United States Office of
Environmental Protection Research and Development
Agency Washington, DC 20460
EPA-600/R-93-067a
April 1993
* EPA Economic Growth
Analysis System:
Reference Manual
conomic
rowth
nalysis
ystem
Prepared for Office of Air Quality Planning and Standards
-------
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/R-93-067a
April 1993
ECONOMIC GROWTH ANALYSIS SYSTEM:
Reference Manual
FINAL REPORT
Prepared by:
Teresa M. Lynch
Terri Young
Karen N. Johnson
Daniel Bowman
Jill Vitas
Terry Wilson
Ajay Chadha
Lois Alpern
TRC ENVIRONMENTAL CORPORATION
100 Europa Drive, Suite 150
Chapel Hill, North Carolina 27514
EPA Contract 68-D9-0173
Work Assignment 3/302
Project Officer: Sue Kimbrough
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
Prepared for:
U.S. Environmental Protection Agency U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards Office of Research and Development
Research Triangle Park, NC 27711 Washington, DC 20460
-------
ABSTRACT
This report presents the results of work completed under four work assignments under
EPA Contract No. 68-D9-0173. The objective of this report was to describe the development
of a prototype Economic Growth Analysis Systems (E-GAS) modeling system. The
E-GAS model will be used to project emissions inventories of volatile organic compounds,
oxides of nitrogen, and carbon monoxide for ozone nonattainment areas and Regional
Oxidation Model (ROM) modeling regions.
This report details the design and development of the E-GAS system, and includes
detailed descriptions of the workings of the E-GAS computer modeling software, and its
relationships with internal modeling software components, like Regional Economic Models,
Inc. (REMI) models, and external software, like ROM, the Aerometric Information Retrieval
System (AIRS), and the Urban Airshed Model (UAM).
-------
TABLE OF CONTENTS
Chapter Page
ABSTRACT ii
LIST OF TABLES vii
LIST OF FIGURES vii
LIST OF ACRONYMS viii
ACKNOWLEDGEMENTS xi
1 INTRODUCTION 1-1
1.1 Background 1-1
1.2 Objectives 1-2
1.3 Scope of E-GAS 1-2
1.4 Overview of E-GAS System 1-8
1.4.1 General 1-8
1.4.2 EPA Guidance on Projecting Emissions 1-8
1.4.3 Photochemical Modeling Demonstrations Required By The
CAAA 1-9
1.4.4 Emission Inventory for the Northeast Transport Region 1-10
1.4.5 Design Decisions 1-10
1.4.6 E-GAS Design. 1-13
1.4.6.1 Tier 1: The National Economic Tier 1-13
1.4.6.2 Tier 2: The Regional Economic Tier 1-15
1.4.6.3 Tier 3: The Growth Factor Tier 1-19
1.5 References 1-22
2 STATUTORY BACKGROUND AND USER REQUIREMENTS FOR USING E-
GAS TO PROJECT EMISSIONS 2-1
2.1 Introduction 2-1
2.2 Potential E-GAS Users 2-2
2.3 Terminology 2-2
2.4 Overview of Reasonable Further Progress Requirements 2-3
2.5 Overview of Attainment Demonstration Requirements 2-5
2.6 Use of E-GAS 2-7
2.7 System Requirements 2-8
2.7.1 Functional Requirements 2-8
2.7.2 Required System Attributes 2-8
2.7.2.1 Easy Data Entry 2-8
2.7.2.2 User Friendly 2-9
2.7.2.3 Quality Assurance 2-9
2.7.2.4 Data Security 2-9
2.7.2.5 State-owned Data 2-9
2.8 Conclusions 2-9
m
-------
TABLE OF CONTENTS
Chapter Page
2.9 References 2"10
3 PROJECTING EMISSION INVENTORIES: EPA GUIDANCE 3-1
3.1 General 3-1
3.2 Point Sources 3"^
3.2.1 EPA Point Source Projection Guidance 3~2
3.2.2 E-GAS Point Source Growth Factors 3'2
3.3 Area Sources 3~3
3.3.1 EPA Area Source Projection Guidance 3'3
3.3.2 E-GAS Area Source Growth Factors 3~4
3.4 Mobile Sources 3'9
3.4.1 EPA Guidance on Projection of Mobile Sources 3-9
3.4.2 E-GAS Mobile Source Growth Factors 3-11
3.5 EPA Guidance on Projecting Emissions from Utilities 3-12
3.5.1 General 3-12
3.6 References 3-14
4 NATIONAL AND REGIONAL ECONOMIC FORECASTS IN E-GAS 4-1
4.1 National Macroeconomic Models 4-1
4.1.1 Overview 4-1
4.1.2 The Role of National Economic Forecasts in E-GAS 4-2
4.1.3 The Effects of the Choice of National Model on Regional
Forecasts 4-3
4.1.4 National Macroeconomic Forecasts 4-5
4.1.4.1 The REMI U.S. Forecast 4-6
4.1.4.2 Council of Economic Advisors 4-6
4.1.4.3 Data Resources, Inc. (DRI) 4-7
4.1.4.4 Research Seminar in Quantitative Economics (RSQE) .... 4-8
4.1.4.5 Wharton Econometric Forecasting Associates (WEFA) ... 4-8
4.1.5 Forecasting Records of the Models 4-9
4.1.6 Summary 4-11
4.1.7 Conclusions 4-13
4.2 Regional Economic Models 4-14
4.2.1 Overview 4_14
4.2.2 REMI Models 4-15
4.2.3 The Use of REMI Models in E-GAS 4-16
4.3 References 4_lg
5 ESTIMATING FUEL CHOICE IN E-GAS 5_1
5.1 Introduction _
IV
-------
TABLE OF CONTENTS
Chapter Page
5.2 Fuel Consumption Data 5-1
5.2.1 Manufacturing Energy Consumption Survey 5-1
5.2.2 Annual Survey of Manufactures (ASM) 5-2
5.2.3 National Energy Accounts 5-2
5.3 Energy Models Reviewed 5-3
5.3.1 NAPAP Model Set . 5-3
5.3.1.1 Industrial Regional Activity and Energy Demand
(INRAD) Model 5-3
5.3.1.2 Commercial Sector Energy Model by State (CSEMS) .... 5-4
5.3.1.3 Household Model of Energy by State (HOMES) 5-4
5.3.2 REMI Model 5-4
5.3.2.1 Commercial and Industrial Fuel Use 5-4
5.3.2.2 Residential Fuel Consumption 5-5
5.3.2.3 Transportation Fuel Consumption 5-6
5.3.3 PC-Annual Energy Outlook (AEO) Model 5-7
5.3.3.1 Residential Fuel Consumption 5-7
5.3.3.2 Commercial Fuel Consumption 5-7
5.3.3.3 Industrial Fuel Consumption 5-8
5.3.4 ENERGY2020 5-8
5.4 Options Considered for E-GAS Fuel Choice Module 5-9
5.5 The E-GAS Fuel Choice Module 5-11
5.5.1 Modifications Made to CSEMS 5-11
5.5.2 Modifications Made to HOMES 5-12
5.5.3 Modifications Made to INRAD 5-13
5.5.3.1 Modifications to INRAD to Include Fossil Fuel Choice . . 5-13
5.6 References 5-17
6 ESTIMATING PHYSICAL OUTPUT IN E-GAS 6-1
6.1 Physical Output: Definition and Data Sources 6-1
6.2 Forecasting Physical Output 6-10
6.2.1 Forecasting Physical Output Using Employment Data 6-10
6.2.2 Forecasting Physical Output Using Value Added Data 6-13
6.3 Physical Output in E-GAS 6-15
6.3.1 Forecasting 6-15
6.3.2 Sources for Which Physical Output Equations Are Developed .... 6-16
6.4 References 6-17
7 METHODOLOGY USED TO FORECAST VEHICLE MILES TRAVELED IN
E-GAS 7-1
7.1 Introduction 7-1
-------
TABLE OF CONTENTS
Chapter Pa8e
7 1
7.2 Trend-based Approaches
7.2.1 The Highway Performance Monitoring System 7'2
7.2.1.1 HPMS Data Collection 7'2
7.2.1.2 HPMS Analytical Process 7~4
7.3 Trends in VMT Indexes 7'5
7.4 Econometric Approaches '"^
7.5 Methodology Used in E-GAS 7'8
7.6 Results from the New England E-GAS Model 7-H
7.7 References 7~13
8 E-GAS CROSSWALK 8-1
8.1 Overview 8-1
8.2 SICs and SCCs 8-2
8.3 Fossil Fuels 8-3
8.3.1 Residential Fossil Fuels 8-3
8.3.2 Industrial Fossil Fuels 8-4
8.3.3 Commercial Fossil Fuels 8-5
8.3.4 Fossil Fuel Consumption at Utilities 8-5
8.4 VMT Estimates 8-7
8.5 Industry-specific Physical Output 8-7
8.6 Other SCCs 8-8
8.7 CROSSWALK Files 8-8
APPENDIX A CROSSWALK FILES: BLS AND SCC MATCHES A-l
APPENDIX B EXAMPLE CROSSWALK OUTPUT FILES B-l
VI
-------
LIST OF TABLES
Number Page
1-1 Classification Of and Counties Within Designated Ozone
Nonattainment Areas 1-3
1-2 Summer Weekday Emission for 1985 By Source Category for
the U.S. Portion of the ROMNET Domain 1-11
1-3 Policy Variables Included in E-GAS 1-17
3-1 EPA-Preferred Growth Indicators for Projecting Emissions for
Area Source Categories 3-5
3-2 E-GAS Growth Factors for Projecting Area Source Emissions 3-7
3-3 Electric Utility NOX Projections Summary 3-12
4-1 Comparison of BLS and WEFA Aggregate, Employment Forecasts
for the United States, 1995 4-4
4-2 Comparison of BLS and WEFA Aggregate, Employment Forecasts
for the Pittsburgh Region, 1995 4-5
4-3 Summary of Economic Forecasts Surveyed 4-12
6-1 Sample of Physical Output Data Available from the Survey of Current Business . . 6-2
7-1 VMT Equations 7-10
7-2 Results from the New England E-GAS Model 7-12
LIST OF FIGURES
Number Page
1-1 Flowchart for the Economic Growth Analysis System 1-14
7-1 Normalized National Personal Transportation Trends 7-6
7-2 Historic VMT Per Capita 7-7
8-1 CROSSWALK Design 8-10
VII
-------
LIST OF ACRONYMS
AADT
AEO
AIRS
AIRS/AFS
ANL
AQMD
ARGUS
ASM
AUSM
BCM
BEA
BEEM
BLS
CAAA
CBO
CEA
CO
COMMEND
CPI
CSEM
CSEMS
CSTM
DOC
DOE
DRI
DVMT
ECM
E-GAS
EIA
EKMA
EPA
EPRI
EPS
ERP
FMVCP
FGD
FGT
FHWA
FRB
GL
GNP
Average Annual Daily Traffic
Annual Energy Outlook
Aerometric Information Retrieval System
AIRS Facility Subsystem
Argonne National Laboratories
Air Quality Management Division
Argonne Utility Simulation Model
Annual Survey of Manufacturers
Advanced Utility Simulation Model
Build Cost Module
Bureau of Economic Analysis
Building Energy End-Use Model
Bureau of Labor Statistics
Clean Air Act Amendments of 1990
Congressional Budget Office
Council of Economic Advisors
Carbon Monoxide
Commercial End-Use Energy Planning System
Consumer Price Index
Commercial Sector Energy Model
Commercial Sector Energy Model by State
Coal Supply Transportation Model
Department of Commerce
Department of Energy
Data Resources, Inc.
Daily Vehicle Miles Travelled
Emissions and Control Module
Economic Growth Analysis System
Energy Information Administration
Empirical Kinetic Modeling Approach
Environmental Protection Agency
Electric Power Research Institute
Emission Preprocessor System
Economic Report of the President
Federal Motor Vehicle Control Program
Flue Gas Desulfurization
Flue Gas Treatment
Federal Highway Administration
Federal Reserve Board
Generalized Leontief
Gross National Product
vm
-------
GRP Gross Regional Product
HOME Household Model of Energy
HOMES Household Model of Energy by State
HPMS Highway Performance Monitoring System
ICARUS Investigation of Costs and Reliability in Utility Systems Model
I/M Inspection and Maintenance
LNDEPTH Industrial End-Use Planning Methodology-Econometric Models
INRAD Industrial Regional Activity and Energy Demand Model
KLEM Capital, Labor, Energy, and Materials
LPG Liquefied Petroleum Gas
MECS Manufacturing Energy Consumption Survey
MPO Metropolitan Planning Organization
MRMP Multiple Region-Multiple Period
MSA Metropolitan Statistical Area
NAAQS National Ambient Air Quality Standards
NAPAP National Acid Precipitation Assessment Program
NBA National Energy Accounts
NEDS National Emissions Database System
NBECS Non-Residential Building Energy Consumption Survey
NERC North American Electric Reliability Council
NOX Oxides of Nitrogen
NSPS New Source Performance Standard
OAQPS Office of Air Quality Planning and Standards
OMB Office of Management and Budget
PSI Pounds per Square Inch
PURHAPS Purchased Heat and Power Systems
QA Quality Assurance
RACT Reasonably Available Control Technology
REEM Regional Energy End-Use Model
REEPS Regional End-Use Energy Planning System
REMI Regional Economic Models, Inc.
RFP Reasonable Further Progress
ROM Regional Oxidation Model
RSQE Reseach Seminar in Quantitative Economics
RVP Reid Vapor Pressure
SCC Source Classification Code
SEDS State Energy Data System
SIC Standard Industrial Classification
SIP State Implementation Plan
SO2 Sulfur Dioxide
SRMP Single Region-Multiple Period
SRSP Single Region-Single Period
TEEMS Transportation Energy and Emissions Modeling System
TSD Technical Support Division
IX
-------
UAM Urban Airshed Model
DEC Unit Energy Consumption
URGE Universities Research Group on Energy
VMT Vehicle Miles Travelled
VOC Volatile Organic Compounds
WEFA Wharton Econometric Forecast
-------
ACKNOWLEDGEMENTS
This report was written by TRC Environmental Corporation for the U.S. Environmental
Protection Agency's Office of Air and Energy Engineering Research Laboratory (AEERL) and
Office of Air Quality Planning and Standards (OAQPS) under the auspices of the Joint Emissions
Inventory Oversight Group (JEIOG). EPA involvement included: E. Sue Kimbrough (Work
Assignment Manager) and Larry Jones of the AEERL Emissions and Modeling Branch; Richard
Wayland and Keith Baugues of the OAQPS Technical Support Division; Sheila Holman of the
OAQPS Air Quality Management Division; Gale Boyd and Don Hanson of Argonne National
Laboratory; and Steve Piccot of Southern Research Institute. Further significant contributions to
this document have been made by staff from Regional Economic Models, Inc. (REMI), Amherst,
MA, and their documents.
XI
-------
CHAPTER 1
INTRODUCTION
1.1 BACKGROUND
On November 15, 1990, the Clean Air Act Amendments (CAAA) of 1990 were signed
into law. The CAAA require that extreme, severe, serious, and multi-State moderate ozone non-
attainment areas use photochemical grid modeling to demonstrate future attainment with the
ozone national ambient air quality standard (NAAQS) [Section 182(e)(2)(A)]. In addition to
photochemical grid modeling, the CAAA require that moderate, serious, severe, and extreme
ozone non-attainment areas submit reasonable further progress (RFP) inventories demonstrating
a 15 percent reduction in emissions from 1990 to 1996 [Section 182(b)(l)(A)]. In addition, RFP
inventories for serious, severe, and extreme areas must include demonstration of a three percent
annual reduction (averaged over three years) from 1996 until attainment is achieved
[Section 182(c)(2)(B)].
Section 182(b)(l)(A) of the CAAA specifies that the 15 percent reduction from baseline
emissions accounts for any growth in emissions after 1990. A key component of the RFP
inventories and photochemical grid modeling demonstrations will be the development of credible
growth factors for the existing inventories. Credible growth factors will require accurate forecasts
of economic variables and the activities associated with the economic variables. An initial design
and model plan for an economic and activity forecast model has been developed. This modeling
concept, known as the Economic Growth Analysis System (E-GAS) has been described in earlier
reports.1
The existing inventories for RFP demonstration and photochemical modeling will be
housed in the Aerometric Information Retrieval System (AIRS). E-GAS will be applied to AIRS
inventories for the development of estimated future emissions out to 2010, the year that extreme
areas must reach attainment. The photochemical models which will be used to show attainment
include the Regional Oxidant Model (ROM) and the Urban Airshed Model (UAM). ROM
accounts for growth of regional inventories of ozone precursors and models expected levels of
ozone formation and transport in the region. This model provides the expected background (or
transported) concentration of ozone for urban nonattainment areas in the region being simulated.
1-1
-------
With background concentration estimates from ROM, UAM is used to model expected levels of
ozone formation in each ozone nonattainment area for specified meteorological conditions.
Chapter 2 of this report describes State modeling and RFP requirements in greater detail.
1.2 OBJECTIVES
The objective of this report is to describe the development of a prototype E-GAS model.
This report includes an overview of the E-GAS modeling system (Chapter 1); a description of
CAAA requirements for which E-GAS may be used (Chapter 2); a review of EPA guidance on
projecting inventories (Chapter 3); a description of national and regional economic forecasts and
their use in E-GAS (Chapter 4); a discussion of which electric utility model will be included in
E-GAS (Chapter 5); a discussion of the fuel choice module (Chapter 6); a discussion of the
physical output module (Chapter 7); a discussion of the VMT module (Chapter 8); and, a
description of the E-GAS CROSSWALK (Chapter 9). Details on the E-GAS user interface,
minimum hardware requirements, and operation and maintenance of the system can be found in
the E-GAS system documentation.
1.3 SCOPE OF E-GAS
E-GAS will be used to project emissions inventories of volatile organic compounds
(VOC), oxides of nitrogen (NOJ, and carbon monoxide (CO) for ozone nonattainment areas and
ROM modeling regions. Therefore, the final structure of E-GAS includes projection capabilities
for sources of VOC, NOX, and CO for ozone nonattainment areas and any attainment portions of
the States associated with the areas, and States included in the ROM modeling domains.
The nonattainment areas modeled were chosen on the basis of their nonattainment
designation. All serious, severe, and extreme areas were modeled, as were multi-State moderate
areas. A list of these areas, their designations, and the counties included in the areas is presented
in Table 1-1. These areas, their designations and area definitions, were announced in the
November 6, 1991, Federal Register.
