Sources and Reliability Statement
INTRODUCTION
Since 1993 the Bureau of Transportation Statistics
(BTS) at the United States (U.S.) Department of Transportation (DOT) has
contracted with Bureau of the Census (Census) at the U.S. Department of
Commerce (DOC) to provide previously unpublished surface transportation data
(other than air or maritime vessel) for U.S. import and export trade with
Canada and Mexico.� This dataset is referred
to as the Transborder Freight Data.
Under the contract, Census provides two sets of data tables to BTS; one
provides detailed transportation flows while the other is commodity based
without as much transportation detail.
A number of changes to improve the quality and usefulness of the data
have occurred since the April 1993 dataset was first made available.� This document discusses the limitations and
overall reliability of the dataset, details the information provided in the
individual data fields, and explains changes and corrections since 1993.
SOURCES
The Transborder Freight Dataset is extracted
from the Census Foreign Trade Statistics Program.� Import and export data are captured from administrative records
required by the Departments of Commerce and Treasury.� Historically, these data were obtained from import and export
paper documents that the U.S. Customs Service (Customs) collected at a port of
entry or exit.� However, an increasing
amount of import and export statistical information is now being captured electronically.� The Automated Broker Interface System (ABI)
collects approximately 98 percent of the value of all U.S. imports
electronically.� On the export side
approximately 80 percent of the value of all U.S. exports is collected
electronically (of that, 35 percent is collected through the U.S./Canada Data
Exchange
[1]
and 45 percent though the Automated Export System (AES)).� The data in the Transborder Surface Freight
Dataset are captured from these same administrative records.�
COVERAGE
This dataset incorporates all statistical shipments
entering or exiting the U.S. by surface modes of transport (other than air or
maritime vessel) to and from Canada or Mexico.
Prior to January 1997 statistics, this dataset also included
transhipments
[2]
in its
detailed tables, that is shipments entering or exiting the United States by way
of Customs ports on the northern or southern borders even when the actual
origin or final destination of the goods was other than Canada or Mexico.� In other Census trade statistics, transhipments
through Canada and Mexico are credited to the true country of origin or final
destination.
Therefore, to make this dataset more comparable to
other Census trade statistics (such as the “U.S. Exports of Merchandise” and
“U.S. Imports of Merchandise” both on CD-Rom and the “FT920: U.S. Merchandise
Trade: Selected Highlights”), detailed information on transshipments has been
removed.� A summary of transhipments is
now presented by country, direction of trade, and mode of transportation for
truck, rail, and all other (mail, pipeline, other, and foreign trade zone
(FTZ)).�
Shipments that neither originate nor terminate in
the United States (i.e., intransits) are beyond the scope of this dataset
because they are not considered U.S. international trade shipments.� (For a complete explanation of the
international trade statistics, refer to the Guide to Foreign Trade
Statistics available from the Government Printing Office and on the Census
Bureau’s web site at http://www.census.gov/foreign-trade/www/ftd.stat.guide.html.)
ROLE OF CENSUS
Census assembles, performs a complex series of
analytical reviews, compiles and publishes the U.S. foreign trade
statistics.�� The Transborder Surface
Freight Data are a subset of these statistics.
For this dataset, Census reports information on the surface method of
transportation, which are not reported separately in any other U.S. trade
statistics.� Census also conducts
analytical reviews for consistency purposes on the transportation related data
(i.e., mode of transportation and port of entry/exit).
RELIABILITY
In general, the reliability of U.S. foreign trade
statistics is very good.� As was noted
in previous sections, the Transborder Surface Freight Data are a subset of
these statistics.� Users should be aware
that trade data fields (such as value, commodity classification) are typically
more rigorously reviewed than transportation data fields (i.e., mode of
transportation and port of entry/exit).
Users should also be aware that the use of foreign trade data to describe
physical transportation flows might not be direct.� For example, this dataset provides surface transportation
information for individual Customs districts and ports on the northern and
southern borders. However, because of filing procedures for trade documents,
these ports may or may not reflect where goods physically crossed the
border.� This is because the filer of
information may choose to file trade documents at one port while shipments
actually enter or exit at another port.
Despite these limitations, however, the Transborder Surface Freight
Dataset is the best publicly available approximation for analyzing transborder
transportation flows.
Users should also note that the Transborder Surface
Freight Dataset represents Census’ first attempt to disaggregate the various
surface modes of transportation in U.S. foreign trade statistics.� Since the dataset was first made available
in April 1993, it has gone through several refinements and improvements.� When improbabilities and inconsistencies
were found in the dataset, extensive analytical reviews were conducted, and
improvements were made to the dataset based on these reviews.�� Therefore, the overall reliability of the
dataset is generally very good.
