet its objectives without
greater involvement of top management in resolving the operational
and implementation problems we have identified.  We believe the
Commissioner of the U.S.  Customs Service and the Director of the
U.S.  Census Bureau need to devote sustained management attention to
AES.  Specifically, these officials need to expeditiously assess the
extent to which the export community's concerns can be addressed, the
likely amount of participation in AES, the likely usefulness of AES
in enhancing enforcement, and the extent to which other agencies will
be able to use AES.  In making this assessment, attention needs to be
given to determining whether

  -- predeparture filing of export data is critical to improved
     export statistics and enforcement of U.S.  laws and regulations
     and, if so, how far in advance inspectors need the information
     for AES to be an effective enforcement tool;

  -- a link between AES and the databases of law enforcement agencies
     can be achieved;

  -- allowing some exporters to file SEDs after departure would
     undermine the objective of achieving improved export data and/or
     render AES ineffective as an enforcement tool; and

  -- the requirements of other agencies can be modified or otherwise
     accommodated to permit their use of AES. 

Once this assessment is done, we believe the agencies need to
consider how or whether to proceed with implementing AES.  If these
problems are not resolved in the near future, we are concerned that
Customs will continue to invest significant monies in a system that
is likely to be of limited benefit. 


   RECOMMENDATIONS
------------------------------------------------------------ Letter :5

We recommend that the Secretaries of the Treasury and of Commerce
direct the Commissioner of the U.S.  Customs Service and the Director
of the Bureau of the Census to delineate the concrete actions needed
to improve AES' potential, and, after doing so, assess the costs and
benefits of continuing to implement AES. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :6

A draft of this report was provided to Customs and Census.  While
Customs agreed that AES should interface with other enforcement and
export data bases and that AES-PASS should be reevaluated in light of
its potential adverse effect on enforcement efforts, both agencies
said that they believed our assessment of the level of participation
in AES was premature.  They said that early in the system's
development, they decided to use a phased implementation approach. 
They also noted that participation in AES has increased since it was
expanded to all modes of transportation in July 1997 and they expect
participation to be greater in the future.  However, they did not
address our recommendation or specify the actions they plan to take
to overcome obstacles to AES' success. 

We disagree with Census and Customs view that our assessment of AES
is premature.  We believe our work provides important insights into
issues that will affect AES' success and that Customs and Census need
to develop a strategy to address these issues.  On the critical issue
of participation, our survey revealed strong resistance among the
export community that has serious implications for future
participation.  Unless AES achieves high participation and provides
an interface among agencies with enforcement and export
responsibilities, it is difficult to envision how the system can meet
its objectives.  We, therefore, continue to believe that Customs and
Census should identify the specific actions needed to improve AES'
potential, and, after doing so, assess the cost and benefits of
continuing to implement AES. 

Census also expressed concern that the results from our surveys and
interviews were not presented in such a way that the reader can
determine the significance of the responses and that our work does
not reflect the views of the entire export trade community.  We used
a variety of techniques to obtain the export community's views
regarding AES, including a nationally representative sample survey of
400 ocean freight forwarders.  Our survey was necessarily limited to
ocean freight forwarders because AES had not been extended to other
modes of transportation.  We believe that the results from this
survey when combined with those from our survey of the top 80
exporters that file paper SEDs, as well as in-depth interviews with
30 exporting companies and 14 of the top AERP filers, provides a
reasonable basis on which to assess the views of a broad cross
section of the export community regarding AES.  We did not suggest
that our assessment was based on a survey of the entire export
community.  Moreover, Census did not offer any studies that produced
results that were inconsistent with what we found.  (See app.  IV for
specific details on our scope and methodology.)


---------------------------------------------------------- Letter :6.1

As agreed with your office, unless you publically announce the
contents earlier, we plan no further distribution of this report
until 2 days after its issue date.  At that time, we will provide
copies of the report to appropriate congressional committees and the
Commissioner of the U.S.  Customs Service and the Director of the
Bureau of the Census.  We will also make copies available to other
interested parties on request. 