To minimize both the number and run time of the models in E-GAS, eight models were
developed. Separate models were developed for EPA Regions 1, 4, 5, 6, 7, and 9. In addition,
1-2
-------
TABLE 1-1. CLASSIFICATIONS OF AND COUNTIES WITHIN
DESIGNATED OZONE NONATTAINMENT AREAS
Designated Area
State(s)
Classification
Counties
Los Angeles-South Coast
Air Basin Area
CA
Extreme
Los Angeles County (part)
Orange County
Riverside County (part)
San Bernardino County (part)
Monterey Bay Area (Monterey,
San Benito, and Santa Cruz
Counties)
Chicago-Gary-Lake County IL-IN
Severe-17
IL: Cook County
Du Page County
Grundy County (part)
Kane County
Kendall County (part)
Lake County
McHenry County
Will County
IN: Lake County
Porter County
Houston-Galveston-Brazoria TX
Severe-17
Brazoria County
Chambers County
Fort Bend County
Galveston County
Harris County
Liberty County
Montgomery County
Waller County
Milwaukee-Racine
WI
Severe-17
Kenosha County
Milwaukee County
Ozaukee County
Racine County
Washington County
Waukesha County
New York- New Jersey-
Long Island
NY-NJ-CT Severe-17 CT: Fairfield County (part)
Litchfield County (part)
NJ: Bergen County
Essex County
Hudson County
Hunterdon County
Middlesex County
Monmouth County
Morris County
Ocean County
Passaic County
Somerset County
(continued)
1-3
-------
TABLE 1-1. CLASSIFICATIONS OF AND COUNTIES WITHIN
DESIGNATED OZONE NONATTAINMENT AREAS (continued)
Designated Area
State(s)
Classification
Counties
New York- New Jersey-
Long Island, continued
NY-NJ-CT Severe-17
NJ: Sussex County
Union County
NY: Bronx County
Kings County
Nassau County
New York County
Orange County
Putnam County
Queens County
Richmond County
Rockland County
Suffolk County
Westchester County
Southeast Desert Modified CA
AQMA
Severe-17
Los Angeles County (part)
Riverside County (part)
San Bernadino County (part)
Baltimore
MD
Severe-15
Anne Arundel County
City of Baltimore
Baltimore County
Carrol County
Harford County
Howard County
Philadelphia-Wilmington- PA-NJ-DE-
Trenton MD
Severe-15
DE: Kent County
New Castle County
MD: Cecil County
NJ: Burlington County
Camden County
Cumberland County
Gloucester County
Mercer County
Salem County
PA: Bucks County
Chester County
Delaware County
Montgomery County
Philadelphia County
San Diego
Ventura Co.
CA
CA
Severe-15
Severe-15
San Diego County
Ventura County
(continued)
1-4
-------
TABLE 1-1. CLASSIFICATIONS OF AND COUNTIES WITHIN
DESIGNATED OZONE NONATTAINMENT AREAS (continued)
Designated Area
State(s)
Classification
Counties
Atlanta
GA
Serious
Cherokee County
Clayton County
Cobb County
Coweta County
De Kalb County
Douglas County
Fayette County
Forsyth County
Fulton County
Gwinnett County
Henry County
Paulding County
Rockdale County
Baton Rouge
LA
Serious
Ascension Parish
East Baton Rouge Parish
Iberville Parish
Livingston Parish
Point Coupee Parish
West Baton Rouge Parish
Beaumont-Port Arthur
TX
Serious
Hardin County
Jefferson County
Orange County
Boston-Lawrence-Worcester MA-NH
(E.MA)
Serious
MA: Barnstable County
Bristol County
Dukes County
Essex County
Middlesex County
Nantucket County
Norfolk County
Plymouth County
Suffolk County
Worcester County
NH: Hillsborough County (part)
Rockingham County (pan)
El Paso
Greater Connecticut
TX
CT
Serious
Serious
El Paso County
Fairfield County (part)
Hartford County
Litchfield County (part)
Middlesex County
New Haven County
New London County
Tolland County
Windham County
(continued)
1-5
-------
TABLE 1-1. CLASSIFICATIONS OF AND COUNTIES WITHIN
DESIGNATED OZONE NONATTAINMENT AREAS (continued)
Designated Area State(s) Classification
Counties
Muskegon
MI
Serious
Muskegon County
Portsmouth-Dover-Rochester NH
Serious
Rockingham County (part)
Strafford County
Providence (all RI)
RI
Serious
Bristol County
Kent County
Newport County
Providence County
Washington County
Sacramento Metro
CA
Serious
El Dorado County (part)
Placer County (part)
Sacramento County
Solano County (part)
Sutler County (part)
Yolo County
San Joaquin Valley
CA
Serious
Fresno County
Kern County
Kings County
Madera County
Merced County
San Joaquin County
Stanislaus County
Tulare County
Sheboygan
WI
Serious
Sheboygan County
Springfield (Western MA) MA
Serious
Berkshire County
Franklin County
Hampden County
Hampshire County
Washington
DC-MD-VA Serious
DC: Entire Area
MD: Calvert County
Charles County
Frederick County
Montgomery County
Prince George's County
VA: City of Alexandria
Arlington County
City of Fairfax
Fairfax County
Falls Church
Loudoun County
(continued)
1-6
-------
TABLE 1-1. CLASSIFICATIONS OF AND COUNTIES WITHIN
DESIGNATED OZONE NONATTAINMENT AREAS (continued)
Designated Area
State(s)
Classification
Counties
Washington, continued
DC-MD-VA Serious
VA: Manassas
Manassas Park
Prince William County
Stafford County
Cincinnati-Hamilton
OH-KY
Moderate
KY: Boone County
Campbell County
Kenton County
OH: Butler County
Clermont County
Hamilton County
Warren County
Huntington-Ashland
WV-KY
Moderate
KY: Boyd County
Greenup County (part)
WV: Cabell County
Wayne County
Louisville
KY-IN
Moderate
IN: Clark County
Floyd County
KY: Bullit County (part)
Jefferson County
Oldham County (part)
St. Louis
MO-IL
Moderate
IL: Madison County
Monroe County
St. Clair County
MO: Franklin County
Jefferson County
St. Charles County
City of St. Louis
St. Louis County (part)
1-7
-------
models which combine the areas in EPA Regions 2 and 3 and EPA Regions 8 and 10 were
developed. Each model includes all extreme, severe, serious, and multi-State moderate areas, as
well as each State and partial State in the region.
1.4 OVERVIEW OF E-GAS SYSTEM
1.4.1 General
Three factors were major considerations during the design phase of E-GAS. First, EPA
guidance on projecting emissions inventories was studied. Second, the role of E-GAS in
projecting inventories for photochemical models was defined. Third, emission inventories used
by Regional Oxidant Model for the Northeast Transport Region (ROMNET) were obtained in
order to determine the largest sources of NOX and VOC in the Northeast. Although the
importance of emission sources to the overall VOC and NOX budgets can vary by geographic
area, the Northeast Transport Region was assumed to provide a good general picture of the
sources which lead to ozone formation. In addition, because of previous modeling efforts for the
region, detailed inventories of VOC and NOX emissions were available.2
1.4.2 EPA Guidance on Projecting Emissions
The general methodology for estimating growth in activity from inventoried emission
sources involves two steps. First, the economic sector which corresponds to the emission-
producing activity is identified. Second, forecasts of growth in the economic sector are used to
project growth in the activity. For example, activity growth at VOC-producing petroleum
refineries may be estimated using growth in Standard Industrial Classification (SIC) 2911. EPA
guidance proposes that economic variables which can be used to project growth in emissions-
producing activity include, in order of preference, product output, value added, earnings, and
employment.3
EPA currently considers economic forecasts from the Bureau of Economic Analysis
(BEA) to be the "preferred data source" for projecting emission source categories.3 This policy
may be modified once E-GAS has undergone its acceptance tests.
1-8
-------
E-GAS is expected to provide more relevant (i.e., physical output) and timely Source
Classification Code (SCC)-specific growth factors than factors based on information available
from BEA. Growth factors based on BEA data reflect employment and earnings growth which
are not as closely related to emissions growth as value added and physical output.
E-GAS provides default, average annual growth factors for ozone nonattainment areas and
for the remainder of the State in which the nonattainment area is located. In the future, E-GAS
growth factors are expected to be very useful to the State and local governments in selecting
growth factors for projecting future VOC emissions. They would provide a starting point for
State and local governments to estimate growth.
Since annual activity growth will fluctuate rather than occur in smooth, year-by-year
increases or decreases, the default factors must be reviewed and modified by State and local
governments based on local knowledge of plant expansions, plant closures, new facility
construction, or similar factors that would be expected to temporarily distort the default growth
projections for any one year.
A detailed discussion of EPA guidance for projecting emissions is presented in Chapter 3.
1.4.3 Photochemical Modeling Demonstrations Required By The CAAA
For photochemical modeling demonstrations, States in the Northeast, Southeast, and
Midwest ROM modeling areas will use estimates from the ROM to determine approximate
background ozone levels from transport of ozone within the region. ROM accounts for growth
in regional inventories of ozone precursors and models expected levels of ozone formation and
transport hi the region. This model provides the expected background (or transported)
concentration of ozone for urban nonattainment areas in the modeled region. These background
concentration estimates will be developed by EPA. In addition, States will need to use a
photochemical grid model, such as the Urban Airshed Model, to estimate ozone formation in the
nonattainment area. UAM uses background concentration estimates from ROM and determines
approximate ozone formation in order to model expected levels of ozone in a nonattainment area
for specified meteorological conditions.
Photochemical modeling and RFP requirements are discussed in detail in Chapter 2.
1-9
-------
1.4.4 Emission Inventory for the Northeast Transport Region
An inventory of VOC and NOX emissions in the Northeast Transport Region in 1985 is
summarized in Table 1-2. As the data in the table indicate, VOC and NOX emission sources
differ greatly by pollutant. Over half of the 1985 NOX emissions in the region are attributable
to point sources, while less than 10 percent of VOC emissions is associated with point sources.
All of the point source NOX emissions are due to fuel combustion. Nonhighway area sources
accounted for over half of the VOC emissions but less than 15 percent of NOX emissions. Only
highway mobile sources are a major source of VOC and NOX: these sources accounted for
approximately 35 percent of NOX and 40 percent of VOC emissions. This inventory does not
include CO; however, CO emissions are associated primarily with sources which also emit VOC
and NOX, namely, fuel combustion.
While two sources of NOX emissions, utility fuel consumption and highway mobile
sources, accounted for over 75 percent of all NOX emissions, only highway mobile sources serve
as a dominant source of VOC emissions. No other VOC source contributed more than 13 percent
of emissions to the inventory.2 This, along with the fact that over half of the VOC was emitted
from area sources, suggests that projecting emissions-producing activity may be more difficult
for VOC than for NOX.
1.4.5 Design Decisions
Based on the information gathered concerning existing EPA projection guidance, the use
of photochemical models in attainment demonstrations, and the 1985 ROMNET inventory, five
major design decisions were made:
1-10
-------
TABLE 1-2. SUMMER WEEKDAY EMISSIONS FOR 1985 BY SOURCE
CATEGORY FOR THE U.S. PORTION OF THE ROMNET
DOMAIN (tons/day)
Point Sources
Fuel Combustion
Utility External - Coal
Utility External - Oil
Utility External - Gas
Utility External - Other
Utility Internal - Oil
Utility Internal - Gas
Industrial External - Oil
Industrial External - Gas
Industrial External - Other
Industrial Internal - Oil
Industrial Internal - Gas
Commercial/Institutional
Aircraft (Internal)
Solvent Metal Cleaning
Printing and Publishing
Dry Cleaning
Automobile Surface Coating
Beverage Can Surface Coating
General Wood Surface Coating
Paper Surface Coating
Miscellaneous Surface Coating
Crude Oil and Gasoline Storage
Bulk Gasoline Storage
Marine Vessel Loading
Service Stations - Stage I
Chemical Manufacture Vents
Chemical Manufacture Fugitives
Petroleum Refinery Fugitives
Refinery Wastewater Treatment
Refinery Vacuum Distillation
Cellulose Acetate Manufacture
Styrene-Butadiene Rubber Mfg.
Polyethylene Manufacture
Vegetable Oil Processing
Paint and Varnish Manufacture
Rubber Tire Manufacture
Carbon Black Manufacture
Coke Oven Byproduct Plants
Other Industrial
Waste Disposal
Total - Point Sources
Highway Mobile Sources
Non-Highway Area Sources
Residential Fuel - Wood
Residential Fuel - Other
Commercial/Institutional Fuel
Industrial Fuel - Coal
Industrial Fuel - Oil
Industrial Fuel - Gas
NOX
5,721
414
176
7
14
8
131
146
6
9
613
62
10
0
0
0
0
7
0
0
0
0
0
0
0
30
0
0
0
19
0
0
13
0
0
0
0
11
425
32
7,851
5,108
7
36
146
111
64
440
voc
23
10
0
1
1
0
12
12
1
0
5
3
6
34
131
0
140
64
30
85
344
67
19
18
1
1
9
16
9
18
30
4
30
2
19
11
2
20
744
3
1,926
8,956
121
2
4
0
3
5
1-11
-------
TABLE 1-2. SUMMER WEEKDAY EMISSIONS FOR 1985 BY SOURCE
CATEGORY FOR THE U.S. PORTION OF THE ROMNET
DOMAIN (tons/day) (continued)
Non-Highway Area Sources, continued
Incineration - Residential
Incineration - Other
Open Burning - Residential
Open Burning - Other
Off-Highway Vehicles
Railroad Locomotives
Aircraft
Vessels - Gasoline
Vessels - Other
Forest Wildfires
Structural Fire
Gasoline Marketed
Degreasing
Drycleaning
Graphics Arts/Printing
Rubber and Plastic Manufacture
Surface Coating
Architectural
Auto Body Repair
Motor Vehicle Manufacture
Paper Coating
Fabricated Metals
Machinery Manufacture
Furniture Manufacture
Flat Wood Products
Other Transportation Equipment
Electrical Equipment
Ship Building/Repair
Miscellaneous Industrial Manufacture
Miscellaneous Industrial Solvent Use
POTWs
Cutback
Chemical Manufacture Fugitives
Bulk Terminals and Bulk Plants
Petroleum Refinery Fugitives
Process Emissions - Bakeries
Process Emissions - Pharmaceuticals
Process Emissions - Synthetic Fibers
Crude Oil/Gas Production Fields
Hazardous Waste TSDFs
Total - Other Area Sources
TOTAL - POINT, MOBILE, AREA
NO,
(continued)
3
17
53
1
706
322
84
7
69
3
6
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
15,035
voc
46
5
279
5
681
78
99
161
17
12
47
1,298
495
319
280
590
1,054
222
58
340
107
51
89
15
3
14
11
847
2,230
11
130
195
405
301
48
45
93
70
787
11,676
22,557
1-12
-------
1. Because BEA economic forecasts for States are only released every five years, it was
determined that BEA was not the best source for economic data. In lieu of economicdata,
it was proposed that sub-national economic models be included in E-GAS because
models, rather than forecasts, would allow the user to base projections on the most up-to-
date economic information available.
2. Because UAM requires estimates of source growth for ozone nonattainment areas, it was
determined that E-GAS should include the capability to produce MSA-level economic
growth factors. After contacting numerous economic modeling firms, Regional Economic
Models, Inc. (REMI), which produces county-level economic models, was located. E-
GAS uses REMI models to produce growth factors, by source, for each of the
nonattainment areas and attainment portions of States in E-GAS. These growth factors
are applied to each county in the modeled area.
3. Regional economic models are driven by forecasts of national economic activity.
Therefore, E-GAS needed a national economic forecasting capability. It was determined
that the system would be most accurate if users were allowed to choose a national
economic forecast. Therefore, the E-GAS system is designed to allow users to make this
choice. However, it was determined that only two national forecasts would be supplied
with the system, Bureau of Labor Statistics (BLS) and Wharton Econometrics Forecasting
Associates (WEFA).
4. Based on the emission source strengths in the 1985 Northeast Transport Region inventory,
it was determined that separate forecasting modules were needed for estimating emissions
growth for the following categories: fuel consumption by electric utilities; industrial
physical output for the major VOC-producing sources; vehicle miles travelled by highway
vehicles; and fuel consumption by the commercial, residential, and industrial sectors.
5. Activity growth estimates should be developed at the level of disaggregation of the
emission inventories to be projected by the growth estimates. Therefore, it was
determined that growth factors would be developed for each of the point, area, and mobile
Source Classification Codes. This level of disaggregation allows E-GAS users to apply
model outputs to existing inventories to project emissions for both photochemical
modeling demonstrations and RFP planning.
1.4.6 E-GAS Design
Figure 1-1 contains the flow chart for E-GAS. As the flowchart indicates, E-GAS is
composed of three tiers: a national economic tier, a regional economic tier, and a growth factor
tier. Each of these tiers will be discussed briefly.
1-13
-------
Housing
Starts
Regional
Houring
Staru
_y Fuel Price
/ Forecasts /
Pud Ma
LJL
HOMES
i
Disenable Income
I Population
I
i Regional Capital.
Labor, MatariaU
Prk»
INRAD
Reddendal
FoasU
Fuel
Industry-Specific Vilue Added Data
WEFA
]
/ Fonscast
1
REMI
Interface
Tim 1
and 2
ACRONYMS LEGEND
REMI - Regional Economic Models, Inc.
WEFA - Wharton Econometric Forecast Anodalai
HPMS - Highway Performance Monitoring Syatem
HOMES - Hoosebold Model of-Energy by State
INRAD - Industrial Regional Activity and Energy Demand Model
CSEMS - Commercial Sector Energy Model by State
VMT - Vehicle Miles Travelled
SCC - Source Omfflrarlon Code
Tter3
Key
/7 Uaor-anecffiedfflei
f~"j Output file goea directly to CroMwalk
I ! Output file la uMdn Input to another model
F~~J Eristtag Model
| | Developed Module
Figure 1-1. Flowchart for the Economic Growth Analysis System
-------
1.4.6.1 Tier 1: The National Economic Tier
The national economic tier includes a REMI model of the United States which includes
a baseline forecast calibrated to the one released by the BLS. Although the BLS forecast is
updated every two years, REMI updates the forecast using data released annually by BEA. In
addition, the E-GAS national economic tier contains the option to use economic forecasts from
WEFA. WEFA forecasts national economic activity under low growth, base case, high growth,
and cyclical growth scenarios.
The function of the national tier in E-GAS is two-fold. First, the inclusion of a national
forecasting capability allows EPA to forecast urban and regional economic growth using a
common assumption about national economic growth. Second, it provides users with the ability
to use the most current national economic forecasts and to simulate the effects of different levels
of national growth on emission-producing activity in nonattainment areas.
The national economic tier is discussed in detail in Chapter 4.
1.4.6.2 Tier 2: The Regional Economic Tier
The regional economic tier includes separate economic models for each of the
nonattainment areas and attainment portions of States. The largest geographic area covered by
an economic model is a State.
The regional economic models included in E-GAS were built by REMI. The models
simulate interaction between the 14 major sectors of an economy and produce estimates of
employment and value added for 210 sectors. The 210-sector outputs are identified by BLS
industrial codes. The BLS codes are closely related to three-digit SIC codes. Outputs from the
regional models are used as input data for the growth factor tier.
The REMI models are designed to forecast future activity in an area and to simulate the
effects of a policy change in an area. The models come with a capability for the user to simulate
the effects of changes in almost 400 economic policy variables and over 70 demographic
variables. The list of policy variables included with E-GAS was reduced to 84 variables. Two
criteria were used for choosing which policy would be included in the system: whether the policy
variable relates to the implementation of the CAAA; and whether the variable is one which local
personnel using E-GAS would be knowledgeable of, particularly changes or proposed changes.
For example, industrial capital costs were included as a variable because that variable satisfies
1-15
-------
the first criterion. This variable will allow users to simulate the effects of control costs
associated with the CAAA. Policy variables that satisfy the second criterion include local tax
rates and State and local government spending. Policy variables which do not satisfy either
criterion, and therefore are not in E-GAS, include demographic variables such as birth and
survival rates, and economic variables such as demand for goods not affected by the CAAA.
Table 1-3 lists the policy variables included in E-GAS.
