COMMODITY FLOW SURVEY CFS

1
The Commodity Flow Survey and Hazardous
Materials Safety Data Perspectives on Use,
Content, and Needs for the Future
Transportation Research Board 80th Annual
Meeting Washington, D.C. January 8, 2001
2
Introduction to 1997 Commodity Flow Survey
Methodology and Results
John Fowler U.S. Census Bureau Service Sector
Statistics Division Commodity Flow Survey Branch
3
Commodity Flow Survey(CFS)

• Conducted through a major partnership between
• U.S. Census Bureau
• U.S. Department of Commerce
• Bureau of Transportation Statistics (BTS)
• U.S. Department of Transportation

4
Commodity Flow Survey

• Component of Economic Census
• Planned for 5-year intervals
• Measures movement of goods in the U.S.

5
CFS Sampling

• Three stages of sampling

• Establishments

• Weeks of the year

• Outbound shipments

6
CFS Sampling

• Establishments
• Sample survey of 100,000 employer establishments
  in

• Mining
• Manufacturing
• Wholesale

• Also includes warehouse locations of retail and
  other multi-establishment companies

7
CFS Sampling (establishments cont.)

• Selected from Census Bureau Business Register
• Sample of Establishments Stratified By

• Industry
• Geography
• Firm Size
• Type of Organization

8
CFS Sampling

• CFS Industry Coverage Excludes
• Crude Petroleum and natural gas extraction
• Farms
• Service Industries
• Government Establishments
• Imports (until shipment reaches 1st domestic
  shipper)
• Trans-border Shipments

9
CFS Sampling

• Sample of Weeks of the Year
• 4 one-week reporting periods assigned to each
  establishment
• 1 reporting period in each calendar quarter
• Sample covers all 52 weeks of the year

10
CFS Sampling

• Sample of Outbound Shipments
• For each reporting period respondents report for
  a sample of outbound shipments
• Sample of shipments is selected by respondents
  using instructions provided
• Average number of shipments reported 25
• Maximum reported set at 40

11
CFS Data Collection

• For each sampled shipment, respondents report
• Origin, destination
• Mode of transportation
• Commodity (5-digit SCTG)
• Value of shipment
• Weight of shipment

12
CFS Data Collection cont.

• Export Information
• export mode
• foreign destination
• Hazardous Material Information
• four digit UN/NA (United Nations/North American
  Code)

13
CFS Modes of Transportation

• - Parcel, delivery, courier, or U.S. Postal
  Service
• - Private truck
• - For-Hire truck
• - Railroad
• - Shallow draft vessel
• - Deep draft vessel
• - Great Lakes
• - Pipeline
• - Air
• - Other
• - Unknown

14
Additional Published Mode Levels

• - All Modes
• - Single Modes
• - Truck Includes for-hire, private, and
  combinations
• - Water Includes shallow draft, deep draft, and
  Great Lakes
• - Multiple Modes
• examples Parcel, delivery, courier, or U.S.
  Postal Service, Truck Rail,
• Rail Water

15
CFS Commodity Classification

• 1993 CFS
• Standard Transportation Commodity Classification
  (STCC)
• Approximately 1,200 codes at the 5-digit level
• Developed by the Association of American
  Railroads for railroad use

16
CFS Commodity Classification

• 1997 CFS
• Standard Classification of Transported Goods
  (SCTG)
• Developed jointly by BTS, Census Bureau, and
  Statistics Canada
• Based on Harmonized System (HS) coding
• Approximately 500 codes at the 5-digit level
• Significantly reduced CFS respondent coding
  problems while maintaining meaningful commodity
  detail

17
Geography Levels of CFS Data

• National
• Census Regions and Divisions
• States
• Top 60 Metropolitan Areas

18
Census Regions and Divisions

• Regions
• Northeast
• Midwest
• South
• West

• Divisions
• New England
• Middle Atlantic
• East North Central
• West North Central
• South Atlantic
• East South Central
• West South Central
• Mountain
• Pacific

19
CFS Produces Summary Statisticson Key Shipment
Characteristics

• Value, Tons, Ton Miles, Average Miles per
  Shipment

• Commodity Shipped
• Modes of Transportation
• Origin/Destination Flows

20
CFS Data

• Available
• Print
• CD-ROM
• Internet

21
Improvements Planned for 2002 CFS

• Enhanced Internet Capabilities
• Automated SCTG commodity coding tool
• On-line assistance Frequently Asked Questions,
  email, etc.