This review was done under the direction of JayEtta Z.  Hecker,
Associate Director.  If you or your staff have any questions
concerning this report, please contact Ms.  Hecker at (202) 512-8984. 
Major contributors to this report are listed in appendix VII. 

Sincerely yours,

Benjamin F.  Nelson,
Director, International Relations
 and Trade Issues


SIX OTHER COUNTRIES' EXPORT DATA
COLLECTION SYSTEMS
=========================================================== Appendix I

We obtained information from officials in six countries--Australia,
Canada, Japan, South Korea, Mexico, and the United Kingdom--on their
export procedures and systems for collecting export data.  (The
information we collected from these countries was self-reported by
these countries--we did not independently verify any information we
obtained.) Almost all of these countries reporting having implemented
automated systems for collecting export data, and most countries
reported that nearly 100 percent of their export data is collected
via their automated systems.  Most countries' automated systems are
voluntary and were automated within the last 5 years.  These
countries require that at least some export information be filed
prior to departure.  Further, almost all of the countries use their
automated system in some way as a targeting tool to help with the
control and enforcement of the countries' export laws. 

Five of the six countries from which we obtained information have
implemented an automated system to collect their export data
(Australia, Japan, South Korea, Mexico, and the United Kingdom);
Canada is currently piloting an automated system to collect export
data (see
table I.1).  With the exception of Mexico,\1 all countries' automated
systems are voluntary (including Canada's new system).  As an
alternative to electronic filing, exporters in Australia, Canada,
Japan, and the United Kingdom can file paper export declarations. 
Most of these countries' automated systems were implemented in the
1990s, although Australia and Japan have had at least a partially
automated system in place since the mid- to late-1980s. 

Most countries require that exporters or their agents file at least
some information prior to departure.  Japan requires that all export
data be submitted prior to departure.  Australia, Canada, South
Korea, Mexico, and the United Kingdom, however, allow approved
exporters to wait to file some of their information after departure. 
Australia requires that approved exporters file an export report as
soon as the information is available; Canada requires that exporters
file a report up to 5 days after the month of departure; South Korea
requires that a report be filed within a day of departure; Mexico
generally requires that a report be filed within a week of departure;
and the United Kingdom requires exporters to file a completed report
within 14 days of departure. 

All of the countries, with the exception of the United Kingdom, use
their automated system to help control exports and target goods for
inspection (Canada plans to use its system for this purpose).  Some
countries, such as South Korea and Japan, use pre-set criteria for
targeting goods for inspection.  Mexico's automated system, on the
other hand, randomly selects shipments for inspection. 



                                    Table I.1
                     
                     Characteristics of Six Countries' Export
                             Data Collection Systems

Characteristic
s               Australia             Canada                Japan
--------------  --------------------  --------------------  --------------------
Method of data  Paper and voluntary   Paper and currently   Paper and voluntary
collection      automated system      piloting voluntary    automated system
(paper or                             automated system
automated
system)

Year of         1988                  Not available         Air cargo: 1985; sea
automation                                                  cargo: 1992

Percent of      98 percent            Not available         Over 90 percent
export data
captured via
system or
percent of
export
community
using
automated
system

Time frame for  Predeparture, with    (1) Paper: air,       Predeparture
data filing     optional post-        truck, rail -
                departure filing for  predeparture; ocean
                approved exporters    -pre-and post-
                (this option          departure;
                generally open only   (2) Automated
                to certain bulk and   system:
                agricultural          predeparture, with
                shippers, but         optional post-
                requires some         departure filing for
                information be filed  approved exporters
                predeparture)

Required        \a                    \b                    \c
export
information

Use of          Yes, system used to   Yes, system's plans   Yes, system used to
automated       monitor exporters'    include targeting     help select goods
system for      compliance and as     shipments for         that should be
targeting       targeting tool        examination           examined or
                                                            inspected