The REMI models and outputs contribute to the development of credible growth factors
for future-year inventories in the following ways:
1. Forecasts of activity from emission-producing sources were to be developed for both the
attainment and nonattainment portions of States, allowing growth rates to differ between
rural and urban portions of a State.
2. Outputs from the models are used to produce area-level estimates of fuel consumption,
VMT, and physical output.
3. The effects of a nonattainment area policy on the surrounding areas can be assessed.
4. Information on local policies can be entered directly into the REMI models. This ability
allows users to include the effects of local policies when developing forecasts.
REMI outputs and the growth factor tier are linked in the following specific ways.
• REMI models provide income forecasts for estimating residential
fuel consumption.
• REMI models provide population and personal income forecasts for
estimating commercial energy consumption.
• REMI models provide the forecasts of the relative costs of capital,
labor, and materials for estimating industrial fuel consumption.
• REMI models provide industry-specific employment and value
added forecasts for estimating physical output.
1-16
-------
TABLE 1-3. POLICY VARIABLES INCLUDED IN E-GAS
Employment Variables
Change in Employment in Durable Goods
Change in Employment in Nondurable Goods
Change in Employment in Mining
Change in Employment in Construction
Change in Employment in Transportation and Public Utilities
Change in Employment in Finance, Insurance, and Real Estate
Change in Employment in Retail Trade
Change in Employment in Wholesale Trade
Change in Employment in Services
Change in Employment in Agriculture, Farm, and Fishing Services
Change in Employment in State and Local Government
Change in Employment in Federal Civilian Government
Change in Employment in Federal Military
Change in Employment in Agriculture
Demand Variables
Final Demand for Durable Goods
Final Demand for Nondurable Goods
.Final Demand for Mining
Final Demand for Construction
Final Demand for Transportation and Public Utilities
Final Demand for Finance, Insurance, and Real Estate
Final Demand for Retail Trade
Final Demand for Wholesale Trade
Final Demand for Services
Final Demand for Agriculture, Farm, and Fishing Services
Personal Consumption Expenditure (PCE).
PCE - Autos and Parts
PCE - Furniture and Household Equipment
PCE - Other Durables
PCE - Food and Beverages
PCE - Clothing and Shoes
PCE - Gasoline and Fuel
PCE - Fuel Oil and Coal
PCE - Other Nondurables
PCE - Housing
PCE - Household Operation
PCE - Transportation and Public Utilities
PCE - Health Services
PCE - Other Services
PCE - Electricity
PCE - Natural Gas
PCE - Bus and Trolley Car Transportation
PCE - Taxicabs
PCE - Commuter Railway Transportation
PCE - Railway Transportation
PCE - Intercity Bus
(continued)
1-17
-------
TABLE 1-3. POLICY VARIABLES INCLUDED IN E-GAS (continued)
Investment
Residential Investment
Nonresidential Investment
Durable Equipment Investment
Fuel Costs
Relative Price of Commercial Electricity
Relative Price of Industrial Electricity
Relative Price of Commercial Natural Gas
Relative Price of Industrial Natural Gas
Relative Price of Commercial Oil
Relative Price of Industrial Oil
State and Local Government Spending
Elementary and Secondary Education
Higher Education
Other Education and Libraries
Health and Hospitals
Public Assistance and Relief
Sewerage
Sanitation
Police
Fire
Corrections
Highways
Water and Air Facilities
Transit Utilities
Other Commerce and Transportation
Gas and Electric Utilities
Water
Urban Renewal and Community Facilities
Natural and Agricultural Resources and Recreation
Other General Government
Local Facilities
New Communications Facilities
New Electric Utility Facilities
New Water and Sewer Supply Facilities
New Gas Utility and Pipeline Facilities
New Roads
New Local Transit Facilities
New Conservation and Development Facilities
Other
Change in Purchasing Power
Corporate Profit Tax Rate
Equipment Tax Rate
Personal Taxes
Property Tax Rate
1-18
-------
REMI models provide population forecasts for use in estimating
VMT.
The regional economic tier is discussed in Chapter 4.
1.4.6.3 Tier 3: The Growth Factor Tier
The third tier of E-GAS is the largest portion of the system. Housed within the third tier
are commercial, residential, industrial, and utility energy models; a VMT module; a physical
output module; and a Crosswalk. Each of these will be discussed.
The energy models in the system were developed by Argonne National Laboratories
(ANL) and are currently being used for the National Acid Precipitation Assessment Program
(NAPAP). The residential energy model, the Household Model of Energy (HOMES), was
modified for use in the NAPAP model set in the mid-1980s. In 1989-1990, ANL updated
HOMES to include the capability to model residential fuel consumption at the State, rather than
Census, level. For use in E-GAS, two changes were made to HOMES. First, the base year of
the model projections was updated to 1990 using data from the State Energy Data Report
(SEDS)* Additionally, the capability to estimate growth in residential fuel consumption at the
sub-State level was developed. REMI forecasts of population data for nonattainment areas and
attainment portions of States are input with State-level fuel price forecasts to develop estimates
of residential fuel consumption growth for seven fuels for each of the nonattainment areas and
attainment portions of States in E-GAS.
The commercial energy model, the Commercial Sector Energy Model (CSEMS), was also
developed for use in the NAPAP model set in the mid-1980s and updated in 1989-1990 to
estimate commercial fuel consumption at the State level. Like HOMES, the model was modified
for use in E-GAS to estimate commercial energy consumption growth for seven fuels for
nonattainment areas and surrounding attainment portions of States. The base year for the model
projections was updated to 1990 using data from SEDS. Inputs to CSEMS include State-level
fuel price forecasts and REMI forecasts of population and personal income at the sub-State level.
The industrial energy model, the Industrial Regional Activity and Energy Demand Model
(INRAD), was developed to predict how energy use will be influenced by energy prices and the
general level of economic activity.5 INRAD was developed to model energy consumption of
1-19
-------
fossil fuels and electricity for seven energy-intensive industries and an eighth "other" category
which aggregates the non-energy-intensive industries. Two modifications to INRAD were made
for use in E-GAS. First, additional industrial categories were modeled. Second, INRAD was
modified to estimate fossil fuel consumption by fuel type. With these modifications, INRAD can
estimate coal, oil, gas, and electricity consumption for the following sectors: food, textiles,
upstream paper products, downstream paper products, upstream chemicals, downstream
chemicals, glass, glass products, and metals. Inputs to INRAD include State-level forecasts of
fuel prices and REMI forecasts of the relative costs of capital, labor, and materials at the sub-
State level.
The VMT module projects growth in VMT for the modeled areas. EPA guidance
indicates that a single VMT projection may be applied to the entire mobile source category. For
short-term forecasts, the growth should be based on extrapolation of VMT trends in the years
1985 to 1990. E-GAS houses equations based on ordinary least squares analysis of VMT per
capita for 1985 to 1990 for each State. These forecasted equations are used with REMI
population data for the nonattainment areas and attainment portions of States. The combination
of EPA and REMI data allows VMT growth to be affected by State trends as well as total sub-
State population forecasts.
The physical output module estimates physical output from value added data generated
by the REMI models. Industrial VOC sources were ranked by their contributions to industrial
VOC emissions and equations were developed for the largest VOC sources. These equations
relate changes in physical output by three-digit SIC categories (as identified by BLS code) with
changes in value added and a time trend to capture technological change. These equations
provide better estimates of VOC-producing activity than value added alone because they estimate
change hi actual material output, which is related to the use of VOC producing materials, such
as surface coatings and degreasers. For industrial VOC categories for which equations were not
developed, activity levels are forecast using value added forecasts from the REMI models.
The Crosswalk is the final component of the E-GAS system. The Crosswalk translates
growth factors from the energy, VMT, and physical output modules into growth by SCC. The
growth factors from the industrial energy and physical output modules are disaggregated to the
two-, three-, and sometimes four-digit SIC level, while growth factors from the electric utility
model can be disaggregated to the plant or county level by type of fuel consumption. The
1-20
-------
commercial and residential sector energy models disaggregate consumption by fiiel type only.
The Crosswalk was developed by individually matching each of the approximately 7000 SCCs
with the appropriate growth factor from the modules. This allows different growth factors to be
applied to different emission sources from the same industrial category. For example, forecasts
of fuel consumption in upstream chemical manufacturing are developed by INRAD, while
forecasts of physical output of upstream chemical products are developed in the physical output
module. This methodology takes into account that future emissions associated with an SIC code
will vary by type of emission. This is consistent with the SCC system of classification which
differentiates according to not only industrial category, but also to processes within that category.
1-21
-------
1.5 REFERENCES
1. Lynch, T.M., T.R. Young, M.G. Smith, and E.S. Kimbrough. Design and Development
of the Economic Growth Analysis System (E-GAS). Paper presented at 2nd Annual
EPA/Air & Waste Management Association Specialty Conference, Emission Inventory
Issues and Progress, Durham, NC. October 19-22, 1992.
2. Possiel, N.C., L.B. Milich, B.R. Goodrich, E.L. Meyer, and K.L. Schere. Regional Ozone
Modeling for Northeast Transport - Development of a Base Year Anthropogenic Emissions
Inventory. EPA-450/4-91-002a (NTIS PB92-108786/AS). Research Triangle Park, NC.
June 1991.
3. U.S. Environmental Protection Agency. Procedures for Preparing Emissions Projections.
EPA-450/4/91-019 (NTIS PB91-242404). Research Triangle Park, NC. July 1991.
4. U.S. Department of Energy. State Energy Data Report Consumption Estimates, 1960 -
1990, DOE EIA - 0214 (90). Office of Energy Markets and End Use, Energy Information
Administration. Washington, DC. May 1992.
5. Boyd, G.A., E.G. Kokkelenberg, and M.H. Ross. Sectoral Electricity and Fossil Fuel
Demand in U.S. Manufacturing: Development of the Industrial Regional Activity and
Energy Demand (INRAD) Model. Argonne National Laboratory. Argonne, IL. February
1990.
1-22
-------
CHAPTER!
STATUTORY BACKGROUND AND USER REQUIREMENTS FOR USING E-GAS
TO PROJECT EMISSIONS
2.1 INTRODUCTION
Title I of the CAAA requires that ozone nonattainment areas classified as serious, severe,
extreme and multi-State moderate use photochemical grid modeling to demonstrate future
attainment with the ozone national ambient air quality standard [Section 182(E)(2)(A)]. In
addition to photochemical grid modeling, Section 182(b)(l)(A) requires that moderate, serious,
severe and extreme ozone nonattainment areas submit a rate of progress plan to demonstrate how
the area will achieve the 15 percent reduction in VOC emissions by 1996. Areas classified as
serious, severe and extreme are also required to demonstrate how the area will achieve a three
percent annual reduction (averaged over three years) in VOC or NOX from 1996 until the area
reaches attainment [Section 182(c)(2)(B)].
EPA is currently drafting guidance to aid States in the development of the rate of progress
plans. The rate of progress plan (due in 1993) must include the 15 percent demonstration, the
control strategy, adopted rules identified in the control strategy and an attainment demonstration
for moderate areas. The post-1996 rate of progress plan (due in 1994) will include the above
information and the attainment demonstration for serious and above areas.
Section 182(b)(l)(A) specifies that the 15 percent reduction from baseline emissions
accounts for any growth in emissions after 1990. A key component of these rate of progress
plans is the projection of emissions that will be required to determine growth in the area. E-GAS
can provide the growth factors necessary to project future emissions.
This chapter discusses potential E-GAS user groups; statutory requirements of the CAAA
for which E-GAS may be used; and system requirements identified during the development of
E-GAS.
2-1
-------
2.2 POTENTIAL E-GAS USERS
States that have ozone nonattainment areas classified moderate and above are required to
submit rate of progress plans that will include projections of emissions for all source categories
within the nonattainment area. In order for the States to prepare accurate estimates of emissions,
appropriate emission growth factors must be developed. E-GAS will support that development.
Probable E-GAS user groups include the following:
• State and Local Air Agencies
• EPA Regional Offices
EPA Office of Air Quality Planning and Standards (OAQPS)
OAQPS Air Quality Management Division (AQMD)
OAQPS Technical Support Division (TSD)
2.3 TERMINOLOGY
The following terms will be used hi this report and hi discussions concerning the use of
E-GAS to project emission inventories for modeling and rate of progress plans.
Rate of Progress EPA has defined rate of progress as the 15 percent emissions reduction from
1990 emissions required by November 15, 1996 [Section 182(b)(l)].
Reasonable Further Progress Reasonable further progress is defined in Section 182(c)(2) as the
three percent per year averaged over consecutive three year periods from November 15, 1996
until the areas are redesignated.
Rate of Progress Plan The rate of progress plan is the portion of the State implementation plan
(SIP) revision (due in 1993) that illustrates the plan for the achievement of the 15 percent
emissions reduction.
Post-1996 Rate of Progress Plan The post-1996 rate of progress plan is the portion of the SIP
revision (due hi 1994) that illustrates the plan for the achievement of the nine percent emissions
reductions every three years.
2-2
-------
Base Year Emission Inventory Section 182(a)(l) defines this inventory as a "comprehensive,
accurate, current inventory of actual emissions from all sources", which includes 1990 emissions
of VOC, NOX, and CO.
Baseline Emissions Section 182(b)(l)(B) defines baseline emissions as the total amount of VOC
or NOX emissions from all anthropogenic sources in the area excluding emissions that would be
eliminated under the regulations described in Section 182(b)(l)(D)(i) and (ii).
1990 Actual Inventory This inventory reflects only emissions within the designated
nonattainment area. EPA has interpreted that the 15 percent reduction must be from sources
within the nonattainment area.
Projected Emission Inventory This inventory is necessary to determine the control strategy that
an area will need to meet the required emission reductions and eventually attain the standard.
1996 Target .Level of Emissions EPA has defined this to be the level of emissions in a
nonattainment area necessary for the area to meet the rate of progress requirements.
Milestone Demonstration Demonstrating achievement of the 15 percent VOC reduction in the
first 6 years after enactment and then subsequently demonstrating achievement of the 3 percent
VOC reduction per year averaged over 3 years from November 15, 1996, are defined as
milestone demonstrations. Milestone demonstrations must be submitted to EPA within 90 days
of the milestone date in accordance with Section 182(g)(2).
User Requirements Analysis The description of user needs in terms of input and output
capabilities.
2.4 OVERVIEW OF REASONABLE FURTHER PROGRESS REQUIREMENTS
Section 182(b)(l) of the CAAA requires all ozone nonattainment areas classified moderate
and above to submit SIP revision to provide for reductions in VOC emissions of at least 15
percent during the first 6 years after enactment.3 The purpose of this specified rate of reduction
program is to establish a consistent requirement for all ozone nonattainment areas classified
moderate and above. The 15 percent reduction requirement is intended to set a minimum level
for emission reductions. The baseline from which the 15 percent reduction is calculated is
defined as all anthropogenic emissions (VOC and NOJ during calendar year 1990 excluding the
emissions that would be eliminated by Federal Motor Vehicle Control Program (FMVCP)
This submission must be made by November 15, 1993.
2-3
-------
regulations promulgated by January 1, 1990 and Reid vapor pressure (RVP) regulations
promulgated by November 15, 1990, or regulations required to be promulgated under section
211(h) which requires RVP no greater than 9.0 pounds per square inch (psi) during the high
ozone season (7.8 psi in the southern portions of the United States) [Section 182(b)(l)(B)].
These expected emission reductions are removed from the baseline prior to calculating the
required 15 percent emission reduction.
Emission reductions from the following types of regulations are not creditable toward the
15 percent progress requirement:
• FMVCP regulations promulgated by EPA by January 1, 1990
• RVP regulations promulgated by EPA by November 15, 1990 or required to be
promulgated under Section 21 l(h) which requires RVP no greater than 9.0 psi during the
high ozone season (7.8 psi in the southern portions of the United States)
• Regulations submitted to correct deficiencies in existing VOC reasonably available control
technology (RACT) regulations as required under Section 182(a)(2)(A)
• Regulations submitted to correct deficiencies in inspection and maintenance (I/M)
programs as required under Section 182(a)(2)(B)
All other emission reductions are creditable.
The expected reductions from FMVCP and RVP are adjusted out of the baseline prior to
calculating the required 15 percent reduction (via the development of the adjusted base year
inventory). By adjusting the baseline for these two programs, States lower the 15 percent
emission reduction requirement. Congress allowed this adjustment to ensure that States would
be fully credited for relevant reductions (i.e., the adjustment recognizes that the reductions from
these programs should have already occurred and therefore lowers the inventory from which the
15 percent requirement is calculated).
A nonattainment area can achieve less than the 15 percent required reductions if the State
can demonstrate that: (1) the area has a New Source Review program equivalent to the
requirements hi extreme areas [Section 182(e)], except that "major source" must include any
source which emits, or has the potential to emit, 5 tons per year of VOC; and (2) all major
sources (those which emit 5 or more tons per year) in the area must have RACT level controls.
2-4
-------
The plan must also include all measures which can be feasibly implemented in the area. Finally,
the State must demonstrate that the plan includes all measures achieved in practice by sources
in the same source category in nonattainment areas of the next higher classification. The waiver
for the 15 percent progress requirement cannot apply to nonattainment areas classified as extreme.
Section 182(c)(2)(B) requires that serious, severe and extreme ozone nonattainment areas
submit a post- 1996 rate of progress plan by November 15, 1994. The plan must provide
reductions in VOC emissions of at least three percent per year averaged over three consecutive
years beginning November 15, 1996 until the area reaches attainment. A nonattainment area can
achieve less than the three percent per year required reductions if the State can demonstrate that
the plan includes all measures which can be feasibly implemented in the area, in light of
technological achievability. Additionally, the State must demonstrate that the plan includes all
measures achieved in practice by sources in the same source category in nonattainment areas of
the next higher classification. The waiver for the three percent per year progress requirement
cannot apply to nonattainment areas classified as extreme. A determination of the waiver from
the three percent per year requirement will be reviewed at each milestone under Section 182(g)
and revised to reflect the availability of any new technologies or other control measures for
sources in the same category. The baseline for the three percent per year reductions and
creditability requirements is the same as for the 15 percent progress requirement under Section
2.5 OVERVIEW OF ATTAINMENT DEMONSTRATION REQUIREMENTS
Section 182(b)(l)(A) requires a SIP revision for a moderate ozone nonattainment area to
provide for reductions hi VOC and NOX emissions "as necessary to attain the national primary
ambient air quality standard for ozone." This requirement can be met through applying EPA-
approved modeling techniques described hi the current version of EPA' s Guideline on Air Quality
Models (Revised).1 The Urban Airshed Model, a photochemical grid model, is recommended for
modeling applications involving entire urban areas. In addition, for moderate areas contained
solely in one State, the city-specific Empirical Kinetic Modeling Approach (EKMA) may be an
acceptable modeling technique. The State should consult with EPA prior to selection of a
modeling technique. If EKMA is used, the attainment demonstration is due by November 1993.
2-5
-------
In other cases, a State might choose to use a photochemical grid model instead of EKMA.
Grid modeling will generally provide a better tool for decision makers and the necessary
additional time may, therefore, be justified. In such cases, States should consult with EPA on
a case-by-case basis on an acceptable approach to meeting the Section 182(b)(l)(A) requirement
through an interim SIP submittal by November 1993 and a completed attainment demonstration
by November 1994. The interim submittal would include, at a minimum, evidence that the grid
modeling has begun and a commitment, with schedule, to complete the modeling and submit it
as a SIP revision by November 1994. The completed attainment demonstration would include
any additional controls needed for attainment. Separate attainment demonstration requirements
apply to multi-State moderate areas, as described below.
