|
||||||||||||||||
|
||||||||||||||||
|
||||||||||||||||
|
| ||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 1
THE NATIONAL ACADEMIES
Advisers lo the Nalion on Science, Engineering, and Medirine
Committee on National Statistics
Mr. Rick Kowalewski
Acting Director
Bureau of Transportation Statistics
United States Department of Transportation
400 7th Street, S.W.
Washington, D.C. 20590
Dear Mr. Kowalewski:
February 23, 2004
500 Fifth Street, NW
Washington, DC 20001
Phone: (202) 334-3096
Fax: (202) 334-3751
We are pleaser! to transmit this letter report from the Committee on the Bureau of
Transportation Statistics International Trade Traffic Study. Convened uncler provisions of
Section 5~ ~5 of the Transportation Equity Act for the 2Ist Century (TEA-21) (PL 105-178), the
Bureau of Transportation Statistics (BTS) requestect the Committee on National Statistics
(CNSTAT) to convene our committee to evaluate the accuracy and reliability of measures of ton-
miles anc] value-miles of international trade traffic carried by highway for each state for use as
formula factors for highway apportionments. (The list of committee members is Attachment A.)
The goals of the study are to review the findings and recommendations of a BTS staff
report prepared in response to PL 105-178 and to explore acIditional data sources and measures
that may be used for this interstate tracle traffic measures. This letter report assesses the validity
of the findings and recommendations in the report prepared in response to Section 5 ~ ~ 5 of TEA-
21 and provides suggestions for areas of investigation and analysis that would improve the report
and assist in interpreting its conclusions. The committee's final report will address data sources,
data collection and estimation methodology, resources, and organizational arrangements for data
collection and will make recommendations for statistical data quality standards.
Since its fornication in August 2003, the committee has conducted two meetings and
commissioned two independent studies. At an initial fact-fin(ling meeting in September 2003,
the committee heard presentations Tom staff of the Bureau of Transportation Statistics and
experts who had contributed to the development of the Bureau's 5 ~ ~ 5 report. Subsequently, the
committee sponsored a workshop in November 2003 at which outside experts and government
officials cliscussed specific aspects of the report. In total, ~ 5 presentations were heard by the
committee in open session. The experts provided advice on the validity and quality of data used
in developing the BTS report, on alternative and emerging means of developing clata on
NATIONALACADEMY OF SCIENCES · NATIONALACADEMY OF ENGINEERING · INSTITUTE OF MEDICINE · NATIONAL RESEARCH COUNCIL
OCR for page 2
international trade traffic data, ant! on matters of access to public and proprietary data sources.
(The list of presenters en c} their topics is Attachment B.)
Report, Bureau of Transportation Statistics
In response to congressional legislation, the BTS has submitted a report to Congress that
presents the methods and findings of a stiffly that (a) estimated the ton-miles anct value-miles of
international tracle traffic carried by highway in each state; (b) evaluates! the accuracy and
reliability of such estimates for use in the formula for highway apportionments; (c) evaluated the
accuracy and reliability of the use of cliese] fuel as a measure of international trade traffic by
state; and (~) iclentified needed improvements in long-term data collection programs to provide
accurate and reliable measures of international traffic for use in the formula for highway
apportionments. ~
There no direct measures of ton-miles and value-miles of international trade traffic
carried by highway in each state. In fact, methods for measuring total volumes of ton-miles and
value-miles passing over state highways are still in development. Thus, ton-miles and value-
miles must be estimated, with data from a multitude of public ant! private sources, none of which
was specifically clesignect for the task. The models used to compute these estimates and the
various data sources have known and unknown errors ant! biases that affect the accuracy and
reliability ofthe estimates.
Ton-miles, clefined as the amount of freight tonnage carried in ~ mile, are derived Tom
separate estimates of tonnage and estimates of miles traveled. Despite growing attention to
measures of tonnage and miles, both of these estimates have many sources of error. Many of the
same limitations apply to value-miles, which are defined as the amount of decIarec! value that
moves ~ mile.