Deadline for    No Customs            None                  None
goods to be at  requirement for
port            goods



Annual value    (1995-96)             (1996)                (1995)
of exports      $59 billion           $202 billion          $443 billion
--------------------------------------------------------------------------------

Characteristic
s               South Korea           Mexico                United Kingdom
--------------  --------------------  --------------------  --------------------
Method of data  Voluntary automated   Mandated automated    Paper and voluntary
collection      system                system                automated system
(paper or
automated
system)

Year of         1996                  1992-93               1992
automation

Percent of      Almost 100 percent    100 percent           20 percent
export data     (export community
captured via    participation level)
system or
percent of
export
community
using
automated
system

Time frame for  Predeparture, with    Predeparture, with    Predeparture, with
data filing     optional post-        optional post-        optional post-
                departure reporting   departure filing      departure filing for
                for bulk cargo and    program (requires     approved exporters
                perishable goods      that invoice          (this option is open
                (requires some        information still be  to both paper and
                information be        presented upon        electronic filers
                provided prior to     export)               but requires some
                departure)                                  information be filed
                                                            predeparture)

Required        \d                    \e                    \f
export
information

Use of          Yes, system designed  Yes, system randomly  No
automated       for targeting;        selects shipments
system for      improvements being    for inspection, as
targeting?      made to targeting     well as inspectors
                ability               to examine shipments

Deadline for    Goods required at     Shipments must        Deadlines set
goods to be at  warehouse 24 hrs.     arrive to Customs     locally at Customs
port            before loading        within 24 hours       ports
                                      after automated
                                      system validates
                                      transaction

Annual value    (1996)                (1996)                (1996)
of exports      $130 billion          $96 billion           $113 billion\ g
--------------------------------------------------------------------------------
\a (Australia) Reference number; type of export, establishment code;
owner's name; owner's phone number; consignee's name; consignee's
city; country of destination; port of loading; port of discharge;
invoice currency; total free-on-board (FOB) value; intended date of
export; number of F.C.L.  containers, if applicable; mode of export;
ship/aircraft identity; number of packages; commodity classification
code; origin code; goods description; net quantity; gross weight;
container type; coal, thermal use indicator; assay details; container
number and seal number, if applicable; permit details, including
permit number and encryption code; information on whether goods are
subject to certain export concession arrangements, FOB value; and
signature. 

\b (Canada) Information required for paper filing generally includes
exporter name and address; consignee name and address; exporter's
business number; country of final destination; province and country
of origin of goods; export permit number; description of goods;
harmonized tariff system code of goods; quantity and unit of measure;
value; signature of responsible party; mode of transportation; and
reason for export.  Goods exported by sea can be reported in a
predeparture interim report that must include the following
information:  exporter name, address, and business number; consignee
name and address; country/province of origin of goods, country of
final destination; number of packages; description of goods; and, if
containerized, container number. 

\c (Japan) User code; exporter code, name, and address; trading
pattern code; airway bill or bill of lading number; description,
number, quantity, value, and statistical code numbers of goods;
destination and its code number; loading and storage place code;
airline code, or name and nationality of vessel; and scheduled
departure date. 

\d (South Korea) Forty-four items, including declarant; manufacturer;
exporter name; buyer; value/quantity of goods; destination;
consignee; letter of credit number; and weight. 

\e (Mexico) Sixty-two data items, including information on the
exporting company and its location; goods' quantity, value, and
classification; transport company name and location; and data of the
foreign trade transaction.  Invoice information that must be provided
upon export in paper form must contain the name of the exporting
company; taxpayer identification number; date and number of the
invoice; a general description, quantity, and value of the goods;
information on the vehicle transporting the merchandise; number of
the consolidated entry; name and signature; and number and license of
the Customs broker. 