22
Improvements Planned for 2002 CFS cont.

• Data Product Improvements
• Additional and Expanded File Formats
• Public Use File with CFS Micro-data

23
CFS Public Use File

• Detailed (shipment level?) CFS Data

Balanced against Census legal responsibility to
protect confidentiality

• Trade-offs being considered
• Commodity vs. Geography vs. Mode of
  Transportation detail

24
Contact Information

• John Fowler
• Chief, Commodity Flow Survey Branch
• Census Bureau
• [phone]
• john.l.fowler_at_census.gov

Census website www.census.gov
25
1997 CFS Hazardous Material Statistics

Ronald Duych Bureau of Transportation Statistics
26
1997 CFS Hazardous Material Statistics

CFS Hazardous Material (Hazmat) Data

• Needed by policy planners
• Addresses public safety concerns
• Expands the body of knowledge

27
Hazardous Materials

• 1993 CFS
• Only a Yes/No Hazmat Indicator for each
  Shipment
• Resulted in Limited Published Data on Hazmat
• Shipment characteristics for
• 3 commodities (one All other)
• 5 modes (one All other)

28
Hazardous Materials

• 1997 CFS
• Using the 4-digit UN/NA Code Produced a Wealth of
  New Hazmat Data
• PIH Hazard Class
  Detailed Modes
• 20 Selected UN/NA Codes Selected SCTG Codes
• Hazard Division Packing Group 1 Intra and
  Interstate
• Major States origin destination

29
CFS Hazmat

• Step 1
• Identified Need for Improved CFS Hazmat Data
• BTS/DOT Public Safety Issues and Policy
• Census Bureau Improved Data Collection
• Industry Groups / Risk Analysis

30
CFS Hazmat

• Step 2
• Improving CFS Hazmat Data
• Proposal to Collect 4-digit UN/NA Code
• Readily Available to Respondents
• Key Building Block for Detailed Hazmat Data
• Comparability With Other Data Sources

31
CFS Hazmat

• Step 3
• Implementing Collection of UN/NA Code
• Replaced yes/no Indicator Used in 1993
• Updated Forms, Instructions Editing
• No Additional Coding Manual Needed
• Hazmat Shippers Already Familiar with UN/NA
  Terminology and Codes
• (Important to the Successful Implementation)

32
1997 CFS Hazmat Data

• Value
• Tons
• Ton Miles
• Average Miles per Shipment
• Modes of Transportation
• State of Origin Destination
• Hazard Class/Division/UN number
• Special Groupings

33
Major Contributions of the 1997 CFS Hazmat Data

• Truck data
• largest volume
• greatest exposure to the public
• Air data
• importance of even limited data
• shipment data importance

34
Special Groupings

• Air Hazmat Shipments by Value
• PIH
• Packing Group 1
• Export
• Inter. vs. Intrastate Truck Shipments

35
Tons of Hazardous Materialsby Mode
All other 1.5
Pipeline 27.6
Truck 55.6
Water 9.1
Rail 6.2
36
Ton-Miles of Hazardous Materials by Mode
Truck 28.4
All other 17.4
Water 25.9
Rail 28.3
37
Value of 1997 CFS Hazmat Data

• Exposure data
• Denominators for incident data
• Allow for the development of ratios

38
Uses of the 1997 CFS Hazmat Data

• Conducting risk assessments
• Industry safety programs
• Development of transportation safety policies and
  programs

39
Examples of 1997 CFS Hazmat Data Uses
40
Examples of 1997 CFS Hazmat Data Uses
41
Types of Hazmat Tables Produced From the 1997
CFS

• Hazard Class and Division Tables - 14
• Selected Hazardous Materials by specific UN s -
  3
• Mode specific tables - 12
• State/geographic tables - 3
• Total 1997 CFS Hazmat Tables - 32