Moderate and above multi-State ozone nonattainment areas must submit attainment
demonstrations which use photochemical grid modeling (or equivalent) (Section 182(j)(l)(B)).
The Urban Airshed Model is recommended for modeling applications involving entire urban
areas. Care should be taken to coordinate strategies and assumptions in a modeled area with
those in other, nearby modeled areas in order to ensure that consistent, plausible strategies are
developed. EPA has further interpreted the requirements of Section 182(j) to supersede the
requirements of 182(b). This means that a State must submit a SIP revision providing for the
15 percent reduction in VOC emissions from 1990 through 1996 by November 15, 1993. A
second SIP revision including the necessary provisions to demonstrate attainment of the NAAQS
is due November 15, 1994. The timing of these submittals is identical to the requirements for
serious ozone nonattainment areas.
Section 182(c)(2)(A) requires a SIP for a serious ozone nonattainment area to provide an
attainment demonstration by November 15, 1994. The "attainment demonstration must be based
on photochemical grid modeling or any other analytical method determined by the Administrator,
in the Administrator's discretion, to be at least as effective" (Section 182(c)(2)(A)). This
requirement can be met through applying EPA-approved modeling techniques for SIP revisions.1
The Urban Airshed Model is recommended for modeling applications involving entire urban
areas.
Serious areas generally must meet all requirements of moderate ozone nonattainment
areas. As previously discussed, moderate areas are required to provide for reductions in VOC
and NOX emissions "as necessary to attain the national primary ambient air quality standard for
2-6
-------
ozone" (Section 182(b)(l)(A)). To determine the "necessary" emission reductions, an attainment
demonstration is generally required by November 1993, if a photochemical grid model is not
used. Serious (and higher) areas, however, must complete photochemical grid modeling analyses
and have longer attainment deadlines. In consideration of the additional time necessary to gather
data to support and to perform a grid modeling analysis, Congress provided an additional year
for serious (and higher) areas to submit their demonstrations of attainment. Due to Congress'
allowance of this additional year, EPA believes that the Section 182(c) requirement for serious
and higher ozone nonattainment areas to submit photochemical grid modeling by November 1994
supersedes the attainment demonstration otherwise applicable under Section 182(b).
2.6 USEOFE-GAS
In developing strategies for complying with the CAAA deadlines, States will examine an
array of complex compliance strategies, and will estimate the impact of these strategies on future
ozone air quality hi nonattainment areas. These assessments will be accomplished by first
estimating future emissions of VOC, NOX, and CO, (all ozone precursors) and then estimating
ambient air quality impacts with atmospheric chemistry models such as the Urban Airshed Model
and the Regional Oxidant Model. Emission forecasts, which are a critical component in both
UAM and ROM analyses, will be estimated both on the anticipated effectiveness of emission
control strategies and on national and local economic growth assumptions.
The primary purpose of E-GAS is to allow State and EPA staff to forecast future growth
in the activity levels of ozone precursor emissions sources. These activity growth estimates can
then be used to project future activity levels and conduct control strategy analyses using emission
estimation models. E-GAS will estimate source-specific growth factors which can serve as input
to the Emissions Preprocessor System (EPS) for UAM, which was developed by the Office of
Air Quality Planning and Standards. E-GAS estimates economic growth projections, employment
growth projections, growth in production and energy consumption, changes in demographic
variables, and other parameters. Outputs from the model were developed in a format that is
compatible with AIRS formats, so that it is possible to use E-GAS outputs to grow AIRS
emission inventories.
2-7
-------
2.7 SYSTEM REQUIREMENTS
Requirements for E-GAS have been organized into the following categories:
Functional requirements - Capabilities identified by customers that could be provided by
the system which can directly support control strategy development activities.
• System attributes - General operating requirements describing user interaction with the
system.
2.7.1 Functional Requirements
The functional requirements discussed in this section are those that have been suggested
by customers that could directly support control strategy development.
Output is in a form consistent with the Emission Preprocessor System (EPS). EPA has
already developed EPS to manipulate the emission inventory data provided by a State to make
them usable for UAM inputs. E-GAS outputs are also in a generic ASCII file that can be input
into AIRS and other systems.
2.7.2 Required System Attributes
In addition to the functional requirements previously described, the following system
attributes have been identified.
2.7.2.1 Easy Data Entry
The need for simple data entry for front-line personnel was identified since data entry
personnel are often responsible for entering only a small subset of data, specific to their function.
They need to be able to quickly view, edit, and update only those data that are important to them,
instead of scrolling through dozens of screens hi order to locate the five or six data elements that
they need to update. Customized user-views of the data are needed to present only pertinent
information to front-line data entry personnel.
2-8
-------
2.7.2.2 User Friendly
In addition, user friendliness and state-of-the-art help features are important since users
are accustomed to PC-type help features (or the help features in a system such as TRIS), such
as the use of the Fl key for field sensitive help, pop-up tables to identify codes and descriptors,
on-line look-up tables to identify acronyms and descriptors, and standard query languages. Users
consider these features critical to the successful use of the system.
2.7.2.3 Quality Assurance
A variety of quality assurance (QA) tools will be needed by State agencies to verify that
data reported by industry are both complete and correct. Typical types of data QA tools would
include edit checks, completeness checks, and reasonableness checks.
2.7.2.4 Data Security
In terms of data security, States often bear the prime responsibility for ensuring that
confidential information supplied by industry is protected. Extreme precautions are needed to
guard against unauthorized access.
2.7.2.5 State-owned Data
States often collect and use data for their own purposes and do not want EPA personnel
to have access to these data. A capability must exist to allow States to protect State-owned data
if desired.
2.8 CONCLUSIONS
The statutory requirements for Reasonable Further Progress and demonstrating attainment
clearly point to the need for a system which will project activity growth factors. E-GAS was
developed to serve that purpose and to aid State and local agencies in the development of their
control strategies for meeting those requirements.
2-9
-------
2.9 REFERENCES
1. U.S. Environmental Protection Agency. Guideline on Air Quality Models, Revised. EPA-
450/2-78-027R (NTIS PB86-245248). Office of Air Quality Planning and Standards.
Research Triangle Park, NC. July 1986.
2-10
-------
CHAPTER 3
PROJECTING EMISSION INVENTORIES: EPA GUIDANCE
In order to determine the growth factors to be used for each emission source, EPA
guidance on emissions projections and inventory requirements for the photochemical air models
was reviewed. Emissions projection guidance is summarized in Procedures for Preparing
Emissions Projections1; and inventory requirements for the photochemical models is contained
in Procedures for the Preparation of Emission Inventories for Carbon Monoxide and Precursors
of Ozone, Volume II: Emission Inventory Requirements for Photochemical Air Quality Simulation
Models.2 This chapter presents a summary of the guidance for projecting point, area, and mobile
sources, along with a discussion of how E-GAS methodologies comply with EPA guidance.
3.1 GENERAL
If emission source growth estimates are not available from individual plants or other local
sources, a surrogate growth indicator must be used. The preferred data source for projecting
stationary source categories is the U.S. Department of Commerce's Bureau of Economic Analysis
growth factors. The BEA has published Metropolitan Statistical Area (MSA), State and regional
growth factors in hard copy and disk format under the titles, Bureau of Economic Analysis
Regional Projections to 2040 Volumes 1, 2, and 3.3'4'5 This source includes personal income,
earnings, and employment data for the MSAs, States, regions, and the entire United States.
Economic variables which may be used as indicators of emission source growth are
product output, value added, earnings, and employment. Product output is measured in physical
units; value added is the difference between the value of industry outputs and inputs; earnings
denotes wage earnings in an industry; and employment measures the number of workers in an
industry. The emission projection guidance suggests that product output is the best indicator of
future emission source growth and that its use is "preferable to any of the [other] indicators, if
it is available".1 If product output projections are not available, value added data should be used,
and if they are not available, earnings data may be used. Finally, employment projections may
be used, but are not considered to be "an effective growth indicator in most cases." The
3-1
-------
emissions projection guidance also indicates that for the purposes of projecting SIP inventories,
States are expected to use earnings, value added, or product output data.1
3.2 POINT SOURCES
3.2.1 EPA Point Source Projection Guidance
Sources of information for projecting point source emissions include the facilities where
the sources are located and local planning agencies. The permit application process may also
yield information on planned construction or expansion of existing capacity.2 However, the
emission projection guidance suggests that plant-specific surveys may not always be a reliable
source of information because much of the information that is relevant to emission projections
(e.g., growth or decline in output, plans for expansion) may be confidential. EPA suggests that
a survey of individual point sources only be performed if the following certain circumstances
apply: (1) the industry is a dominant industry in the region; (2) the industry's growth may not
be captured in the regional projections; and (3) it is expected that the industry may experience
significant growth or decline.1 Finally, emissions growth at a plant may be projected from
information obtained for other point sources in that area and category. This procedure uses a
growth trend developed from information from a group of facilities and applies it to a facility for
which there is no available information.
When information is not available from plants, permit applications, or local planning
agencies, projected economic variables may be used to estimate emission source growth. These
factors were previously discussed in Section 1.4.
3.2.2 E-GAS Point Source Growth Factors
E-GAS will be used to project the AIRS point source inventories which are housed in the
AIRS Facility Subsystem (AIRS/FS). These projected inventories will be used in photochemical
grid modeling and RFP inventories. Because the AIRS/FS inventories will be projected on a
source-specific basis, the user will be able to choose each growth factor. For example, if a user
has information from permits or plant surveys about the expected growth of a point source, the
3-2
-------
user may use that information to predict future growth of that source within E-GAS. The ability
of the user to override default growth factors may be most important for electric utilities, which
are permitted sources and are major emitters of NOX. E-GAS produces default growth factors
for commercial and industrial energy consumption, fuel consumption by electric utilities, and
physical output by Bureau of Labor Statistics code, which represent groups of three- and four-
digit SICs. These growth factors are then translated, via the E-GAS CROSSWALK, into default
growth factors by SCC. Because there is no direct linkage between E-GAS and AIRS, users may
alter the E-GAS growth factor file based on information that they have on specific emission
sources.
E-GAS uses the following information for projecting point source growth factors:
1. Value added estimates for 210 non-farm industrial categories
2. Physical output estimates for some major VOC-emitting sources
3. Estimates of fuel consumption by type of fuel for the commercial, industrial, and electric
utility sectors
The CROSSWALK, which translates economic and energy consumption forecasts into
activity growth by SCC, is discussed in detail in Chapter 8.
3.3 AREA SOURCES
3.3.1 EPA Area Source Projection Guidance
The major difference between area and point source projection is the need for the area
source growth to be allocated to grid cells. E-GAS does not project growth factors by grid cell,
but provides area source growth factors for each nonattainment area, the remaining portion of
surrounding State(s), and each State in one of the ROM domains. However, it is beyond the
scope of this portion of the projection methodology to allocate these growth factors to sub-MSA
areas.
3-3
-------
EPA guidance on projecting inventories includes preferred growth indicators for area
sources. These indicators are listed in Table 3-1, along with potential information sources
provided by EPA. Because area sources are not individually projected, information from permits
or specific plants cannot be used directly. Local studies or surveys, however, may provide
information that can be used to develop surrogate growth factors.1
3.3.2 E-GAS Area Source Growth Factors
EPA-preferred growth indicators are listed in Table 3-1. In Table 3-2, these indicators
are listed along with E-GAS outputs which can be used as growth factors. As Table 3-1 shows,
metropolitan planning organizations (MPOs) will probably be the best source of information for
some of these sources. There are sources for which E-GAS does not provide growth factors
because there are no outputs from E-GAS which match or can approximate a recommended EPA
growth indicator. These sources are primarily biogenic sources. When there is no appropriate
growth factor for a source, E-GAS assigns a factor of 1.0 (no growth).
E-GAS uses the following information for projecting area source categories:
1. Value added estimates for 210 non-farm industrial categories
2. Physical output estimates for some major VOC-emitting sources
3. Estimates of fuel consumption by type of fuel for the commercial, industrial, and electric
utility sectors
4. Vehicle miles travelled
5. Population
Each emission source in the AIRS Area and Mobile Source (AIRS/AMS) inventories is matched
by the CROSSWALK with the appropriate growth factor. These growth factors correspond to
the E-GAS outputs identified in Table 3-2. As with the point sources, the user may override an
SIC or SCC growth factor and enter his/her preferred value.
3-4
-------
TABLE 3-1. EPA-PREFERRED GROWTH INDICATORS FOR PROJECTING
EMISSIONS FOR AREA SOURCE CATEGORIES
Source Category
Growth Indicators
Information Sources
Gasoline Marketing
Dry Cleaning
Degreasing (Cold Cleaning)
Architectural Surface Coating
Automobile Refinishing
Small Industrial Surface Coating
Graphic Arts
Asphalt Use - Paving
Asphalt Use - Roofing
Pesticide Applications
Commercial/Consumer Solvent Use
Publicly Owned Treatment Works
(POTWs)
Hazardous Waste Treatment,
Storage and Disposal Facilities
(TSDFs)
Municipal Solid Waste Landfills
Residential Fuel Combustion
Commercial/Institutional Fuel
Combustion
Industrial Fuel Combustion
Aircraft (Commercial and General)
Aircraft, Military
Railroads
Ocean-going and River Cargo
Vessels
Vessels, Small Pleasure Craft
Off-Highway Motorcycles
Agricultural Equipment
projected gasoline consumption
population; retail service
employment
industrial employment
population or residential dwelling
units
industrial employment
industrial employment
population
consult industry
industrial employment; construction
employment
historical trends in agricultural
operations
population
site-specific information
State planning forecasts
State waste disposal plan
residential housing units or
population
commercial/institutional
employment; population
industrial employment (SIC 10-14,
50-51); or industrial land use
site-specific forecasts
site-specific forecasts
revenue ton-miles
cargo tonnage
population
population
agricultural land use; agricultural
employment
MOBELE4 fuel consumption model
solvent suppliers; trade associations
trade associations
local Metropolitan Planning
Organization (MPO)
BEA
BEA
State planning agencies; local MPO
consult industry
local industry representatives
State department of agriculture;
local MPO
local MPO; State planning agencies
State planning agencies
State planning agencies; local MPO
local MPO; State planning agencies
local MPO
local MPO; land use map
projections
local MPO; land use projections;
State planning agencies
local airport authority and
commercial carriers
local airport authorities; appropriate
military agencies
American Association of Railroads
and local carriers
local port authorities; U.S. Maritime
Administration; U.S. Army Corps
of Engineers
local MPO
local MPO
local MPO; Census of Agriculture
(continued)
3-5
-------
TABLE 3-1. EPA-PREFERRED GROWTH INDICATORS FOR PROJECTING
EMISSIONS FOR AREA SOURCE CATEGORIES (continued)
Source Category
Growth Indicators
Information Sources
Construction Equipment
Industrial Equipment
Lawn and Garden Equipment
On-site Incineration
Open Burning
Fires: Managed Burning,
Agricultural Field Burning, Frost
Control (Orchard Heaters)
Forest Wildfires
Structural Fires
industry growth (SIC Code 16)
industrial employment (SIC codes
10-14, 20-39, 50-51) or industrial
land use areas
single-unit housing
based on information gathered from
local regulatory agencies
based on information gathered from
local regulatory agencies
areas where these activities occur
historical average
population
local MPO
local MPO
local MPO
local regulating agencies and MPO;
State planning agencies
local agencies; State planning
agencies; local MPO
U.S. Forest Service, State
agricultural extension office
local, State, and federal forest
management officials
local MPO; State planning agencies
3-6
-------
TABLE 3-2. E-GAS GROWTH FACTORS FOR PROJECTING AREA
SOURCE EMISSIONS
Source Category
EPA Preferred Growth
Indicators
Relevant E-GAS Growth Factors
Gasoline Marketing
Dry Cleaning
Degreasing (Cold Cleaning)
Architectural Surface Coating
Automobile Refinishing
Small Industrial Surface Coating
Graphic Arts
Asphalt Use - Paving
Asphalt Use - Roofing
Pesticide Applications
Commercial/Consumer Solvent Use
Publicly Owned Treatment Works
(POTWs)
Hazardous Waste Treatment,
Storage and Disposal Facilities
(TSDFs)
Municipal Solid Waste Landfills
Residential Fuel Combustion
Commercial/Institutional Fuel
Combustion
Industrial Fuel Combustion
Aircraft (Commercial and General)
Aircraft, Military
Railroads
Ocean-going and River Cargo
Vessels
Vessels, Small Pleasure Craft
Off-Highway Motorcycles
projected gasoline consumption
population; retail service
employment
industrial employment
population or residential dwelling
units
industrial employment
industrial employment
population
consult industry
industrial employment; construction
employment
historical trends in agricultural
operations
population
site-specific information
State planning forecasts
State waste disposal plan
residential housing units or
population
commercial/institutional
employment; population
industrial employment (SIC 10-14,
50-51); or industrial land use
site-specific forecasts
site-specific forecasts
revenue ton-miles
cargo tonnage
population
population
value added in petroleum refinery
value added in laundry and cleaning
services
value added in specific industry
population
value added in automobile repair
value added in specific industry
value added in commercial printing
value added in asphalt, paving, and
roofing materials
value added in asphalt, paving, and
roofing materials
value added in non-manufacturing
services
population (consumer)
value added in sanitary services
population
population
estimate from E-GAS fuel module
commercial fuel consumption
estimates from E-GAS fuel module
industrial fuel consumption
estimates for 2-digit SICs from
E-GAS fuel module
value added in air transportation
value added in air transportation
value added in railroad
transportation
valued added in water
transportation
population
population
(continued)
3-7
-------
TABLE 3-2. E-GAS GROWTH FACTORS FOR PROJECTING AREA
SOURCE EMISSIONS (continued)
Source Category
EPA Preferred Growth
Indicators
Relevant E-GAS Growth Factors
Agricultural Equipment
Construction Equipment
Industrial Equipment
Lawn and Garden Equipment
On-site Incineration
Open Burning
Fires: Managed Burning,
Agricultural Field Burning, Frost
Control (Orchard Heaters)
Forest Wildfires
Structural Fires
agricultural land use; agricultural
employment
industry growth (SIC Code 16)
industrial employment (SIC codes
10-14, 20-39, 50-51) or industrial
land use areas
single-unit housing population
based on information gathered from population
local regulatory agencies
based on information gathered from population
local regulatory agencies
areas where these activities occur —
value added in fanning
value added in construction
value added in manufacturing
historical average
population
population
3-8
-------
3.4 MOBILE SOURCES
3.4.1 EPA Guidance on Projection of Mobile Sources
EPA guidance on projection of mobile source emissions can be found in Procedures for
Preparing Emissions Projections.1 This guidance covers highway vehicles as well as some non-
highway mobile sources (aircraft and railroads). Additional guidance specific to highway mobile
source inventory forecasting and tracking for CO nonattainment areas is contained in Section 187
VMT Forecasting and Tracking Guidance6, a document required by Section 187 of the Clean Air
Act Amendments. These two documents discuss the same basic methods and sources for mobile
source projections. In order of preference, these include:
1. Use of projections based on a network-type travel demand model for the area of concern
2. Use of projections based on data generated by the Federal Highway Administration
(FHWA) Highway Performance Monitoring System (HPMS) for the subject area
3. Use of "any reasonable methodology" for areas not covered by HPMS
Details on the information presented in the two guidance documents are discussed below.