\f (United Kingdom) Filing requirements depend on the procedure being
used, but a completed declaration via the automated system generally
includes consignor/exporter; number of items declared; total
packages; reference number; name and address of person or company
making the declaration; code for country of ultimate destination;
information on shipment container, if appropriate; identify and
nationality of active means of transport crossing the border; mode of
transport at the border; place of loading; location of goods;
packages and description of goods, including marks and numbers,
container numbers, and number and kind of goods; item number; tariff
classification commodity code; net weight; any additional
information, documents produced, certificates and authorizations; and
value of goods.  Participants in the U.K.  automated system
post-departure filing program generally must file the following
information predeparture:  name and address of exporter or agent;
brief commercial description of goods; number and kind of
packages/goods; marks and numbers on packages; net weight; and any
additional information, documents produced, certificates, and
authorizations. 

\g Excludes exports to other members of the European Community. 

Source:  Data provided by Australia, Canada, Japan, South Korea,
Mexico, and the United Kingdom. 


--------------------
\1 Mexico requires that exporters use agents, called Customs brokers,
to file export data on exporters' behalf.  The Customs brokers must
submit all export data electronically. 


THE AUTOMATED EXPORT SYSTEM
PROCESS
========================================================== Appendix II

As currently implemented, the Automated Export System (AES) allows
exporters or their agents to electronically transmit Shipper's Export
Declaration (SED) information directly to Customs.  The process
begins when either the exporter or agent transmits commodity data
directly into AES or when the carrier transmits a receipt of goods
message.  (AES participants can transmit their commodity data either
by developing their own software, using software from various
AES-certified software vendors, via the Internet, or using facilities
of a port authority or service center.) If the carrier transmits data
via AES before the exporter, an "I owe you" (IOU) is established
noting that the exporter has not yet transmitted commodity data.  The
commodity data passes through built-in edits that check for accurate
and complete information and match it against U.S.  agency
requirement files.  The system also matches commodity data sent by
the exporter with transportation data (such as the name and flag of
the vessel) sent by the carrier.  The carrier is free to load the
cargo unless it receives a "hold" message.  AES will reject the
shipment if core information, such as the commodity code, country
name, or exporter name is invalid or incomplete.  These "fatal
errors" must be corrected before merchandise is exported.  (See fig. 
II.1.)

   Figure II.1:  The AES Process

   (See figure in printed
   edition.)

Source:  U.S.  Customs Service. 


RESULTS OF FREIGHT FORWARDER AND
EXPORTER SURVEYS
========================================================= Appendix III

Provided in the following section are questions and responses for our
surveys of 400 U.S.  freight forwarders and Non-Vessel Operating
Common Carriers (NVOCC) and 80 U.S.  exporting companies that file
paper SEDs.  All results are reported as percentages, and for each
question, we present the number of respondents answering the
question.  (Certain questions were only to be answered by a subset of
respondents, that is, those possessing a certain characteristic or
giving a particular answer to a previous question.) For questions
requesting a numerical answer (such as the number of employees) we
present descriptive statistics, such as the median and the range of
responses.  In addition, for several questions where we report in the
letter on the subset of respondents who plan to use AES, we provide
results both for this subset group and for all respondents. 



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


OBJECTIVES, SCOPE, AND METHODOLOGY
========================================================== Appendix IV

To determine whether AES is likely to achieve its objectives of
improving export data, enhancing enforcement efforts, and
streamlining export data collection, we interviewed Customs and
Census headquarters officials and representatives of 12 government
agencies with export-related responsibilities.  We also visited 13
Customs ports, including air, sea, and land border ports, where we
observed export processing and enforcement operations and where we
interviewed numerous supervisory and line inspectors involved in
these operations.  We conducted interviews with over 30 potential
users of AES, including 12 ocean and air carriers and all AES
participants as of April 1997.  We also interviewed over 10 of the
top 16 AERP users in terms of value and volume of AERP filers.  We
also met with several trade groups representing various segments of
the export community.  In addition, we analyzed Customs' and Census'
AES planning documents and Customs' strategic plans regarding its
process for checking goods to be exported.  We also reviewed data
provided by both Customs and Census regarding their actual and
projected costs for AES development.  We did not independently verify
the validity of their cost estimates. 