Because the data sources are not designect to directly yield ton-mile and value-mile
measures, the estimates require application of moclel-basect estimation techniques ant! untested
assumptions. The models that were used in the BTS study includes! the Oak Ridge National
Laboratories highway network mo(le] to derive estimates of the distances travelecl by imports ant!
exports. The accuracy of the distance estimates is unknown. Moreover, data on imports were
available only at the state level. The accuracy of the assumptions used to clistnbute state-level
imports to counties within a state is unknown because of a lack of data at the substate level.
The BTS study directed considerable emphasis to assessment of the statistical properties
of the periodic Commodity Flow Survey (CFS), collectecl by the Census Bureau, given the
central role that this survey plays in the development of core data inputs to the ton-miles
estimates. The BTS stiffly concludes that CFS data, used to calculate the tonnage, value, organ
and destinations, and distance traveled of export freight movement by truck, suffer significant
shortcomings for this nuroose. For example, the CFS data do not provide cntical estimates of
. . . . ....
~ ~ ~ ,
. ~ .
. .
Imports; In action, because ot sample coverage, the data exclucle such major sectors as farms,
~ U.S. Department of Transportation, Bureau of Transportation Statistics, Measurement of Ton-Miles and Value-
Miles of International Trade Traffic Carried by Highway for Each State, December 2003.
2
OCR for page 3
fisheries, construction, transportation, and of} and gas extraction companies. The CFS also has
sampling and nonsampling variability issues that have not been fully considered.
The BTS study also evaluated the accuracy and reliability of the use of diesel fuel data as
a measure of international tracle traffic by state. This analysis of the relationship between diesel
fuel use and ton-miles of international freight determined that there is no direct way of using
diesel fuel use as measure of the international tracie traffic by state. The BTS study concludes
that, at present, there are no convenient means of distinguishing international and domestic use
of diesel fuel, ant! the diesel use data are plaguer! with several sources of error; consequently,
estimates of use on a state-by-state basis are quite problematical.
Finally, the BTS study addressee! issues of data coverage, quality, and access. The BTS
report found that, although international trade data are extensive ant! detailed, the administrative
data collected by the Customs and Border Protection ant! processed by the Census Bureau are
limiter! in their usefulness for the purpose of measuring international tracle traffic on state
highways. Critical data elements necessary to identify destinations and transportation modes are
not available in the data. Other data collections, including the U.S. Foreign Waterborne
Transportation Statistics, the Port Port/Export Reporting Service (PIERS), and the
Transborcler Surface Freight Database, provicie important pieces of information, but each suffers
from issues of data quality and compatibility. Furthermore, the BTS study found that many of
the sources of data required to prepare and improve the estimates of international trade traffic by
highway by state are collected under pledges of confidentiality and are protected from other use
by legal mandate. Some of the potential data sources are developed in the private sector ant! are
of unknown quality.
Committee Findings and Conclusions: Summary
In this letter report the committee selectively addresses several major aspects of the BTS
report focusing on initial analysis of the requirements for data quality and reliability associated
with the purpose of the allocation formula. The committee concurs with the BTS report that
available freight data are not of sufficient coverage ant! quality to permit precise determination of
international trade traffic by state. The committee's preliminary review of the availability of
valiclate(1 data for the estimates suggests neetlecl improvements in data collection programs to
enhance the accuracy and reliability of international tracle traffic measures for formula
apportionment purposes.
Data Quality and Reliability
The committee concurs with the central finding that the present state of available
freight data does not permit precise determination of international trade traffic by state.
However, the committee was limited in its ability to evaluate the accuracy and reliability of the
data for use as formula factors because no allocation formula that depends on ton- or value-miles
of international trade traffic has been developed or proposed. Without knowing the formula and
its purpose, it is not possible to assess how accurate estimates of international tracle traffic must
be. If, for example, the central purpose of the formula is to compensate states for wear and tear
on their highway infrastructure, then even highly precise measures of value-miles will bear little
3
OCR for page 4
relationship to the objective of the formula, and ton-miles will only roughly reflect the goal of
the formula. Less accuracy is needled if the formula is insensitive to variations in ton-miles than
if the formula is highly sensitive.