The Procedures for Preparing Emissions Projections1 states that the preferred method for
performing VMT projections for on-road mobile sources is to use a validated travel demand
model. Travel demand models are locality-specific computerized models which simulate travel
on a network representing an area's transportation system. The number of cities with a current
travel demand model is limited and there are many nonattainment areas without such models.
Resources involved in developing a model are substantial, and creating a model for inventory
purposes alone may not be warranted. For areas that do not have a validated travel demand
model, this guidance permits VMT projections to be based on the FWA's HPMS. For areas
outside the domain of a travel demand model and/or HPMS reporting area, the use of an
historically-based extrapolation method is allowed. An example trend projection method,
requiring the quantifying of road mileage and associated VMT, is outlined; however, details on
these methodologies are not provided.
3-9
-------
The Section 187 VMT Forecasting and Tracking Guidance6 makes the following specific
recommendations for procedures for CO nonattainment areas, distinguishing between the baseline
inventory, forecasting and estimates to be made in the future for tracking purposes. Only the
forecasting requirements are directly relevant to the projections of concern to E-GAS, but the
other requirements are included for clarity.
1990 Baseline CO Inventory Use of HPMS to develop the 1990 baseline CO
inventory, with alternate use of a travel demand model permitted only if the model
available is believed to be strong and the HPMS data for the area are weak.
VMT Forecasting Preferred use of a travel demand model to forecast VMT, based
on the assertion that these models are the best predictors of VMT growth (but not
necessarily of absolute VMT). If a model cannot be made available, an "historical
area-wide VMT" method can be used, based on a regression analysis of the area's
1985 to 1990 HPMS-reported data. However, States forecasting beyond 1996 are
required to use a travel demand forecast. States may use "any reasonable
methodology" to forecast VMT for the portion of the VMT Tracking Area outside
of the Federal Aid Urbanized Area.
VMT Tracking Annually, beginning in 1993, estimates of actual VMT for each
CO nonattainment area are to be made for tracking against the forecast VMT.
Since repeated use of a travel demand model is very costly, EPA recommends use
of each year's HPMS data for tracking, with upgrading of the area's HPMS
sample to specific standards for this purpose.
Specific guidance has not been issued for forecasting and tracking mobile source emissions in
ozone nonattainment areas, but it is stated hi the CO guidance6 that these procedures, when
issued, "are expected to be consistent" with the CO guidance described above.
3-10
-------
Guidance for SIP inventory projections for nonhighway mobile sources1 is as follows.
For aircraft, major commercial airports should be surveyed individually to determine their
specific growth plans. The best source of information on potential growth in rail travel is the
railroad companies themselves, along with the American Association of Railroads. Similar
sources are recommended for obtaining data on projected growth in commercial water vessel
activity. Guidance on gasoline marketing provides fuel efficiency ratios that allow forecasting
based on fuel economy changes due to fleet turnover, as well as changes in the total number of
vehicle miles traveled in the area under consideration.
3.4.2 E-GAS Mobile Source Growth Factors
Although the first choice in the EPA VMT projection guidance is the use of
locality-specific travel demand models, it is beyond the scope of E-GAS to include these models
in the system. The main difficulties are the variety of model types and data formats, and the
amount of interaction that would be required to accomplish this task for the relatively small
number of areas with current travel demand models. The system, however, includes the
capability to insert area-specific modeling results from travel demand models. When this option
is chosen, results from the E-GAS VMT module are replaced by user-specified estimates of
VMT. The user may provide an overall VMT growth estimate, VMT estimates by road type,
or VMT estimates by road and vehicle type.
To conform with the next level of the EPA guidance, E-GAS mobile source projection
methods are based on State-level Federal Aid Urbanized Area HPMS data for 1985 to 1990. The
methodology uses regression analysis of these data to establish short-term State-level trends in
VMT per capita growth. The methodology and data are discussed in detail in Chapter 8.
For the "rest-of-State" areas outside the nonattainment areas, EPA guidance indicates that
VMT can be projected by a method such as (1) performing similar regressions of historic HPMS
State-level VMT statistics; (2) obtaining "rest-of-State" projections by subtracting out individual
projections for any cities (obtained as described above); and (3) bounding the "rest-of-State"
projection as done for the REMI cities. A discussion of the methodology used to project "rest-of-
State" VMT is presented in Chapter 7.
3-11
-------
3.5 EPA GUIDANCE ON PROJECTING EMISSIONS FROM UTILITIES
3.5.1 General
EPA guidance on projecting emissions from electric utilities includes projection
methodologies for existing, planned, and additional electricity-generating units. The guidance
is summarized in Table 3-3.
TABLE 3-3. ELECTRIC UTILITY NO, PROJECTIONS SUMMARY
Estimate emissions from existing units:
- determine State-level growth factors
- estimate unit-level future year capacity factors
- determine unit-level NOX control requirements
Estimate emissions from planned units:
- obtain listing of planned units
- determine most likely siting for undesignated units
- determine applicable unit-level NOX emission rates (and default data)
Estimate emissions from generic units:
- determine amount of additional generation needed (if any)
- estimate NOX emission rate
- determine siting for generic units
The guidance requires that the methodology for projecting emissions from existing units
should be based on State-level electricity growth factors, estimated unit-level capacity factors,
and unit-level NOX control requirements. The estimated unit-level capacity factors for any future
year should not exceed 80 percent, unless the 1990 capacity factor exceeded 80 percent. The
emission rate requirements should be based on the most stringent regulation (local, State, or
federal) which applies to the unit.1
For planned units, the guidance requires the use of announced plants; the Department of
Energy annually publishes a list of plants that are expected to come on line in the next ten years.
In addition, for announced plants without a designated site location, it should be assumed that
any future unit whose pollutant contribution would exceed the amounts retired in a nonattainment
3-12
-------
area would be located outside the boundaries of the nonattainment area. The NOX emission rate
assigned to new units will depend on the year that the unit comes on line, as future standards
[e.g., the revised new source performance standard (NSPS) for 1994] will determine the most
stringent standard which applies to the unit.1
For "generic" units (the term given to estimated future electricity generation which cannot
be met by existing or planned capacity) the first step in calculating future emissions is the
determination of the amount of expected future generation from these units. Expected generation
from these units will equal the difference between expected demand and the amount of electricity
that will be generated by existing and planned capacity. The NOX emission rate for these units
should be assumed to equal the revised NSPS standards required by the CAAA. These units also
need to be sited; the assumptions which should be used are the same as those for planned units,
which were discussed in the previous paragraph.
3-13
-------
3.6 REFERENCES
1. U.S. Environmental Protection Agency. Procedures for Preparing Emissions Projections.
EPA-450/4-91-019 (NTIS PB91-242404). Office of Air Quality Planning and Standards.
Research Triangle Park, NC. July 1991.
2. U.S. Environmental Protection Agency. Procedures for the Preparation of Emission
Inventories for Carbon Monoxide and Precursors of Ozone, Volume II: Emission
Inventory Requirements for Photochemical Air Quality Simulation Models. EPA 450/4-
91-014 (NTIS PB91-216176). Office of Air Quality Planning and Standards. Research
Triangle Park, NC. May 1991.
3. U.S. Department of Commerce. BEA Regional Projections to 2040, Volume 1: States.
Bureau of Economic Analysis. Washington, DC. 1990.
4. U.S. Department of Commerce. BEA Regional Projections to 2040, Volume 2:
Metropolitan Statistical Areas. Bureau of Economic Analysis. Washington, DC. 1990.
5. U.S. Department of Commerce. BEA Regional Projections to 2040, Volume 3: BEA
Economic Areas. Bureau of Economic Analysis. Washington, DC. 1990.
6. U.S. Environmental Protection Agency. VMT Forecasting and Tracking Guidance,
Section 187. EPA (NTIS PB92-164961). Office of Mobile Sources. Ann Arbor, ML
January 1992.
3-14
-------
CHAPTER 4
NATIONAL AND REGIONAL ECONOMIC FORECASTS IN E-GAS
4.1 NATIONAL MACROECONOMIC MODELS
4.1.1 Overview
National macroeconomic models are used to forecast and simulate economic behavior at
the national level. These models are useful for predicting the level of future economic activity
for industries and consumers as well as explaining past economic behavior. In the public sector,
macroeconomic models are used to estimate the effects of potential and actual government
policies on the United States' economy. In the private sector, the models can be used to predict
future levels of demand for products, interest rates, and cost of factor inputs.
The E-GAS model allows the user to specify national macroeconomic forecasts. The
choice for national economic forecasts rather than a full national economic modeling capability
for E-GAS is explained in an earlier report.1 Although E-GAS does not contain options for
allowing users to develop their own forecasts using a national economic model, the REMI U.S.
model is embedded in E-GAS. This model is included because the regional REMI models need
forecasts of specific national economic indicators. The REMI U.S. model calibrates national
forecasts specified by the user to produce the outputs necessary to run the regional models. This
calibration is performed using an interface procedure developed by REMI to accommodate the
use of various national forecasts.
E-GAS is designed such that emission projection scenarios for each nonattainment area
and attainment portion of States can be made using a common assumption about future U.S.
economic activity. This chapter discusses these assumptions, or forecasts, and provides
information on available national forecasts, their characteristics, and the role and effects of the
national forecasts on final emission projections. In addition, this chapter discusses regional
economic models and their use in E-GAS.
The role of national forecasts in E-GAS is discussed in Section 4.1.2. Section 4.1.3
compares forecasts from a REMI model of Pittsburgh using different national forecasts. In
Section 4.1.4, selected national forecasts are discussed. In Section 4.1.5, available information
CH-93-09
4-1
-------
on the track record of forecasts is summarized. Section 4.1.6 contains a summary of the national
economic forecasts, while Section 4.1.7 discusses decisions regarding the choice of forecasts for
inclusion in E-GAS. Section 4.2. of this chapter focuses on regional economic models. Section
4.2.1 gives an overview of regional economic models. Section 4.2.2 discusses the REMI regional
economic models. Section 4.2.3 discusses the use of REMI models in E-GAS.
4.1.2 The Role of National Economic Forecasts in E-GAS
The use of a national macroeconomic model to drive a regional forecast model reflects
the inter-relationships between national and regional economies. National forecasts of final
demand provide estimates of national consumption demand; regional models capture the amount
of this consumption that is located in the regional economies and, perhaps more importantly, the
amount of demand that will be satisfied by each regional economy. The use of U.S. forecasts
provides consistency by assuring that regional final demands and supplies sum to the national
final demand and supply for goods and services.
The primary purpose of national economic forecasts in E-GAS is to provide a common
forecast with which to forecast regional economic growth. The nature of ozone formation
dictates that attention be paid to not only the level of economic activity, but also the location of
activity. A national forecast will provide an estimate of total economic activity. The regional
models will partition this activity between U.S. urban areas, States, and regions. The geographic
level of the regional forecasts will be dictated by the needs of the photochemical models used
by the ozone non-attainment areas.
The primary purpose of E-GAS is to project emission inventories for use in UAM and
ROM modeling, as well as Reasonable Further Progress inventories required by the C AAA. This
will require the use of emissions inventories, emission source growth projections, and estimates
of future emission controls. The inclusion of a national economic forecasting capability in E-
GAS allows EPA to forecast urban and regional growth under a common assumption about
national growth (i.e., GNP) and provides State users with the ability to simulate the effects of
different levels of national growth on ozone attainment regions.
CH-93-09 4_2
-------
4.1.3 The Effects of the Choice of National Model on Regional Forecasts
EPA will have the capability to develop ozone precursor emission projection scenarios
for each of the nonattainment areas using a common GNP assumption. This will ensure that the
levels of estimated future VOC and NOX emissions are not based on inconsistent assumptions
about each region's growth. The REMI U.S. model and interface procedure allow the EPA and
State users to base estimates of economic growth on GNP forecasts from respected economic
firms. State users of E-GAS may use their own assumptions about national economic activity
or may base their estimates of economic growth on forecasts from respected economic firms.
EPA may then compare baseline scenarios submitted by the States with baseline scenarios
developed by EPA. The results of a study using a REMI model of Pittsburgh suggest that the
choice of GNP scenario can significantly affect regional economic, and therefore, emission
forecasts.
In the REMI regional models, growth is affected by a number of factors, including the
performance of the national economy and the relative costs of doing business in the modeled
region. The relative costs of doing business are determined endogenously, although the user may
simulate policies which would affect the relative costs in a region. The growth or decline of the
national economy, however, is determined outside of the regional model. The choice of national
forecast is left solely to the user. This choice can have a large impact on the estimates of growth
in the region being modeled.
As part of a 1989 study performed at the University of Pittsburgh, a REMI model of
Pittsburgh was run using two forecasts, the Bureau of Labor Statistics (BLS) forecast and the
WEFA forecast.2 Although the BLS forecast is in part based on the WEFA national model, the
forecasts are based on different assumptions. The BLS forecast explicitly models national
economic cycles and includes in its forecast a recession in the early 1990s. The WEFA forecast
does not try to capture cycles in the national economy but instead uses a trend forecast. A
comparison of the BLS and WEFA forecasts is presented in Table 4-1.
Table 4-2 compares the Pittsburgh forecasts produced using the BLS and WEFA forecasts.
The WEFA and BLS forecasts of United States' manufacturing employment differed by 5
percent; this 5 percent difference resulted in a 10 percent difference in estimated manufacturing
CH-93-09
4-3
-------
TABLE 4-1. COMPARISON OF BLS AND WEFA AGGREGATE,
EMPLOYMENT FORECASTS FOR THE UNITED STATES, 1995
Employment in Thousands
Manufacturing
Durables
Nondurables
Nonmanufacturing
Mining
Construction
Transport and Public Utilities
Retail Trade
Finance, Insurance, and Real Estate
Wholesale Trade
Services
Agriculture, Forestry, and Fishing
Total Government
State and Local Government
Federal Government— Civilian
Federal Government— Military
Farm Employment
Total Employment
US BLS
1995
18,769
11,146
7,623
98,591
1,027
7,477
6,574
23,710
11,121
6,996
40,331
1,355
20,938
15,004
3,115
2,819
3,071
141,369
US WEFA
1995
19,704
11,542
8,162
103,487
1,019
8,314
6,588
24,222
11,746
7,148
43,005
1,445
21,686
15,788
3,096
2,802
3,071
147,948
Percent
Difference
5.0
3.6
7.1
5.0
-0.8
11.2
0.2
2.2
5.6
2.2
6.6
6.6
3.6
5.2
-0.6
-0.6
0.0
4.7
employment in Pittsburgh. Over all, differences in the national forecasts were magnified in the
forecasts of economic behavior in Pittsburgh. There are two important issues to note:
1.
2.
The BLS and WEFA forecasts both use the WEFA model as a basis for their projections
of U.S. economic activity. However, the use of different assumptions, including the
inclusion of business cycles hi one of the forecasts, resulted in an almost 5 percent
difference in the forecasts of total employment in 1995. The differences in the two
forecasts for the construction sector were over 11 percent2.
The almost 5 percent difference in estimates of total national employment was magnified
into a 7.6 percent difference in the estimates of total employment in Pittsburgh. The
sensitivity of the Pittsburgh estimates to the national estimates held for all sectors. The
national forecasts of manufacturing and non-manufacturing employment differed by
5 percent. The Pittsburgh forecasts for these sectors differed by 9.8 and 7.9 percent,
respectively.2
CH-93-09
4-4
-------
TABLE 4-2. COMPARISON OF BLS
FORECASTS FOR THE
AND WEFA AGGREGATE EMPLOYMENT
PITTSBURGH REGION, 1995
Manufacturing
Durables
Nondurables
Nonmanufacturing
Mining
Construction
Transport and Public Utilities
Retail Trade
Finance, Insurance, and Real Estate
Wholesale Trade
Services
Agriculture, Forestry, and Fishing
Total Government
State and Local Government
Federal Government— Civilian
Federal Government— Military
Farm Employment
Total Employment
PGH BLS
1995
129,658
85,343
44,315
913,441
6,518
61,316
57,912
224,329
82,331
66,567
409,846
4,621
129,756
95,970
20,070
13,716
8,431
1,181,285
PGH WEFA
1995
142,426
93,947
48,479
985,809
6,492
70,716
58,179
234,315
89,417
69,795
451,655
5,240
134,749
101,170
19,947
13,632
8,431
1471,414
Percent
Difference
9.8
10.1
9.4
7.9
-0.4
15.3
0.5
4.5
8.6
4.9
10.2
13.4
3.9
5.4
-0.6
-0.6
0.0
7.6
4.1.4 National Macroeconomic Forecasts
This section reviews forecasts from the Council of Economic Advisors to the President;
Data Resources, Incorporated; Research Seminar on Quantitative Economics (RSQE); REMI; and
Wharton Econometrics Forecasting Associates. The emphasis of this section is on national
economic forecasts rather than the national economic models which produce the forecasts.
However, the E-GAS model plan may be consulted for brief summaries of the modeling
techniques, input assumptions, and theoretical rationale of the national economic models used by
REMI; Data Resources, Incorporated; and the U.S. Bureau of Labor Statistics.1
CH-93-09
4-5
-------
4.1.4.1 The REMI U.S. Forecast
Regional Economic Models, Inc. has developed a U.S. model for use in a national and
regional economic forecasting and simulation framework. Regional models developed by REMI
include models for each of the 50 States, as well as sub-national (region, State, or sub-State area)
models as requested by clients. The REMI U.S. forecast is based on the BLS Trend-2000
forecast and will be referred to as the REMI/BLS forecast for the remainder of this report. The
BLS forecast also provides "fundamental information" for use in the REMI national and regional
models. This information includes historical and forecast data about technologies employed by
specific industries and the resulting "production recipe" (the type and amount of inputs) and inter-
industry relationships.3 The information on technology for each industry, which is implicit in the
production recipe, is contained in the U.S. input-output tables in the Trend-2000 forecast. These
input-output tables capture the inter-industry relationships in 1982 and 1986, and project the
relationships for 2000. The input-output tables are used to determine the technical coefficients
matrix for each industry. The technical coefficients represent the amount of intermediate goods
(e.g., products from other industries, fuel) required to produce a given amount of output from
each industry.
The methodology for projecting U.S. final demand by industry relies on the creation of
technical coefficients matrices for each historical and forecasted year. This methodology involves
developing an input-output model for the years for which BLS provides input-output accounts
(1982,1986, and 2000).3 The BLS forecasts include employment and output by industry, as well
as Gross National Product (GNP). The final demand components of the BLS forecast are used
to drive the input-output models, resulting in a prediction of intermediate demand for and output
by industries. The REMI national model may also take forecasts from other national economic
models to project industry-specific output estimates. The use of other forecasts to drive the
REMI national model will be discussed in a later section.
4.1.4.2 Council of Economic Advisors
The Council of Economic Advisors (CEA) was established by Congress through the
Employment Act of 1946 to provide economic advice and analysis to the President. Each year,
the CEA submits an annual report on the state of the U.S. economy; this report is contained in
CH-93-09 4_g
-------
the Economic Report of the President (ERP)4, which is delivered to Congress in February of each
year.
The CEA provides the Administration with forecasts of the major components of
economic growth. These projections "are not intended to be year-to-year forecasts; rather, they
are meant to reflect underlying economic trends and Administration policies".4 The forecasts
cover a five-year period and project growth or decline in real GNP, real compensation per hour,
output per hour (productivity), inflation, employment and unemployment, and are accompanied
by a short essay from the President and a report by CEA staff on economic issues of the past
year.