As part of our effort to determine the trade community's plans for
using AES, we conducted two surveys of potential AES users--U.S. 
ocean freight forwarders and exporters.  A detailed summary of our
methodology for these two surveys follows. 


      FREIGHT FORWARDER SURVEY
      METHODOLOGY
------------------------------------------------------ Appendix IV:0.1


         STUDY POPULATION AND
         SAMPLE
---------------------------------------------------- Appendix IV:0.1.1

The freight forwarder study population consisted of active licensed
ocean freight forwarders and NVOCCs listed in the Federal Maritime
Commission's December 1996 Regulated Persons Index.  The 1,939
freight forwarder headquarters and 2,341 NVOCC listings were merged
and duplicates were eliminated, resulting in a total population of
3,209.  A simple random sample of 400 cases was selected from the
combined list. 

Twelve cases, although listed in the index, had not or were not
currently providing freight forwarding or NVOCC services and
therefore were considered ineligible for the survey.  An additional
six companies were found to be subsidiaries of others on our list. 
In these instances, a single respondent was chosen to respond on
behalf of both companies. 

We sent certified letters to companies we were unable to contact by
phone.  We received confirmation from the Postal Service that three
of those cases were not located at the listed address, nor did the
Postal Service have forwarding address information for those cases. 
The bonds and tariffs of two cases were cancelled by the Federal
Maritime Commission.  The population was adjusted to reflect these
inactive cases.  Applying the same adjustment to the sample resulted
in a final sample size of 376.\1


--------------------
\1 One additional case was removed from the sample because its
representative was a participant in discussions of AES with GAO in
another capacity. 


         DATA COLLECTION
---------------------------------------------------- Appendix IV:0.1.2

Telephone interviews were completed with 331\2 freight forwarders and
NVOCCs, for a response rate of 88 percent.  Forty-five sample members
either refused to participate (30), could not be scheduled for an
interview during the study's time frame (4), or could not be
contacted to confirm eligibility (11). 


--------------------
\2 Includes two cases for which the data were lost due to computer
malfunction. 


         SAMPLING AND NONSAMPLING
         ERROR
---------------------------------------------------- Appendix IV:0.1.3

Because this study is based on a probability sample, our estimates
involve some statistical uncertainty.  Percentages and other
estimates contained in the report are the midpoints of the 95-percent
confidence intervals for the value being estimated.  The results
present intervals for items quoted in the letter. 

To minimize nonsampling sources of error, such as question wording or
sequencing effects and interviewer differences, the survey was
pretested with 16 active freight forwarders and NVOCCs following
intensive interviewer training and practice.  The item nonresponse
rate (the rate of interviewers not recording an answer to a question
that should have been answered) for reported items ranged between 0
and 2 percent for questions asked of all respondents and between 0
and 5 percent for questions asked only of those not planning to use
AES. 


         NONRESPONSE ANALYSIS
---------------------------------------------------- Appendix IV:0.1.4

We examined the Federal Maritime Commission's database to determine
whether systematic differences held between our sample and the parent
population as well as whether systematic differences distinguished
nonrespondents from our respondents.  We examined each group in terms
of number of branch offices, as an approximate measure of size, the
mixture of cases from the freight forwarder or NVOCC listings, and
the region of the country in which they operated.  All nonrespondents
are listed as having single offices, and about 6 percent of
respondents are listed as having two or more offices.  Respondents
and nonrespondents alike were equally divided between the freight
forwarder and NVOCC source listings.  No difference was found between
respondents and nonrespondents in terms of their geographic location
nor between the sample and its parent population. 