More generally, the accuracy that is needed for formula allocation may clepenc! on
whether the objective is constructing, maintaining, and operating infrastructure; reducing
congestion; minimizing the impact of the highways on the environment; improving the safety of
the network; or any of a number of other worthwhile objectives. When evaluating the quality
and reliability of data, such factors as error, bias, and transparency will each be weighed
differently, depending on the objective of the apportionment formula.
It is instructive to compare the quality of international trade traffic data with that of the
data used in existing fecleral highway apportionment formulas. Each major fecleral highway
program has a legislated formula that relies on data obtained from the joint Fecleral Highway
Administration Highway Performance Monitoring System. The primary measures user! in the
current allocation formulae inclucle lane miles, vehicle miles traveled, diesel fuel usage, state
population, urbanized area population, and nonhighway recreational fuel use. Each of these data
elements has its own error measures, as well as an established validation procedure that is
clocumentec! in guiclelines that were carefully developed in a federal-state cooperative venture.
It is important to note that each of the data elements used in the existing formula
allocation is directly and independently measured. The measures are also transparent: their
sources are well known, and they use carefully monitored procedures for verification. The
current formula allocation factors do not rely on model-based estimation procedures. Thus, each
has a directly measurable error structure in which bias and variance can be specified.
The characteristics of verifiability and transparency now deemed important for
data used as allocation factors do not apply to the estimates computed in the BTS report.
The BTS model-basec! estimates of ton-miles ant! value-miles of international trade traffic have
their genesis in a multiplicity of data sources anct methodologies. As the BTS report
appropriately notes, "limitations in the data have a negative effect on their accuracy and
reliability." Even when it is possible to specify the error associated with one of the data inputs,
such as the Commodity Flow Survey, it is not possible to specify with precision the quality and
reliability of the overall estimates, which reflect the accumulation of the errors of several data
sources. it is also not possible to compute the variability that is introducer! into the estimates by
several of the key methodologies, such as gravity models, substate allocation procedures, ant!
highway network assignment procedures.
The BTS report appropriately addresses the need for separately iclentifying international
tracle traffic data for an allocation formula. However, the committee is concerned that in the long
run, a focus on estimating ton- and value-miles of international trade traffic by state may distract
attention from the larger issues of characterizing sources of highway wear ant} tear and
congestion for purposes of funds allocation. Wear ant! tear, congestion, ant! other concerns are
not a function of whether the traffic is international or domestic; rather, they relate to such
measures as axle loac! factors, vehicle types ant! categories, and weight. It was observed in the
4
OCR for page 5
committee's workshop that pavement doesn't care whether traffic is international or domestic.
Moreover, even total ton-miles are only roughly related to wear and tear ant! other concerns.
It is likely that total ton-miles of all types of freight traffic moving by state highway can
be estimated more accurately than the disaggregated measures of domestic and international
tracle traffic. The committee notes that the Fecleral Highway Administration's Office of Freight
Management and Operations uses a variety of data sources in the clevelopment of the Freight
Analysis Framework (FAF), which estimates (as of 1998) and forecasts (to 2010 and 2020) total
freight tonnage flows on U.S. highways, as well as U.S.-Canadian truck traffic. The base year of
the FAF is the ~ 993 economic census year, for which Commoctity Flow Survey data are also
available. However, the FAF is assembled from existing data sources. Therefore, it is subject to
the same problems of low frequency of acquisition and questionable accuracy as other data
sources.
Part of the problem of accurately estimating international tracle traffic stems from issues
of ctisaggregation. We offer two illustrations, on disaggregating diesel fuel use ant! on
computing substate estimates.
Disaggregating diesel fuel use into domestic ant! international components. The FITS
report assesses the accuracy and reliability of diesel fuel data as a measure of ton-miles of
international tracle traffic on state highways. As notec! above, diesel fuel use is aireacly a factor
in the apportionment of funds for the National Highway System program, but that measure has
several weaknesses when pressed to serve as data for allocation purposes. It is difficult to
estimate the number of gallons consumed for road use in each state because fuel may be
purchased in one state and used in another, ant! there are differences in state reporting of motor
fuel sales. Procedures used to adjust sales data to use data (by means of reports submitted by
carriers uncler the International Fuel Tax Agreement) acid variability to the estimates. Other
factors further reduce the reliability of these estimates so that fuel figures are not as reliable as
mileage ciata. Nonetheless, the diesel fuel data are currently user! as a factor in the
apportionment formula. The BTS report states that these estimates "inherently capture
international trade traffic."