However, accuracy does vary by direction of trade and individual data
field.� For example, import data are
generally more accurate than export data.
This is primarily due to the fact that the Customs uses import documents
for enforcement purposes while it performs no similar function for exports.
MAJOR
REPORTING CHANGES
Several significant reporting changes have occurred
since the release of the first data month, April 1993.�� These are noted here, and also discussed in
detail in subsequent sections.
April 1994
Increased commodity detail and geographic detail
began to be reported together in a number of files.
For exports, the 98 Schedule B 2‑digit commodity groups replaced
the previous 11 Schedule B Groups; the U.S. state of origin replaced the U.S.
region of origin; the Canadian province and Mexican state of destination
replaced Canadian and Mexican regions of destination; and the National
Transportation Analysis Region (NTAR) of the U.S. exporter was added.
For imports, the 98 2‑digit Harmonized Tariff Schedule of the
United States (TSUSA) commodity groups replaced the 11 TSUSA groups; the
Canadian province of origin replaced the Canadian region of origin; and the
U.S. state of destination replaced the U.S. region of destination. (Note: At
the 2-digit level the commodities of Schedule B and TSUSA are the same.)
April 1995
Shipping weight for Mexican imports was
added.� For U.S. exports to Canada, the
cost of moving goods from the place of direct shipment in the U.S. to the
consignee in Canada (data field, “Freight”) was added.��
The mode of transport,
“foreign trade zones”, (or DISAGMOT 9) was added to the import tables.� This new mode of transportation (MOT)
category was added in recognition of the increased activity in foreign trade
zones along the U.S./Mexican and U.S./Canadian borders.� Although FTZ is being treated as a MOT in
this dataset, the mode of transportation for a specific shipment into or out of
a foreign trade zone is unknown because Customs does not collect this
information.
January 1996
Shipping weight for truck and rail shipments
imported into the United States through Canada or Mexico from a third country
(i.e. transshipments) was added. (Note: Beginning with the January 1997 data
month, transhipment data were removed.
(See January 1997 changes.))
January 1997
Transshipments from a third country through Canada or
Mexico to the U.S. or from the U.S. to a third country through Canada or Mexico
were removed from the dataset.� A
summary of transhipments is now presented by country, direction of trade, and
mode of transportation for truck, rail, and all other (mail, pipeline, other,
and FTZ).� The deletion of these
transhipments will allow for more comparability with other Census trade data
(such as the “U.S. Exports of Merchandise” and “U.S. Imports of Merchandise”
both on CD-Rom and the “FT920: U.S. Merchandise Trade: Selected
Highlights”).�
Additional port detail was
added.� Canadian and Mexican border
customs districts now include all public ports.��� In addition, some additional non-border or inland ports are now
identified separately.� Non-border ports
with low activity are combined at their parent Customs district and reported by
an� “XX” (i.e., “35XX”).
Due to increased geographic specificity, individual
record counts were deleted from the dataset.
A summary of record counts is now presented by country, direction of
trade, and mode of transportation.
January 2003
With the release of
December 2002 statistics on February 20, 2003, the U.S. Census Bureau will
discontinue the state export series based on the exporter location (EL).� Beginning with January 2003 statistics,
published in March 2003, the U.S. Census Bureau will compile and release state
exports based only on the origin of movement (OM) series.
Users should be
cautious in interpreting the exporter location (EL) series and are advised that
comparisons of 2000 and 2001 state totals to those of previous years may be
misleading. The state export series based on the exporter's location, the EL
series has changed significantly since late 1999. An analysis of the locations
reported by exporters in 2000 and 2001 as compared to 1999 has shown that most
of the changes involve shipments reported electronically through the Automated
Export System (AES). AES, a joint effort of the U.S. Customs Service and the
Census Bureau, was first implemented in 1995. The results of their analysis
coincide with the surge in reporting through AES in late 1999 through 2001,
when the former Automated Export Reporting Program (AERP) was discontinued and
alternative ways of filing through AES (AES DIRECT, PC Link, and Web Link) were
introduced.
While AES has
significantly improved the overall quality and coverage of the export data, it
has changed filing practices, especially the addresses reported for multiple
location companies. The exporter's location is based on the ZIP code in the
exporter's address as reported on the Shipper's Export Declaration or its
electronic equivalent. The Census Bureau is currently exploring whether it is
possible to better define the address of each shipment in order to establish a
stable EL state export series for the future.
As a result of these changes
BTS can no longer provide four tables that have been released in prior
years.� Those tables are:
3B Exports
to Mexico with State of Exporter and Commodity Detail,
4B Exports to Canada with State of Exporter and Commodity Detail,
5B Exports to Mexico with NTAR of Exporter and Geographic Detail,
and 6B Exports to Canada
with NTAR of Exporter and Geographic Detail.