         SUMMARY OF RESPONDENT
         CHARACTERISTICS
---------------------------------------------------- Appendix IV:0.1.5

The freight forwarders we interviewed are predominantly small
companies.  The great majority (94 percent) have single offices and
few employees.  Nearly one-half have 5 or fewer full-time employees,
and
74 percent have fewer than 15.  Collectively, our respondents employ
a total of about 87,000 people, and they have a total of 490 office
locations.  Their home offices are located in 25 states and Puerto
Rico.  They served an estimated 132,000 clients during 1996 and had
gross revenues of about $4.6 billion. 

We did not attempt to verify the accuracy of information, such as the
cost of implementing AES, supplied by businesses during our
interviews and surveys. 


      EXPORTER SURVEY METHODOLOGY
------------------------------------------------------ Appendix IV:0.2


         STUDY POPULATION
---------------------------------------------------- Appendix IV:0.2.1

The study population for the exporter survey consisted of the
companies responsible for the greatest number of paper SEDs and/or
those of the highest value filed with the Census Bureau in September
1996.  Collectively, these companies filed or had their agents file
34,340 SEDs for exports worth $3.9 billion.  The number of SEDS filed
by individual filers ranged from 2 to 2,293, and the value of goods
exported ranged from about $2.4 million to about $492 million. 

We obtained from the Census Bureau the names of the top 49 filers in
terms of volume of SEDs filed and the top 49 filers in terms of the
value of SEDs filed in September 1996.  The two lists were combined
and purged of duplicates.  In addition, foreign embassies and U.S. 
foreign military sales units were removed from the list.  The
resulting list contained 80 filers located in 22 states.  During the
course of the study, we learned that for some companies, a single
individual was responsible for one or more additional filers. 
Multiple cases for a single respondent were combined into a single
case, leaving a final study population of 72 filers.  Sixty-three of
these companies responded to the survey, a response rate of 88
percent.  Responding companies accounted for 92 percent of the SEDs
filed by the total study population and 88 percent of their total
value. 


         NONRESPONSE ANALYSIS
---------------------------------------------------- Appendix IV:0.2.2

To determine whether systematic differences distinguished
nonrespondents from respondents, the two groups were compared in
terms of the value and number of SEDs filed as well as their
geographic location.  Independent sample t-tests of the means of SED
value and volume revealed no difference between the groups on either
dimension. 

Because the distributions of these variables were nonnormal, a second
test, which grouped cases according to whether they fell in the top
or bottom half of each distribution, was performed.  The comparison
revealed no difference between the two groups.  The geographic
distribution of nonrespondents also paralleled that of respondents. 
Item nonresponse for reported items ranged from 0 percent to 2
percent for questions asked of all respondents in this survey and
from 0 to 5 percent for questions asked only of those planning to use
AES. 

We did our work between November 1996 and August 1997 in Washington,
D.C., and in various Customs port locations across the United States,
in accordance with generally accepted government auditing standards. 




(See figure in printed edition.)Appendix V
COMMENTS FROM THE DEPARTMENT OF
COMMERCE
========================================================== Appendix IV



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)

See comment 1. 

Now on pp.  2, 5, and 6. 

See pp.  20-21. 



(See figure in printed edition.)

Now on pp.  5 and 7. 

See comment 2. 

See comment 3. 

Now on pp.  16-17. 

See comment 4. 



(See figure in printed edition.)

Now on p.  13. 

See pp.  20-21. 


The following are GAO's comments on the Department of Commerce's
letter dated October 30, 1997. 

GAO COMMENTS

1.  Our draft report stated that predeparture filing is a major
concern for some segments of the export community, including certain
industry groups, airlines and air couriers, and companies that export
bulk goods or grain commodities.  However, we do not state that most
companies we interviewed cited the requirement for predeparture
filing as the main reason for not participating in AES. 

We note that nearly 40 percent of both the freight forwarders and
exporters we surveyed reported that they have at least "some" to
"very great" difficulty filing SEDs predeparture.  In addition, about
40 percent of both groups said they filed SEDs late in 1996.  Among
companies that reported little or no difficulty filing SEDs
predeparture, 28 percent of all ocean freight forwarders and 19
percent of exporters we surveyed said they filed SEDs late in 1996. 