Several computations are necessary to separate the international from the domestic diesel
fuel use. The BTS report uses a regression analysis of the relationship between the estimate of
total ton-miles by truck by state and diesel fuel use by state to produce a formula for predicting
ton-miles based on diesel fuel use. The report finds that the regression relationships for many
states are quite weak, suggesting that "(liege] fuel usage estimates do not provide good predictors
of total trade ton-miles on a state-by-state basis." The committee is concerned that the next step-
-to break clown the weak total relationships into domestic and international components--wouic!
require comparing estimates of international trade traffic ton-miles from the BTS stucly with
estimates of total ton-miles by state from other sources. The committee concurs with the BTS
report that the process of disaggregating diesel fuel use into its domestic and international
trade traffic components wouIc! compound variability in that it would apply a ratio of two
estimates with significant variability with state diesel fuel use data that have known quality
problems.
s
OCR for page 6
Computing substate estimates of import trade traffic flows. Existing data sources provide
estimates of imports flowing to each state, but not of the distribution of imports within states.
The distribution of imports within a state is estimated by assuming that imports to each county
are proportional to annual payrolls computed from data in the ~ 997 County Business Patterns.
The accuracy of this assumption is unknown.
Data Sources
Because there are no direct measures of ton-miles and value-miles of international trade
traffic carried by highways for each state, data from a variety of sources, none of which was
specifically developed for this purpose, have to be used in the estimates. Previous National
Research Council studies have assessed several ofthese data sources.2 ~ its final report, the
committee will address the strengths and weaknesses ot each ot~these data sources in some
detail. In this initial report, the committee presents only a few key observations on one key data
source that was gleaned from presentations at its October 2003 workshop meeting.
... ... .. ,. , .. . . ,` . ~ ,.
The Commodity Flow Survey, undertaken as a partnership between the Census Bureau
and the Bureau of Transportation Statistics, is an invaluable source of information on the how of
goods by highway in the United States. This survey plays an absolutely critical role in the
computation of total miles and international ton-miles and value-miles. Data from the CFS are
used throughout the BTS report. They underpin the computation of total ton-miles and value-
miles and the computation of the export flow estimates for the international trade traffic. The
survey data are also used in the Oak Ridge National Laboratories' hi~hwav network model and
. ... . . , . , . . .. ~ · . . · . · ~
In public and private ctatanases, such as the freight Analysts framework and the Reebie
Associates TRANSSEARCH database.
The (?FLS has many advantages. It nrc~vides data con shipments c~ri~inatin~ from
-------of rip ---a-------- ~--o---~-----o -
manufacturing, mining, wholesale, and selected retail establishments in the 50 states and the
District of Columbia. The survey is conducted every 5 years, as part of the Economic Census
program, and it is mandatory. Respondents report total number of outbound shipments and
information on value, weight, primary commodity, domestic destination or port, airport, or
boarder crossing of exit, foreign destination, and mode of transportation for a sample of
shipments for a 1-week reporting period each quarter. The coverage is extensive: it is estimated
the 2002 CFS gathered information on more than 2.5 million shipments. The establishments are
selected from a frame constructed from the Census Bureau's Business Register.
However, the limitations of the CFS data restrict the accuracy and reliability of the
international trade traffic estimates. Survey coverage omits most retail establishments, services,
transportation, agriculture, government, construction, and oil and gas extraction; thus, overall,
only about 60-70 percent of shipments are covered. Imports are not included until they reach the
first domestic shipper covered by the CFS.
As a stratified random sample survey, the CFS has sampling and nonsampling error
components. An analysis ofthe 1997 response rates indicated an overall nonresponse of ~5
2 For example, see Committee to Review the Bureau of Transportation Statistics (BTS) Survey Programs,
Transportation Research Board Letter Report on the Commodity Flow Survey March 30 2003.