42
Summary

• Expansion of the Hazmat Data Resulted From
• Understanding Data Needs
• Identifying Opportunities for Improvement
• Successful Implementation
• Teamwork government and private industry
• Use of existing survey

43
Contact Numbers and E-mail Addresses

• Email Ronald.Duych_at_bts.gov
• Phone [phone]
• Product Distribution

BTS website www.bts.gov/btsprod
Census website www.census.gov
44
Measuring Commodity Flows

• William C. Davie Jr.
• U.S. Census Bureau
• Service Sector Statistics Division
• Program Research Development Branch

45
Topics

• History of the Commodity Transportation Survey
  (CTS) and the Commodity Flow Survey (CFS)
• Transportation Data Needs
• Defining the Target Population
• Options for Collecting Shipment Data

46
History of the CTS 1963, 1967, 1972

• Conducted in 1963, 67, 72, 77, 83
• Used similar survey designs for 63, 67, and 72
• Covered establishments in selected manufacturing
  industries in the 48 contiguous States
• Requested respondents to sample shipping
  documents for preceding year
• Used straight-line distances

47
History of the CTS1977 CTS

• Improvements changes made for the 77 CTS
• Covered ALL manufacturing establishments in the
  U.S.
• Collected shipment value to allow the estimates
  to be adjusted to the results of the 1977 Census
  of Manufactures
• Requested respondents to sample shipping
  documents for preceding year (mail-out/mail-back)

48
History of the CTS1983 CTS

• Major changes made for the 83 CTS
• Expanded coverage to include selected mining
  industries and grain wholesalers
• Requested summary data (no respondent sampling)
• Did not collect commodity nor shipment distance
  info
• Did not publish results due to large non-sampling
  error

49
History of the CFS 1993 CFS

• Improvements/changes made for the 93 CFS
• Expanded coverage - mining, manufacturing,
  wholesale, selected retail and service industries
• Increased sample size to nearly 200,000
  establishments
• Estimated shipment distances using Oak Ridge
  National Laboratorys transportation network

50
History of the CFS 1997 CFS

• Improvements/changes made for the 97 CFS
• More efficient sample design
• Reduced sample size and respondent burden
• Improved edits
• 1999 ASA paper (Black, Davie, Jonas)

51
Transportation Data Needs

• Identify major transportation centers and
  corridors (volume flows)
• Measure shipment flows into, out of, within, and
  through major geographic areas of the U.S.
• Develop standardized measures across time.

52
Transportation Data Needs cont.

• Evaluate the effect of Federal and State
  regulations
• Determine whether user charges are adequate and
  equitable
• Produce data for safety/risk assessment
  (denominator data)

53
Defining the Target Population

• Industry
• Geography
• Commodity

• Mode
• Shipments
• Reference Period

54
Survey Unit - Primary Sampling Unit (PSU)

• Shippers (Origin)
• Carriers (En-route)
• Businesses/Households (Destination)

55
Shipper-Based Survey

• Advantages
• Private truck activity is captured
• Shipment characteristics known (e.g., commodity,
  value, weight)
• Ultimate origin and ultimate destination known

56
Shipper-Based Survey

• Disadvantages
• Carrier models and routes may vary from models
  sometimes only initial mode known
• Potentially more expensive to canvass due to the
  size of the universe of shipping establishments
• Doesnt capture import movements from the border
  to initial domestic location excludes
  pass-through shipments (e.g. Mexico to Canada)

57
Carrier-Based Survey

• Advantages
• Potentially less expensive to canvass because the
  universe is smaller than that of shipping
  establishments
• Able to capture imports as soon as they enter the
  country
• Routes well known could ask carrier for mileage

58
Carrier-Based Survey

• Disadvantages
• Private truck not captured
• Shipment characteristics (commodity, value) are
  often unknown
• Difficult to identify multi-mode shipments
• Ultimate origin and ultimate destination often
  not known

59
Reporting Shipment Data

• Three Possibilities
• Respondent provides summary information
• Respondent selects a sample of shipping documents
• Respondent provides all shipments electronically
  (all items or just shipment value?)