E-GAS users may formulate regional economic forecasts using GNP forecasts. The REMI
national forecast uses information from the Bureau of Labor Statistics' fifteen-year forecast. This
forecast serves as the default national forecast for the REMI regional models. Thus, the GNP
projections reported in the ERP could be included as an option for E-GAS users because the
REMI national-regional interface can calibrate the REMI national forecast to a user-specified
GNP. The ERP does not forecast final demand components, so the forecasts cannot be used to
provide a detailed alternative forecast for E-GAS users. The CEA projections are available to
E-GAS at no cost and could be updated annually. The CEA forecasts do not include other
variables required by E-GAS such as housing starts and energy prices; forecasted values of these
variables must be taken from other sources.
4.1.4.3 Data Resources, Inc. (DRI)
The DRI quarterly forecasts contain over 1200 variables. The forecasts include short- and
long-term forecasts. The long-term forecasts typically extend 15 years, but DRI will produce
longer forecasts at a client's request. Each forecast is released with an accompanying report
which explains the forecast assumptions and results for various sectors of the U.S. economy.
These forecasts may be purchased separately or may be received as part of a yearly subscription.
A subscription includes forecasts and publications, as well as client support and on-line access
to DRI economic databases.5
CH-93-09
4-7
-------
The DRI quarterly model houses the 1200 model equations in ten major sectors:
1. private domestic spending
2. production and income
3. government
4. international transactions
5. financial
6. inflation and productivity
7. supply
8. expectations
9. population
10. aggregates and miscellaneous
The forecasts are issued with an accompanying report which discusses the forecast results,
reviews the sector results, and provides tables detailing the sector forecasts. Three forecasts are
released: base case, low, and high forecasts.5
Housing and energy variables in the forecast include energy production, demand, taxes,
and price variables, and existing housing stock, start, and price variables.
The REMI models may be run using 92 forecasted variables from DRI. These 92
variables include 25 final-demand variables. Other variables hi the DRI forecasts which may be
used hi E-GAS include energy and housing variables. The DRI quarterly model forecasts nine
categories of energy variables including energy price, spending, and production variables. The
model also forecasts housing variables including housing starts, sales, stocks, and prices.
4.1.4.4 Research Seminar in Quantitative Economics (RSQE)
RSQE declined to participate hi supplying forecasts for E-GAS.
4.1.4.5 Wharton Econometric Forecasting Associates (WEFA)
The WEFA Group produces short- and long-term economic forecasts of U.S. economic
activity. The short-term forecasts range from 10 to 13 quarters (2.5 to 3.25 years) and are issued
monthly. The long-term forecasts are 25-year forecasts which are issued quarterly. In addition
to the baseline short-term forecast, the WEFA Group provides two alternative forecasts focusing
on macroeconomic risks and their probable effects on industries. The 25-year forecasts include
trend, cycle, and two alternative forecasts.6
The WEFA Group uses Mark 9, a quarterly economic model developed at WEFA, to
produce its short- and long-term forecasts. The model is comprised of over 1200 equations and
CH-93-09
4-8
-------
contains a "satellite" industry model which produces detailed industrial forecasts using outputs
from the core macroeconomic model.7 The Mark 9 model contains nine major sectors:
1. personal consumption expenditures 6. labor market
2. fixed investment 7. wages and prices
3. inventory investment 8. financial market
4. government 9. income
5. international trade
Variables in the model include consumption, investment, income, and inflation data from the
National Income and Product Accounts; population, employment, and wage rate data from the
BLS; industrial production data from the Federal Reserve Board; and demand, production, and
price data for the auto, housing, and energy sectors of the economy.7
The long-term economic forecasts are issued in a two-volume report. The first volume
of the report covers the trend or moderate growth scenario and contains an overview of the
forecast results and detailed sector reviews of the population, housing, investment, government,
inflation, labor market, industrial activity, and energy forecasts in addition to tables detailing the
sector forecasts.7
The REMI models may be run using 92 forecasted variables from WEFA. These 92
variables include 25 final demand variables. WEFA also forecasts housing and energy variables
which may be used in E-GAS development and simulations. Mark 9 forecasts detailed energy
price, supply, demand, and consumption variables. The model also forecasts housing variables
including housing starts, sales, stocks, and prices. A REMI interface for WEFA data has been
developed and tested.
4.1.5 Forecasting Records of the Models
Rating the track records of economic forecasters is difficult. There is no systematic
method for comparing the records of economic forecasters; the availability of published forecasts
varies and the forecasts often contain different variables.8'9 When forecast comparisons are
published, they often neglect the track record of newer or lesser-known forecasters. However,
published comparisons do exist; these will be summarized in this section.
CH-93-09
4-9
-------
In the effort to compare forecasts, each of the forecasters discussed in Section 4.1.4 was
contacted and a literature search of all journal articles on economic forecasting published between
1984 and 1991 was performed. DRI did not provide any materials on forecasting history. RSQE
declined to participate in providing information. WEFA provided a copy of a forecast for 1990
to 2005 along with the accompanying report provided to subscribers, but did not provide a
forecast history or materials comparing then- forecasts with other forecasts.
This section will focus on the forecasting records of DRI and WEFA for two reasons.
First, the other forecasts examined in Section 4.1.4 are not appropriate for inclusion in E-GAS
for reasons discussed in that section. Second, published comparisons on the track records of the
forecasters were found for DRI, WEFA, and CEA, but not for RSQE.
A 1987 study of the forecasting records of fifteen of the "best-known forecasters" ranked
the forecasters according to the percentage of absolute error (i.e., the absolute value of the
difference between the forecast and actual values as a percent of actual value) for four
commonly-forecast economic indicators: GNP, Consumer Price Index (CPI), unemployment rate,
and three-month Treasury Bill rate. The forecasters were rated for each year from 1983 to 1986
and were awarded an overall ranking based on the sum of the errors for the four indicators for
the four-year period. The fifteen forecasts ranked included forecasts from DRI and WEFA. Two
government forecasters, the Congressional Budget Office (CBO) (the Congressional economic
agency) and Office of Management and Budget (OMB) are included in the rankings.8 OMB, like
the Council of Economic Advisors, is controlled by the President and Executive Branch. The
forecast provided by CEA for the annual Economic Report of the President was not included in
the rankings, nor were forecasts from RSQE.
DRI and WEFA ranked first and second, respectively, while CBO ranked fourth and OMB
ranked fifteenth. The third-rated forecaster, DuPont, is not discussed in this report because it is
not on the list of potential forecasters for E-GAS. Although DuPont's track record for 1983-86
is strong, it is probably not appropriate to use an industry forecaster for E-GAS. The score for
DRI for the four-year period was 2.898, which translates into a total error of 290 percent for the
four indicators for the four years. This implies an average error of 18.1 percent per indicator per
year. The cumulative error for the WEFA forecast was 293 percent, implying an average error
of 18.3 percent per indicator per year.8 The difference between the average errors of the forecast
CH-93-09 4-10
-------
is one percent. The average error of the CBO and OMB forecasts was 19.6 and 34.0 percent,
respectively.
Another study, published in 1988, compared DRI, WEFA, and Chase Manhattan forecasts
of GNP and CPI for the years 1980 through 1984 (Chase Manhattan's forecasting division is not
discussed because it has merged with WEFA). WEFA had the lowest forecast error in 1981,
1982, and 1983 while DRI had the lowest forecast error in 1984. Again, the difference in the
forecast errors between WEFA and DRI was about one percent.10
Finally, a related study concerning the bias in government forecasts was examined. The
forecasts released by CEA, as mentioned in Section 4.1.4.2 of this report, are "not intended to
be year-to-year forecasts; rather, they are meant to reflect underlying economic trends and
Administration policies." As such, these forecasts are sometimes characterized as biased or
optimistic. Based on a statistical analysis of GNP, CPI, and unemployment forecasts from CEA
and CBO for the 1976 to 1987 period, it was concluded that the null hypothesis that the forecasts
are unbiased could not be rejected, i.e., the study did not find evidence that the forecasts were
biased.11
4.1.6 Summary
This chapter is intended to summarize the characteristics of macroeconomic forecasts from
CEA, DRI, RSQE, and WEFA. Table 4-3 summarizes the relevant information on the forecasts.
This table includes length of forecast, and variables included in the forecast. The REMI national
forecast, which uses information from the BLS 15-year forecast, is included in the table. This
forecast is referred to as the REMI/BLS forecast in Table 4-3.
The BLS forecast, from which the REMI national model extracts information for
developing a national forecast, is a 15-year forecast which is updated every 2 years. Because the
forecast is in the public domain, there is no cost for using it. The latest forecast was released
in November, 1991; this bi-annual forecast will not be updated by BLS until November, 1993.
REMI, however, updates its forecast by including national data from the Bureau of Economic
Analysis when data become available each year.
CH-93-09
4-11
-------
TABLE 4-3. SUMMARY OF ECONOMIC FORECASTS SURVEYED
Final Housing,
Number of Length of Proprietary Demand Energy
Forecast Release Forecasts Forecast Response Issues Variable Variables
REMI/BLS
CEA
DRI
RSQE
WEFA
'Can be extended
Bi-annual 1 15 years N/A N/A Y
Annual 1 5 years N/A N/A N
Quarterly 3 15 years' D Uncertain Y
N/A N/A N/A D N/A N/A
Monthly/ 4 25 years A No Y
Quarterly
at client's request.
N
N
Y
N/A
Y
The CEA forecast published in the Economic Report of the President extends five years
and, therefore, could be used by E-GAS users whose forecast horizons are five years or less. The
forecast does not contain final demand variables or housing and energy forecasts. The forecast
contains assumptions about the Administration's policies and then- effects and is not intended as
a "year-to-year forecast." The forecast's short forecasting period and purpose make it an
inappropriate choice for an alternative forecast for E-GAS. However, if CEA forecasts were
chosen as a national forecast for E-GAS, EPA could request forecasts for a longer time horizon.
CEA has developed a 40-year forecast of general economic indicators such as GNP and
productivity growth which could be used in E-GAS if the forecast were updated annually and
released to EPA.12 However, other characteristics of CEA forecasts (purpose, lack of final
demand variables) indicate that even if longer forecasts could be secured, CEA is not a good
choice for a national forecast to drive E-GAS.
The DRI forecast, though typically 15 years, may be extended at a client's request. The
forecast provides three scenarios (low growth, base case, and high growth) and contains housing
and energy variables which could be used in the growth factor tier of E-GAS. A yearly
subscription includes quarterly forecasts and reports, on-line access to DRI economic databases,
and client support. Finally, the participation of DRI as a supplier of national forecasts for E-GAS
CH-93-09
4-12
-------
may be considered a conflict of interest with other groups at DRI. Due to these problems, DRI
is not included in E-GAS.
The Research Seminar in Quantitative Economics at the University of Michigan was
contacted for information on their model and forecasts. The Director of RSQE declined to
participate in providing information or to be considered as a source for national forecasts for
E-GAS.
The WEFA long-term forecast extends 25 years and contains four scenarios: low growth,
base case, high growth, and cyclical growth. The cyclical growth forecast has recessions for
1991 and 1996 built into the forecast. The forecast contains final demand variables, as well as
housing and energy variables. The yearly subscription price is $18,200 and includes four
quarterly long-term forecasts and accompanying reports explaining the forecasts, monthly short-
term forecasts, on-line access to WEFA economic databases, two on-site presentations by WEFA
senior staff, an annual historical data book, and invitations to two of the four yearly U.S.
Economic Outlook Conferences. The use of WEFA forecasts in E-GAS will not involve
proprietary issues as WEFA allows subscribers to use purchased forecasts as a tool for analysis;
proprietary rights are a concern to WEFA only if the forecasts are being used by a client for
monetary gain.6
4.1.7 Conclusions
Sections 4.1.4 and 4.1.5 of this chapter summarized the characteristics and track record
of selected model forecasts. Summaries were developed from information provided by forecast
vendors, conversations with personnel at the forecasting firms, and journal articles discussing
forecasts and track records of the best-known forecasters. Based on this information, it is
concluded that the forecasts most appropriate for use in E-GAS are those provided by WEFA.
A first-generation E-GAS model was completed on September 30,1992. This model was
sent to States for testing, but did not include the individual regional economic models being
developed by REMI. The use of the REMI/BLS national forecasts in the first-generation model
was sufficient for testing purposes, although dummy data sets for other national forecast options
were included to allow the user to test the ease of use of the E-GAS national tier.
CH-93-09
4-13
-------
This version of the model will be used by States for Reasonable Further Progress
demonstrations and photochemical modeling requirements, as specified in the Clean Air Act
Amendments of 1990. The inclusion of an up-to-date respected national economic forecast will
allow States to derive the best possible estimates of national and regional economic activity and
emission estimates.
Section 4.1.3 of this chapter compared forecasted economic activity in the Pittsburgh area
using a REMI model of Pittsburgh with different national forecasts. The comparison showed that
differences in projected national economic activity lead to even larger differences in regional
economic forecasts. This suggests that the best available national forecast should be used to
drive the E-GAS modeling system to achieve the best estimates of future emission levels for non-
attainment areas and States in ROM regions.
Final information from DRI on conflict-of-interest concerns was not received, so a
complete comparison of DRI and WEFA forecasts could not be made. Comparing the forecast
outputs and track records suggests that the better forecast for E-GAS cannot be clearly
determined. Both forecasts have very good track records and both contain over 1000 variables,
including variables which may be used in the growth factor tier of E-GAS. However, the
confirmation from WEFA personnel that the use of its forecasts would not cause proprietary or
conflict-of-interest concerns resulted in the decision that WEFA forecasts should be used to drive
E-GAS.
4.2 REGIONAL ECONOMIC MODELS
4.2.1 Overview
Regional economic models were developed by REMI for use in the E-GAS system.
Models were developed for each of the nonattainment areas and remaining (attainment) portions
of the State, as well as for each State hi one of the ROM modeling regions. This detailed level
of geographic separation for economic activity is an important component of the E-GAS system,
as it will allow the user to distinguish between growth in a nonattainment area and growth in the
surrounding areas. Because the outputs of E-GAS will be used in ROM and UAM modeling, this
CH-93-09 4-14
-------
ability translates into a more precise breakout between growth in UAM modeling areas and
growth in ROM modeling areas, which is important in ozone formation and transportation issues.
This type of modeling system explicitly recognizes that while ozone formation is a local
phenomenon, ozone transport is a regional phenomenon.
The economic projections from the REMI models are used to estimate growth in physical
output of industries, fuel consumption, and VMT. The estimates of nonattainment area economic
growth, along with the estimates for the remaining portions of the State(s), will be used to
estimate fuel consumption and VMT for nonattainment and attainment areas. The existing fuel
consumption and VMT models do not project sub-State estimates. The REMI outputs will allow
sub-State estimates to be extracted from State estimates of fuel consumption and VMT.
Finally, the regional models are used to simulate the effects of policies in the non-
attainment areas on the surrounding area. A policy which increases the cost of doing business
in a nonattainment area will reduce economic activity in that area. Ozone concentrations in the
area, however, will be affected less if businesses move from the nonattainment area to locations
immediately outside the nonattainment area than if businesses were to leave the region. Although
the REMI models do not specifically trace industrial movements (i.e., though economic location
decisions are implicit in the models, the models do not capture movements of businesses from
one location to another), the user will be able to examine the relative costs of doing business for
each area, and will be able to determine the net effect of a policy on an area.
4.2.2 REMI Models
The REMI models were developed by Regional Economic Models, Inc. located in
Amherst, MA. The company was established in 1980 in response to the demand for regional
economic models for use in forecasting and policy simulation. The methodology used in building
the models pre-dates the establishment of REMI. In the mid-1970s, the methodology was
developed by George Treyz, Ann Friedlander, and Benjamin Stevens. The methodology, which
was named the TFS methodology after its authors, was implemented in 1977 in the Massachusetts
Economic Policy Analysis model, and has been used extensively since. REMI currently has
clients in over 20 States. They analyze a variety of policies including environmental,
transportation, energy, utility, and taxation policies. REMI recently developed a model for
CH-93-09
4-15
-------
California's South Coast Air Basin which was used to analyze the costs and benefits of the air
quality management plan for achieving federal and State air quality standards.
REMI models can be developed for any combination of counties and States in the United
States. The standard REMI economic-demographic model (the EDFS-14) forecasts supply and
demand for conditions for 14 sectors, 17 occupations, 25 final demand sectors, and 202
age/gender cohorts.13 In addition, employment estimates are produced for 210 sectors, and value
added and earnings are forecast for 14 sectors. With the addition of (purchased) input-output and
occupation matrices, value added can be estimated for 446 sectors and employment by 585
occupations.
In addition to forecasting, the REMI models are developed to allow the user to simulate
the effects of a policy change on an area. A large variety of economic and demographic policy
variables may be changed by the user, including almost 400 economic policy variables and over
70 demographic policy variables. The effects of these results can be determined by examining
the 664 economic and 849 demographic variables which are forecast by the REMI models.
While many regional models rely solely on regional data, the REMI models use regional
and national data in their model development. The use of national data provides a longer time
series of data and a larger set of data. The national data are used to construct national
econometric response functions which can be calibrated to each region based on regional data.
The philosophy behind this approach is summarized in a description of the TFS methodology:
...there is little reason to believe that economic units in one part of the country
have measurably different behavioral characteristics from those in another. The
differences among regions in their reactions to external event are substantial; but
they are mainly due to differences in industrial composition, regional purchase
coefficients and other variables which can be modelled, rather than to 'unique'
interregional differences in firm or household motivation and behavior.14
4.2.3 The Use of REMI Models in E-GAS
The REMI models are a key component of E-GAS. The inclusion of REMI models for
each nonattainment area, "rest-of-State" (i.e., surrounding attainment portion of the State), and
State in a ROM region provides distinct capabilities which can be used to assess emission-
CH-93-09 4-16
-------
producing activity in UAM and ROM modeling areas. The REMI models and outputs will
contribute to the development of credible growth factors for future-year UAM and ROM
inventories in the following ways:
Forecasts of emission-producing activities will be developed for both the attainment and
nonattainment portions of States, allowing growth rates to differ between the rural and
urban portions of a State.
Outputs from the REMI models will be used to produce State-level estimates of fuel
consumption and VMT, and regional (sub-State) estimates of physical output.
Information on the relative economic growth rates of attainment and nonattainment
portions of States will provide a basis for sharing State-level fuel consumption and VMT
estimates.
The effects of a policy implemented in a nonattainment area on the surrounding areas can
be assessed.
The effects of different GNP assumptions on nonattainment activity and emissions growth
can be determined.
Information on local policies (e.g., tax increase) can be entered directly into the REMI
model. This ability allows users to update forecasts based on new information.
Specific linkages between REMI outputs and the fuel choice, VMT, and physical output modules
include:
REMI models supply population forecasts to the fuel choice module for estimating
residential fuel consumption
REMI models supply population and personal income forecasts to the fuel choice module
for estimating commercial fuel consumption
REMI models supply forecasts of the relative costs of capital, labor, materials, and energy
to the fuel choice module for estimating industrial fuel consumption
REMI models provide industry-specific employment forecasts to the physical output
module for estimating physical output
CH-93-09
4-17
-------
4.3 REFERENCES
1. Lynch, T.M., T.R. Young, M.G. Smith, and E.S. Kimbrough. Design and Development
of the Economic Growth Analysis System (E-GAS). Paper presented at 2nd Annual
EPA/Air & Waste Management Association Specialty Conference, Emission Inventory
Issues and Progress, Durham, NC. October 19-22, 1992.