, , ,
6
OCR for page 7
percent, with differences by trade area, size of establishment, and state that count bias the result.
Item nonresponse for shipments was 2.7 percent for entries for value ant! 4.0 percent for entries
for weight. Over the years, sample variability has increased rather than diminished, as the size of
the sample has been reduced from 200,000 establishments in ~ 992 to 50,000 in 2002.
Export activity in the CFS is relatively rare, and the variance on the estimates of value-
miles and ton-miles is relatively high. Exports accounted for about ~ percent of estimated total
value anct 8.2 percent of ton-miles in ~ 997, ant! the domestic part of those shipper! by truck
accounted for 7 percent of the total value shipped by truck and 5 percent of the ton-miles. The
coefficients of variation for the estimates of exports shipped by truck were 5.5 percent for value
and 12.4 percent for ton-miles. These errors pass into the estimates of international trade traffic
ton-miles ant! value-miles carried by trucks on state highways.
Techniques for Improving Data Accuracy and Reliability
The congressional legislation implicitly recognizec! many of the difficulties that wouic! be
encountered in developing measures of ton-miles and value-miles of international tracle traffic
carried on highways in each state, and it directed the BTS to identify long-term clata
improvements to provide accurate and reliable measures for use in highway apportionments.
This section summarizes a few major observations about the potential of current initiatives to
sufficiently improve data to enable its use in formula allocation. The observations in this section
will be amplifies! in the committee's final report.
The BTS report primarily focuses on issues of data coverage, collection, quality, and
access and primarily on the international tracle statistics program. International trade statistics
are collected! by Customs and Borcler Protection for most U.S. international transactions and are
processecl by the Census Bureau's Foreign Trade Division. This database is impressively large,
containing more than 3 million import and 2 million export transactions per month.3 If the
records in this database includes! key data elements, they could be used directly to report
international tracle volumes, destinations, ant! mules. However, the data are largely
administrative in nature. They are provided by shippers, not carriers. The records contain value
data but no weight data; have limited information on geographic location of origin ant!
destination; have uncertain assignment of port geography; and fail to capture all mocles of
transportation. The export statistics are of particularly questionable quality since they receive
little agency scrutiny. Estimates suggest that undercoverage of exports is between 3 and 10
percent. All of these factors tend to limit the utility of the data for apportionment formula
purposes.
The committee observes that there are several fortuitous projects uncler way which, in the
long term, hold promise of filling information gaps. The Custom ant! Border Protection Agency
is streamlining the international transaction data systems uncler the Automated Commercial
Environment (ACE) system, a $1 billion program. This program is linked to a government-wide
program that wouicl meet the needs of the more than ~ 00 agencies for trade clata. Over the next
3 C. Harvey Monk, Jr., U.S. Census Bureau, "Access to Export/Import Trade Data," Presentation to the International
Trade Traffic Study Workshop, November 21, 2003.
7
OCR for page 8
several years, more than $100 million will be invested! in the International Trade Data System
(ITDS) to modernize the data and access to them.4
Among other promising emerging technologies and techniques are projects that would
capture information on weights and vehicle types (weigh-in motion), radio Frequency (RF) tag
technologies that wouIc! identify vehicles ant! cargos while in motion, and advances in traffic
counting technology. Although these cutting-eclge applications hold promise of providing much
of the necessary information in real time for total tonnage on highways, none of the technologies
appears to solve the problem of identifying international trade traffic and associating that traffic
with miles traveled.
The committee will more fully explore promising techniques in our final report. We
hope this assessment of the BTS report is helpful to your work on this important topic.
.
.
Joe] Horowitz, Chair
Committee on the Bureau of Transportation Statistics International Tracle Traffic Stucly
Attachments
4 M.J. Fiocco, U.S. Department of Transportation, "Overview of International Trade Date System (ITDS) &
Automated Commercial Environment (ACE)." Presentation to the International Trade Traffic Study Workshop,
November 21, 2003
8
Representative terms from entire chapter:
trade traffic