2.  In commenting on the percent of companies that report only to
Census and Customs, Census did not take into account those companies
that reported having licensing requirements.  Therefore, the
statistics they cite are inaccurate.  About 61 percent of all ocean
freight forwarders, and, of the exporters we surveyed, 32 percent
have no license or other agency reporting requirements. 

3.  We note in our report that AES was designed to serve as a source
of export data for agencies with export requirements and reduce
redundancies in filing and paperwork associated with various export
control requirements.  This paperwork includes license application
data.  We revised the text to make clear that AES was not designed to
replace an agency's authority to regulate exports. 

4.  We do not suggest that paper filers need to automate their
procedures in order to file via AES.  Rather, our report lists
various options available to companies that want to convert from
filing paper SEDs to filing via AES
(see p.  4). 




(See figure in printed edition.)Appendix VI
COMMENTS FROM THE U.S.  CUSTOMS
SERVICE
========================================================== Appendix IV

See comment 1. 



(See figure in printed edition.)

Now on p.  4. 

See pp.  20-21. 



(See figure in printed edition.)

Now on pp.  3-4
and 14. 



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)

See comment 2. 

Now on p.  16. 



(See figure in printed edition.)


The following are GAO's comments on the U.S.  Customs Service's
letter dated October 27, 1997. 

GAO COMMENTS

1.  Our report does not state that the majority of the trade
community supports full automation or that a majority recognizes the
benefit of one-stop filing.  Instead, our survey shows that 45
percent of all ocean freight forwarders and 55 percent of the
exporters we surveyed cited the convenience of automation as an
incentive to use AES.  Only 16 percent of ocean freight forwarders
and 18 percent of the exporters we surveyed cited one-stop filing as
an incentive to use AES.  Similarly, our work does not validate that
80 percent of the information required predeparture is in fact
available predeparture.  We note in our report that representatives
from some companies participating in Customs' 1996 evaluation of AES
stated that they believe that 80 percent of the time they have
information needed to complete the SED prior to predeparture of the
vessel.  We did not attempt to determine whether this was an
universal view among companies in the exporting community. 
Conversely, our surveys show that nearly 40 percent of both the
freight forwarders and exporters we surveyed reported that they have
at least some to very great difficulty filing SEDs predeparture. 
About 40 percent of both groups said they filed SEDs late in 1996. 

2.  We note in our report that AES was designed to serve as a source
of export data for agencies with various export control requirements
and to reduce redundancies in filing and paperwork.  This paperwork
includes license application data.  However, we also state that AES
is unlikely to achieve its objective of providing exporters with
"one-stop shopping" because most agencies' export requirements cannot
be fully satisfied through AES.  We also note that AES will not
reduce or eliminate agency paperwork or the electronic filing
associated with the issuance of export licenses, certificates, or
permits.  However, we have revised the text to make clear that AES
was not designed to replace an agency's authority to regulate
exports. 


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

NATIONAL SECURITY AND
INTERNATIONAL AFFAIRS DIVISION,
WASHINGTON, D.C. 

Elizabeth J.  Sirois, Assistant Director
Elizabeth Morrison, Evaluator-in-Charge
Carolyn Black-Bagdoyan, Senior Evaluator
Sara B.  Denman, Senior Evaluator
Mary M.  Park, Evaluator
Olivia L.  Parker, Evaluator
Rona H.  Mendelsohn, Senior Communications Analyst
Waverly E.  Sykes, Assistant Director
Marilyn C.  Mauch, Assistant Director
Kathleen M.  Joyce, Senior Social Science Analyst
Arthur L.  James, Jr., Senior Statistician

OFFICE OF THE GENERAL COUNSEL

Mark C.  Speight, Senior Attorney

LOS ANGELES REGIONAL OFFICE

Daniel R.  Garcia, Senior Evaluator
Edward J.  Laughlin, Senior Evaluator
Larry S.  Thomas, Senior Evaluator


*** End of document. ***