2. Giarratani, F. "Some Perspectives on the Value and Limitation of Long-Term Forecasts:
An Application of the Pittsburgh REMI Model", Journal of Geographic Analysis. April,
1991.
3. Shao, G., and G. Treyz. Building a U.S. and Regional Forecasting and Simulation
Model. Research Paper, Regional Economic Models, Inc. Amherst, MA. 1991.
4. Council of Economic Advisors. Economic Report of the President. Government Printing
Office (transmitted to the Congress). Washington, D.C. February, 1990.
5. Braman, Stuart, Data Resources, Incorporated. Telecon with Teresa Lynch, Alliance
Technologies Corporation. May, 1992.
6. Randall, Tony, Wharton Econometric Forecasting Associates. Telecon with Teresa Lynch,
Alliance Technologies Corporation. April, 1992.
7. Wharton Econometric Forecasting Associates. Mark 9 Model Reference. The WEFA
group. Bala Cynwyd, PA. January, 1990.
8. Wolf, Charles. "Scoring the Economic Forecasters", The Public Interest. National Affairs,
Inc. New York, NY. Summer, 1987.
9. Strongin, Steven and Paula S. Binkley. "A Policymakers' Guide to Economic Forecasts",
Economic Perspectives. Volume XII, Issue 6. Federal Reserve Bank of Chicago.
Chicago, IL. November/December, 1988.
10. Bretschneider, Stuart and Larry Schroeder. "Evaluation of Commercial Economic
Forecasts for Use in Local Government Budgeting", International Journal of Forecasting.
Volume 4, Issue 1. North Holland Press. Amsterdam, 1988.
11. Belongia, Michael T. "Are Economic Forecasts by Government Agencies Biased?
Accurate?", Federal Reserve Bank of St. Louis Review. Volume 70, Issue 6. Federal
Reserve Bank of St. Louis. St. Louis, MO. December, 1988.
12. Schmalensee, R, Council of Economic Advisors. "Long-Term Forecasts", memorandum
to Larry Jones, U.S. Environmental Protection Agency. June 4, 1991.
CH-93-09 4-18
-------
13. Regional Economic Models, Inc. Operator's Manual for a Single Region EDFS-14
Conjoined Forecasting and Simulation Model. REMI Reference Set, Volume 2.
Amherst, MA. 1991.
14. Treyz, G., and B. Stevens. "The TFS Regional Modeling Methodology", Regional
Studies, 19(6). 1985.
CH-93-09
-------
CHAPTER 5
ESTIMATING FUEL CHOICE IN E-GAS
5.1 INTRODUCTION
As part of the E-GAS model system, fuel choice modules were developed to provide
growth factors for emissions from energy consumption. The approach chosen for development
of the modules was based on a review of existing energy data and models, and the structure and
function of E-GAS. The structure of E-GAS includes economic models for each of the
nonattainment areas required to use photochemical modeling and for each of the States in the
ROM modeling region. The outputs from these models include detailed economic forecasts of
industrial and commercial employment and factor costs, and population. These factors are used
to estimate energy consumption in the residential, commercial, and industrial sectors.
Section 5.2 describes available fuel consumption data. Section 5.3 describes existing
energy models. Section 5.4 discusses options considered for the fuel choice portion of E-GAS.
Section 5.5 describes the methodology used to estimate fuel choice hi E-GAS.
5.2 FUEL CONSUMPTION DATA
5.2.1 Manufacturing Energy Consumption Survey
The Manufacturing Energy Consumption Survey1 (MECS) provides detailed energy
consumption data by industry for the United States and the four Census regions. MECS is a
triennial survey which began in 1985. Data for 1991 will be published soon. Most data hi
MECS are at the two-digit SIC level, although there are data for the energy-intensive four-digit
SIC categories. Data for 1985 did not cover smaller establishments. However, the level of detail
of the 1988 data is sufficient for estimating future energy consumption patterns. Finally,
although the 1988 MECS data have sufficient detail for analyzing energy choice and consumption
patterns, there are no historical values for these data. Because annual data are not available,
MECS could not be used in E-GAS.
5-1
-------
5.2.2 Annual Survey of Manufactures (ASM)
The Annual Survey of Manufactures1 (ASM) published figures on purchased fuels and
electric energy used for heat, power, and electricity generation for the years 1974 to 1981. The
industrial breakout includes fuel consumption data by four-digit SIC for 1974 to 1981. However,
these data do not include total (purchased and produced) consumption of fuels and electric energy
for heat, power, and electricity generation. Because emissions are related to energy consumption,
not amount of energy purchased, these data may not be useful in E-GAS development.
The ASM does not present a complete series of data because information on purchase of
specific fuels is not covered in the post-1981 period. The 1982 and 1987 Census of
Manufactures3 published data on purchased fuels. Data on the energy purchases are currently
being developed by staff at the U.S. Department of Energy's Energy Information Agency. These
data must be estimated for 1983, 1984, 1985, and 1986, which are the post-1981 years in which
neither MECS nor ASM was published. Because annual data are not available, ASM could not
be used in E-GAS.
5.2.3 National Energy Accounts
The National Energy Accounts4 (NEA), prepared by Jack Faucett Associates for the
U.S. Department of Commerce, report annual national time series data for 35 energy products
for 1958-1985. The NEA data use data from MECS and ASM. The 1985 MECS data were used
to allocate total four-digit SIC fuel expenditure data from ASM to specific fuels. This was done
by trending 1981 ASM fuel shares by four-digit SIC to 1985 based on trends in fuel shares at
the two-digit SIC level.4 Other energy consumption data sources that were used in the
construction of the NEA include the 1982 Census of Manufacturers3, the 1982 to 1985 ASM, the
1985 MECS, and the 1985 State Energy Price and Expenditure Report.5
NEA includes a complete data set for many categories of energy use for 1958-1985, and
is considered to be the best source for industrial energy-use data. For these reasons, NEA was
chosen to be a source of energy data for E-GAS.
5-2
-------
5.3 ENERGY MODELS REVIEWED
5.3.1 NAPAP Model Set
5.3.1.1 Industrial Regional Activity and Energy Demand (INRAD) Model
INRAD provides estimates of industrial electricity and fossil fuel demand. INRAD was
developed to predict how energy use will be influenced by fuel prices and the general level of
economic activity. The model also accounts for technological change. In two sectors, steel and
pulp and paper, the technological change estimates are specific to the industries. For the other
sectors, a general declining energy intensity (due to technological change) is employed.
Specifically, INRAD accounts for the increasing use of electric arc furnaces in the steel industry
and thermochemical pulping in the pulp and paper industry. The model uses the econometric
technique of seemingly unrelated regressions and uses national data from 1958 to 1985 to build
industry-specific equations for eight industrial categories. The eight categories include seven
industries and an eighth "other" category which aggregates the non-energy-intensive industries.
The seven industries modeled at the two-digit SIC level are food, textiles, paper, chemicals, glass,
glass products, and metals.6
The model estimates electricity and fossil fuel consumption in each sector as a function
of energy costs, capital, labor, materials cost, and capacity utilization in the industry. The model
predicts how energy consumption in the eight industrial sectors changes with changes to prices
in factor inputs. Simulated energy use from INRAD for 1985 was compared to 1985 data from
MECS. INRAD estimates were within five percent of actual electricity consumption and within
six percent of actual fossil fuel demand.6 The model's overall results are better than the industry-
specific results. For State- and urban-level modeling, the estimates of energy use will depend
significantly on the industrial composition of the area. If an area has large segments of an
industry for which INRAD may over- or underestimate fuel consumption, the model results could
be biased for that area. Modifications to the INRAD model which were made to improve the
model's ability to estimate industry-specific and State-level fuel consumption are discussed in
Section 5.5.
5-3
-------
5 3.1.2 Commercial Sector Energy Model by State (CSEMS)
The Commercial Sector Energy Model (CSEM) was chosen in the mid-1980s by Argonne
National Laboratory for use in the NAPAP model set. CSEM forecasts commercial energy
consumption by census region, and develops State estimates using a sharing algorithm. In 1989-
1990, ANL modified CSEM to directly forecast commercial energy consumption at the State
level; this version of the model was termed CSEMS. The CSEMS forecasts energy consumption
in the commercial sector for seven fuels.7
The input data to CSEMS include fuel prices, disposable personal income, and population.
Outputs include consumption of electricity, natural gas, distillate and residual fuel oil, and
liquefied petroleum gas (LPG) for warehouse, institution, office, hotel/motel, retail/wholesale, and
miscellaneous building types for three vintages of buildings. Kerosene, coal, and motor gas can
also be modeled but are usually omitted from model runs due to the insignificant amounts of
these fuels that are consumed in the commercial sector.7
5.3.1.3 Household Model of Energy by State (HOMES)
The Household Model of Energy (HOME) was chosen in the mid-1980s by Argonne
National Laboratory for use in the NAPAP model set. HOME forecasts residential energy
consumption for seven fuels by census region, and develops State estimates using a sharing
algorithm. In 1989-1990, ANL modified HOME to directly forecast residential energy
consumption at the State level; this version of the model was termed HOMES.8
The input data to HOMES include housing starts and income per household. Outputs
from HOMES include consumption of electricity, natural gas, distillate and residual fuel oil,
wood, and liquefied petroleum gas (LPG) for single and multi-family buildings end-use.
Kerosene, coal, and motor gas can also be modeled but are usually omitted from model runs due
to the insignificant amounts of these fuels that are consumed in the residential sector.8
5.3.2 REMI Model
5.3.2.1 Commercial and Industrial Fuel Use
The REMI model does not estimate commercial and industrial fuel consumption explicitly,
but uses fuel price in capital and labor factor demand equations. The price of fuel is a factor in
5-4
-------
thfe capital and labor demand equation in recognition that fuel, capital, and labor are inputs to the
production process, and that substitution possibilities exist among the three inputs. The optimal
level of fuel consumption is determined by the relative prices of capital, labor and fuel.
The REMI models calculate each region's costs of labor, capital, and fuel relative to the
entire United States. These relative costs are used to determine the amount of labor demanded
by each industry in a region. The relative costs of labor and capital, as well as anticipated
employment, are then used to estimate capital demands in a region. The demand for fuel is not
explicitly estimated. The cost of fuel, however, directly enters the labor demand equation and
indirectly enters the capital demand equations.
The "fuel price" used in REMI is a weighted average of the costs of natural gas,
electricity, and residual oil. The price of coal is not included in the fuel price except as it
indirectly affects the price of electricity.
E-GAS requires estimates of changing fuel consumption patterns in the commercial and
industrial sectors. While REMI provides information that may be useful in estimating future fuel
consumption, REMI does not provide estimates of commercial and industrial coal, gas, oil, and
electricity consumption. Because of this, the fuel consumption module must be developed
independently of REMI. A feedback loop between estimates of energy costs and consumption
with the REMI models could be included in E-GAS. This feedback would allow the user to
utilize the relationships between capital, labor, and fuel which are specified in the REMI models
and could improve the REMI estimates of labor and capital demand.
REMI outputs that may be useful for estimating industrial fuel consumption by fuel type
include relative (regional) costs of labor, capital, and fuel, as well as relative costs of total factor
inputs and intermediate inputs.
5.3.2.2 Residential Fuel Consumption
The REMI model produces estimates for one residential fuel category, fuel oil and coal.
These estimates are a function of two factors. The first factor that determines the amount of fuel
oil and coal consumption is the real disposable income of the region. The second factor is the
area's consumption of fuel oil and coal as a proportion of its real disposable income. This factor
is based on a consumer survey performed by the Department of Labor. The results of this survey
are published in the Consumer Expenditure Survey.9
5-5
-------
These estimates capture the income effect on residential oil and coal consumption and
may be a good indicator of the income effect on all household heating expenditures. However,
the REMI estimates cannot capture three other components of household energy demand:
substitution and conservation of fuels caused by a change in the price or relative price of a fuel;
change in heating efficiency as older heating units are replaced; and, the use of energy in the
household for purposes other than heating. Finally, because REMI models do not produce
estimates of residential electricity and natural gas consumption, the models are not sufficiently
detailed for use in E-GAS.
5.3.2.3 Transportation Fuel Consumption
The REMI model estimates consumption, in dollars, of gasoline and oil (motor). This
estimate is derived using the same basic methodology that is used to estimate residential fuel
consumption. The estimate is a function of two factors. The first factor that determines the
amount of gasoline and motor oil consumption is the real disposable income of the region. The
second factor is the area's consumption of gasoline and motor oil as a proportion of its real
disposable income. This factor is based on a consumer survey performed by the Department of
Labor. The results of this survey are published in the Consumer Expenditure Survey.9
The growth in real disposable income drives changes in gasoline and motor oil
consumption. Changes in real disposable income in a region depend on changes in population
and per capita real disposable income. Changes in motor oil and gasoline expenditures may be
a fairly good proxy of changes in VMT growth in the short-run. The guidance on projecting
VMT suggests that a time trend of VMT growth may be used to project VMT. REMI estimates
of gasoline and motor oil consumption capture changes in real disposable income and population,
both of which are related to VMT.
533 PC-Annual Energy Outlook (AEO) Model
53.3.1 Residential Fuel Consumption
The PC-AEO model uses the Residential Energy End-Use Model (REEM) to project
residential energy consumption for the Annual Energy Outlook.10 The REEM model projects fuel
consumption by Census region, type of service demand, type and vintage of residential structure,
5-6
-------
fuel, and year. Projected fuels include distillate oil, natural gas, LPG, electricity, kerosene, and
other fuels, and projections are made for each year through 2010.
There are four models within REEM; housing stock, service demand, service capacity, and
technology choice. The housing stock model uses new housing projections from Data Resources,
Incorporated and estimates annual housing starts by three housing types and three vintages for
each of the Census regions. The service demand model estimates the demand for energy
services. These services include heating and cooling, hot water, and refrigeration and appliances.
Service demands are modeled for each housing type, vintage and region. The service capacity
model calculates the amount of new and existing energy demand that must be met and the
technology choice model estimates the shares of each technology which will be used to meet this
demand.10
5.3.3.2 Commercial Fuel Consumption
The PC-AEO framework estimates commercial energy consumption using the Building
Energy End-Use Model (BEEM). BEEM is based on the Nonresidential Buildings Energy
Consumption Survey (NBECS-86), a 1986 Department of Energy survey on energy use in
commercial buildings. The model forecasts consumption of eight fuel types by year through
2010. The forecasted fuels are residual and distillate oil, natural gas, electricity, LPG, coal,
motor gasoline, and kerosene.
The four basic components of BEEM are models of building floorspace, service demand,
service capacity, and technology choice. The floorspace model is considered a key component
of BEEM because of the sensitivity of commercial energy consumption to the amount of
floorspace. The floorspace model projects new and existing floorspace by year for each of the
four Census regions.
The NBECS-86 includes data on total commercial consumption by fuel, measured in Btus,
and uses "conditional demand analysis" to estimate the amount of fuel that is consumed for each
energy service (e.g., heating) per foot of floorspace.
Final energy demand by region is calculated from estimated service demand, average
efficiency of each fuel used to meet service demand, fuel shares by type of service demand, and
price and employment effects. Data on macroeconomic variables, fuel prices and elasticities are
from the PC-AEO model.
5-7
-------
BEEM can run as part of the PC-AEO model or can serve as a stand-alone model.
Macroeconomic data and fuel prices would need to be fed to BEEM if it were to be run alone.
However, the usefulness of BEEM in E-GAS might be limited by the level of disaggregation
(Census region) of the output.10
5.3.3.3 Industrial Fuel Consumption
The PC-AEO industrial model estimates fuel consumption for eight industries. Five of
these industries are two-digit SIC industries and three industries are aggregations of two-digit SIC
industries. Fuels that are modeled include purchased electricity, natural gas, steam coal, residual
oil, and distillate oil. The model forecasts energy consumption at the national level through
2010.
The general form used to estimate fuel consumption specifies an industry's fuel
consumption as a function of output in the industry, the price of the fuel being modeled, and the
price(s) of competing fuel(s). The national equations are used to forecast Census region forecasts
by benchmarking the national equations to aggregations of State Energy Data Systems data.
Regional fud consumption can then be forecast using regional macroeconomic and price forecasts
provided by the PC-AEO macroeconomic model. The use of benchmarking assumes that
sensitivities of energy consumption to changes hi fuel prices and output are the same for all
regions.
5.3.4 ENERGY2020
One model that can simulate energy demand (and supply) at the sub-State level using
detuled economic inputs is the ENERGY2020 model. Early versions of this model were
de eloped for DOE. Total investment hi the model exceeds 250 experience-years of model
development and usage and $15,000,000 of model development and testing. A 1989 California
Energy Commission study concluded that ENERGY2020 was the best energy and planning
analysis model of the 26 models tested.11 A more detailed description of the ENERGY2020
model can be found hi the E-GAS preliminary model development plan.
The model is fairly large and a demonstration model was tested on two machines. First,
the model was run on a 386SX/20 without a math co-processor. The model reached solutions
5-8
-------
for all fuels and sectors for 1990-2000 in 95 minutes. The model was then run on a 486/33 with
a math co-processor and the model reached the final solutions in 35 minutes.
Though ENERGY2020 is a respected model and could be calibrated to analyze fuel choice
in the areas modeled in E-GAS, the model is very sophisticated and may be costly. Based on
this information, it was determined that the use of a purchased model had three drawbacks: the
cost would be higher than the cost associated with using the NAPAP models with an improved
version of INRAD; there will be an additional cost associated with learning the model; and the
model has not been peer-reviewed by EPA.
5.4 OPTIONS CONSIDERED FOR E-GAS FUEL CHOICE MODULE
In the preliminary E-GAS research plan, two general options were presented for the fuel
choice module. The first option was to modify and use an existing model set. The second
option was to build a fuel choice module by developing equations for each urban area to be
modeled in E-GAS. Both of these options were considered for the model plan. Section 5.3
presented reviews of the NAPAP models, ENERGY2020, and the fuel consumption modeling in
the REMI model. The criteria for ranking the models are the ability of the model to forecast at
the State or sub-State level; the input data required to run the models; the resources needed to
modify the model for use in E-GAS; and finally, an assessment of the quality of the model.
The models that met all of the above criteria are the NAPAP and ENERGY2020 models.
NAPAP and ENERGY2020 can forecast at the State level and could be modified to estimate sub-
State energy consumption. In addition, neither model requires technological inputs which may
be difficult for the user to estimate (e.g., future energy intensity of commercial floorspace). Both
models have strong theoretical frameworks that have been modified and improved since their
initial development. While both models could be used in E-GAS, NAPAP was deemed the better
choice because of cost, familiarity to the E-GAS team members, and its status as an EPA peer-
reviewed model. The other alternatives were rejected for a number of reasons.
Electric Power Research Institute (EPRI) models (i.e., Residential End-Use Energy
Planning System (REEPS), Commercial End-Use Energy Planning System, and Industrial End-
Use Planning Methodology - Econometric Models) and the related Electric and Gas Utility
Modeling System were reviewed, but were not considered appropriate for E-GAS for three
5-9
-------
reasons: (1) the models currently forecast at the NERC-region level; (2) the required inputs to
the commercial and residential models are technological in nature and may be difficult to
estimate for sub-national regions; and (3) the Electric and Gas Utility Modeling System was
never finalized. The PC-AEO models from DOE appear similar hi technique and structure to the
NAPAP models. However, the models forecast at the Census-region, rather than at State, level.
The REMI models are economic models which model energy costs and consumption of fuel oil
and natural gas hi the residential sector and motor oil and gasoline in the transportation sector.
The models, however, do not explicitly forecast commercial and industrial fuel use and therefore
could not be used in E-GAS.
The second approach introduced in the preliminary model plan was to develop MSA-level
equations for fuel consumption. However, data needs were researched and it was determined that
the data needed to estimate energy elasticities were not available. The Annual Survey of
Manufacturers has data for 1974-1981 on energy consumption by two-digit SIC for each MSA
in the United States, but the data set is not complete for the 1980s. In addition to incomplete
energy data, employment and value added may not be available for two-digit SIC codes for
MSAs. These data are collected, but are often suppressed due to plant disclosure concerns.
Even if complete energy and employment data were available or could be estimated for
each MSA and "rest-of-State" area, this approach would probably not be appropriate for the
E-GAS modeling system. E-GAS will include economic models and emission projection
capabilities for 28 ozone nonattainment areas, as well as each State in a ROM modeling area.
Development of energy consumption estimates for SICs 20 through 39 for just the 28
nonattainment areas would involve estimating 532 sets of equations. Each set of equations would
include consumption estimates for coal, oil, natural gas, electricity, and "other" fuel. Therefore
approximately 2600 equations would have to be developed to estimate fuel choice in the
nonattainment areas hi E-GAS. Estimates for each State and "rest-of-State" in the ROM
modeling region would also have to developed. This approach could not be completed within
the schedule and budget constraints of the project.
Finally, the idea of developing equations for each of the nonattainment areas is based on
the assumption that economic behavioral characteristics differ between regions of the United
States. The theoretical structure of the REMI models (as well as the NAPAP and PC-AEO
models) is based on the belief that behavioral characteristics are similar hi all regions of the
5-10
-------
country and that differences in regional economic factors, such as level of fuel consumption, are
based on attributes, not consumer behaviors, of the regional economy.
5.5 THE E-GAS FUEL CHOICE MODULE
Based on the information presented in Sections 5.1, 5.2, and 5.3, it was determined that
the NAPAP models would be the best option for estimating fuel consumption in E-GAS. It was
also determined that significant modifications would be made to the models in order to use them
in E-GAS. The following sections describe the modifications made to HOMES, CSEMS,
INRAD, and ARGUS during the development of E-GAS and the use of the models in E-GAS.
5.5.1 Modifications Made to CSEMS
Three major modifications were made to CSEMS in order to include it in E-GAS. First,
the model was re-coded to allow it to run hi an MS-DOS environment. Second, the model was
modified to predict commercial energy consumption at the sub-State level. Third, the base year
of the model was updated to 1990.
The original version of CSEMS was coded to run on a mainframe computer; it was then
modified to run in a UMX environment; finally, for E-GAS, the model was re-coded to allow
it to run hi an MS-DOS environment. The programming language used is C.
During the re-coding of the model, CSEMS was updated to forecast commercial energy
consumption growth for the modeling areas defined in E-GAS. In order to forecast sub-State
energy consumption, CSEMS was modified to accept input data from the REMI models. These
areas include nonattainment areas and attainment portions of States. Although the model was
re-coded, the approach used in the CSEMS model hi E-GAS is consistent with the model used
for the NAPAP assessments. This approach relies on forecasting three factors: commercial
floorspace, demand for commercial energy end-use services (e.g., air conditioning) and the
proportion of each fuel type which will be used to satisfy demand for an end use, and the
efficiency with which fuel will be used in commercial buildings. Forecasts of these three factors
are used to project consumption of energy, by fuel type, in the commercial sector.
5-11
-------
CSEMS produces sub-State commercial consumption estimates using State-level fuel
prices, and sub-State forecasts of population and disposable personal income from the REMI
models. CSEMS produces growth factors for electricity, coal, fuel oil (distillate and residual
oils), liquefied petroleum gas (LPG), motor gasoline, kerosene, and natural gas for nonattainment
areas and attainment portions of States. The growth factors for the fossil fuels are used by the
Crosswalk to grow the appropriate SCCs for each region. The growth factors for electricity are
sent to an electric utility model pre-processor, where they are used to develop an electric demand
growth factor using a weighted average of regional electricity demand growth for the residential,
commercial, and industrial sectors.
5.5.2 Modifications Made to HOMES
Three major modifications were made to HOMES to include it in E-GAS. These
modifications parallel the modifications made to CSEMS. The model was re-coded, modified
to predict residential energy consumption growth at the sub-State level, and the base year of the
model was updated to 1990.
The original version of HOMES was coded to run on a mainframe computer; it was then
modified to run in a UNIX environment; finally, for E-GAS, the model was re-coded to allow
the model to run in an MS-DOS environment. The programming language used is C. During
the re-coding of the model, HOMES was updated to forecast residential energy consumption for
the modeling areas defined in E-GAS. These areas include nonattainment areas and attainment
portions of States.
During the re-coding of HOMES, the approach used to estimate residential fuel
consumption remained consistent with the techniques used for the NAPAP assessments. This
approach relied on forecasting three factors: housing stock, demand for residential energy end-use
services (e.g., water heating) and the proportion of each fuel type which will be used to satisfy
demand for an end use, and the efficiency with which fuel will be used in residential housing
units. Forecasts of these three factors are used to project consumption of energy by fuel type in
the residential sector.
HOMES produces sub-State residential consumption estimates using State-level fuel
prices, and sub-State forecasts of household income from the REMI models. HOMES produces
5-12
-------
growth factors for electricity, coal, fuel oil, liquefied petroleum gas (LPG), natural gas, motor
gasoline, and wood for nonattainment areas and attainment portions of States. The growth factors
for the fossil fuels are used by the Crosswalk to grow the appropriate SCCs for each region. The
growth factors for electricity are sent to an electric utility model pre-processor where they are
used to develop an electric demand growth factor using a weighted average of regional electricity
demand growth for the residential, commercial, and industrial sectors.
5.5.3 Modifications Made to INRAD
Three modifications were made to INRAD for its use in E-GAS. First, the model, which
had been written to run in a UNIX environment, was re-coded to run in an MS-DOS
environment, the programming language used is C. Second, the model was modified to accept
sub-State level inputs from the REMI models and to forecast industrial fuel consumption growth
at the sub-State level. Third, equations for disaggregating fossil fuel consumption into coal, oil,
and natural gas consumption were added to the model.
5.53.1 Modifications to INRAD to Include Fossil Fuel Choice
The INRAD model is based on factor demand equations derived from the Generalized
Leontief (GL) flexible functional form. They are estimated for seven energy intensive industries.
Only electricity and total fossil fuel are forecast by INRAD. To support E-GAS, a second level
of hierarchical equations was developed to forecast fuel use by fuel type. These fuel share
equations can then be used in conjunction with INRAD to forecast growth in individual fuel
types. The general issues and methodology for the INRAD/E-GAS fuel share equations are
discussed in this section.
To maintain simplicity, equations are developed for three fuel types only: coal, total oil
(aggregated mostly from residual and distillate fuel oil use), and total gas (mostly natural gas,
but also including some LPG). As was the case for INRAD, national data series for the three
different fuel types were drawn from the National Energy Accounts for the years 1958-1985.
The same level of industrial disaggregation was used, i.e., SIC 20, 22, and 32, upstream
production sectors of SIC's 26, 28, and 33, and an "other" sector which includes all non-energy-
5-13
-------
intensive industries.8 An index of natural gas curtailments was also used to control for the gas
shortages in the early seventies.
INRAD predicts the level of fossil fuel use, given all factor prices. The fuel choice
equations give the predicted share of energy use by type. The methodology assumes the fossil
fuel choice follows a GL cost function. The GL approach generates a system of equations that
predict the i* share of fossil fuel use, f/F, as a function of the relative prices of each fuel type.
The form of the equations is:
/*'
where,
f; is the i* fuel
i represents the 3 fuels i = c,o, and g (coal, oil, and gas)
j represents the 3 fuels i = c,o, and g (coal, oil, and gas)
F is total fossil fuel use
Pj and PJ are the fuel prices
T is a technology trend
Z represents non-price influences
Py , TJ and Yi are parameters that are estimated, with Py representing the cross-
price elasticity of fuels i and j
TJ representing the technology elasticity of fuel i demand
YJ representing the non-price elasticity of fuel i demand
For forecasting, there are no non-price terms. The term TJ T is approximately an exponential
time trend so that Equation (1) can be re-written as:
'In INRAD, the aluminum sector was also a separate sector. Almost all energy use in that sector is electricity. The remainder is included
in "other" for purposes of forecasting fuel shares.
5-14
-------
«
Unfortunately, there is no guarantee that the predicted values for the fuel share will sum to one.
One approach is to drop one of the equations, e.g., the oil equation. The equation would then
be written:
+) (3)
F F
The other alternative is to renormalize the equations:
/, /,/*"
(4)
»i
i=l
Equation 4, while not based on any economic theory, is the recommended choice since it does
require that a particular fuel be singled out and treated differently. This is the approach used in
E-GAS.
Due to the paucity of State-level data on energy use by 2-digit SIC and due to the need
to aggregate the total energy use by fuel type to apply to the boiler SCC records, which may not
reliably identify the SIC code of the point source, a method to forecast regional fuel use by fuel
type and industry in 1990 and beyond is required.
Two alternatives to implementing these equations are available. The first is to use
regional price data to predict s—f/F (i=c,o,g) normalize using Equation (4), and apply these
predicted fuel shares to the State-level predictions of total fossil fuel use. The second choice is
to construct a new base year file of State-level fossil fuel use by type and SIC. The forecast for
any year t would then be:
where,
5-15
-------
(6)
and Pt /FM is based on the fossil fuel forecasts from the original INRAD equations. This
approach better accounts for State-level variations in fuel choice, but requires slightly more data
and programming. This approach is taken in E-GAS because the energy consumption numbers
need to be applied to benchmark consumption estimates developed for the nonattainment areas
and attainment portions of States. The model will be run using 1980 energy consumption data;
from these output, 1990 benchmark data will be estimated. The 1980 estimates of State-level
energy use by 2-digit SIC and fuel type are available from the Purchased Heat and Power
Systems (PURHAPS) model and database. Argonne National Laboratory will supply this
benchmark data for E-GAS. No other more recent data are available.1*
The growth rate equation used hi E-GAS is developed from re-writing (5) as:
(7)
where fifFt and s^ represent growth rates.
There are 1981 data available in PURHAPS and for the INRAD benchmark year. However, this year was the beginning of a severe recession
in many energy intensive sectors and would not be a good choice for a benchmark.
5-16
-------
5.6 REFERENCES
1. U.S. Department of Energy. Manufacturing Energy Consumption Survey. Washington,
DC. Various years.
2. U.S. Department of Commerce. Annual Survey of Manufactures. Washington, DC.
Various years.
3. U.S. Department of Commerce. Census of Manufactures. Washington, DC. Various
years.
4. Jack Faucett Associates. National Energy Accounts, 1958-1985. Prepared for the U.S.
Department of Commerce. Washington, D.C. February, 1989.
5. U.S. Department of Energy. State Energy Price and Expenditure Report. 1985.
6. Boyd, G.A., E.G. Kokkelenberg, and M.H. Ross. Sectoral Electricity and Fossil Fuel
Demand in U.S. Manufacturing: Development of the Industrial Regional Activity and
Energy Demand (INRAD) Model. Argonne National Laboratories. Argonne, IL.
February, 1990.
7. U.S. Department of Energy. Model Documentation: Commercial Sector Energy Model.
Prepared for the Energy Information Administration. DOE/EIA-0453. August, 1984.
8. Holte, J. A. Model Documentation: Household Model of Energy. Prepared for the U.S.
Department of Energy, Energy Information Administration. DOE/ELA-0409. February,
1984.
9. U.S. Department of Labor. Consumer Expenditure Survey. Bureau of Labor Statistics.
Washington, D.C. 1983.
10. U.S. Department of Energy. PC-AEO Forecasting Model for the Annual Energy Outlook
1990 (model documentation). Energy Information Administration. Washington, D.C.
March, 1990.
11. Systematic Solutions, Inc. Introduction to the ENERGY 2020 Model. Vandalia, Ohio.
October, 1991.
5-17
-------
CHAPTER 6
ESTIMATING PHYSICAL OUTPUT IN E-GAS
6.1 PHYSICAL OUTPUT: DEFINITION AND DATA SOURCES
EPA guidance suggests that, when possible, physical output be used to forecast emission
source growth.1 Physical output is measured using direct physical units such as tons of steel,
barrels of motor gasoline, numbers of computers, etc. When these data are not available, indexes
of physical production must be calculated. The ability to forecast physical output is particularly
important for VOC sources where the emissions are related to materials used in the production
process, such as surface coating operations. These emissions are directly related to the amount
of physical output produced and, therefore, growth in physical output is a better indicator of
emissions growth than value added, industrial earnings, or employment.
There are two ways that physical output is measured. The first is simply the direct
measure of actual physical output of an industry (e.g., tons of steel). The second method is
indirect and is used when direct measures data are not available. This measure is termed
"constant dollar output" and is calculated by converting value of shipments and inventory change
into constant dollars. The value of shipments added to an industry's inventory change over the
course of a year equals the value of goods produced in that year. These dollar output values are
then translated to dollar outputs for a base year (e.g., 1982) using price deflators (the ratios of
the price of the output in the current year divided by the price of the output in 1982) developed
for each year. These data series are usually termed industrial production or constant dollar
indexes.
The Survey of Current Business, which is considered a comprehensive source for physical
output data, compiles available physical output data on an annual basis.2 A sample of the
industries and products for which data are compiled is presented in Table 6-1. A second source
of physical output data is the Federal Reserve Board (FRB). The FRB has completed an index
of industrial production which contains data from 1977 to the present. This index is a
compilation of both actual physical output data and constant dollar indexes. The index is
constructed using data obtained from the Federal Reserve System, various government agencies,
trade associations, and the Bureau of Labor Statistics, compiled at the three- and four-digit SIC
6-1
-------
TABLE 6-1. SAMPLE OF PHYSICAL OUTPUT DATA
AVAILABLE FROM THE SURVEY OF
CURRENT BUSINESS
Transportation and Communication
Air Carriers
Domestic Operations
Scheduled service
Revenue passenger-miles (billions)
Cargo ton-miles (millions)
Mail ton-miles (millions)
Air Carriers - International Operations
Scheduled service
Revenue passenger-miles (billions)
Cargo ton-miles (millions)
Mail ton-miles (millions)
Urban Transit Industry - Passengers carried (millions)
Motor Carriers
Tonnage hauled (revenue) (millions of tons)
Railroads and Travel
Traffic
Revenue ton-miles (net) of freight (billions)
Chemicals and Allied Products
Chemicals
Inorganic Chemicals
Production (thousands of short tons)
Aluminum sulfate, commercial
Chlorine gas
Hydrochloric acid
Phosphorus, elemental
Sodium hydroxide
Sodium silicate, anhydrous
Sodium sulfate
Titanium dioxide, composite and pure
Sulfur, native (Frasch) and recovered (thousands of metric tons)
Production
Stocks (producers'), end of period
Inorganic Fertilizer Materials
Production
Ammonia, synthetic anhydrous
Ammonium nitrate original solution
Ammonium sulfate
Nitric acid
(continued)
6-2
-------
TABLE 6-1. SAMPLE OF PHYSICAL OUTPUT DATA
AVAILABLE FROM THE SURVEY OF
CURRENT BUSINESS (continued)
Nitrogen solutions
Phosphoric acid
Sulfuric acid
Superphosphate and other phosphate fertilizers
Production
Stocks, end of period
Potash sales
Imports
Ammonium nitrate
Ammonium sulfate
Potassium chloride
Sodium nitrate
Chemicals and Alcohol
Industrial Gases - Production (millions of cubic feet)
Acetylene
Hydrogen - high and low purity
Nitrogen - high and low purity
Oxygen - high and low purity
Organic Chemicals - Production (millions of pounds, except as noted)
Acetylsalicylic acid (aspirin)
Ethyl acetate
Formaldehyde
Glycerin - refined, all grades
Methanol synthetic (millions of tax gallons)
Phthalic anhydride
Ethyl Alcohol and Spirits (millions of tax gallons)
Production
Stocks, end of period
Alcohol, Plastics Materials, Paints, Varnish and Lacquer
Denatured Alcohol (millions of wine gallons)
Production
Consumption (withdrawals)
Total
For fuel use
Stocks, end of period
Plastics and Resin Materials, Production (millions of pounds)
Phenolic resins
Polyethylene and copolymers
(continued)
6-3
-------
TABLE 6-1. SAMPLE OF PHYSICAL OUTPUT DATA
AVAILABLE FROM THE SURVEY OF
CURRENT BUSINESS (continued)
Polypropylene
Polystyrene and copolymers
Polyvinyl chloride and copolymers
Paints, Varnish, and Lacquer, Shipments (millions of $), total
Food and Kindred Products; Tobacco
Alcoholic Beverages
Beer (millions of barrels)
Production
Distilled Spirits (millions of tax gallons, except as noted)
Total Production
Whisky
Production
Effervescent Wines (millions of wine gallons - 231 cubic inches)
Production
Still wines (millions of wine gallons)
Production
Distilling materials produced at wineries
Dairy Products (millions of pounds, except as noted)
Butter
Production (factory)
Cheese
Production (factory), total
Condensed and evaporated milk
Production, case goods
Fluid milk
Production on farms
Dry milk
Production
Grain and Grain Products
Barley (millions of bushels - 48 pounds, except as noted)
Production - crop estimate for the year
Corn (millions of bushels - 56 pounds, except as noted)
Production - crop estimate for the year, grain only
Oats (millions of bushels - 32 pounds, except as noted)
Production - crop estimate for the year
Rice (millions of pounds, except as noted)
Production - crop estimate for the year (millions of bags-100 Ib.)
Rye (millions of bushels - 56 pounds)
Production - crop estimate for the year
(continued)
6-4
-------
TABLE 6-1. SAMPLE OF PHYSICAL OUTPUT DATA
AVAILABLE FROM THE SURVEY OF
CURRENT BUSINESS (continued)
Wheat (millions of bushels - 60 pounds, except as noted)
Production (crop estimate for the year), total
Wheat Flour (thousands of sacks - 100 pounds, except as noted)
Production
Poultry (millions of pounds)
Slaughter
Eggs (millions of cases)
Production on farms
Cattle and Calves (thousands of animals)
Slaughter - federally inspected
Hogs (thousands of animals)
Slaughter, federally inspected
Sheep and Lambs (thousands of animals)
Slaughter, federally inspected
Tobacco (millions of pounds)
Leaf Production - crop estimate for year
Manufactured products
Leather and Products
Leather
Exports (thousands of square feet)
Footwear
Production (thousands of pairs), total
Lumber and Products
Lumber (all types) (millions of board ft.)
Production, total
Softwoods (millions of board ft.)
Douglas Fir
Production
Southern Pine
Production
Softwoods and Hardwood Flooring (millions of board ft.)
Softwoods
Western pine
Production
Hardwood Flooring - Oak flooring
Shipments
Metals and Manufactures
Iron and Steel (thousands of short tons, except as noted)
Iron and Steel Scrap
Production
(continued)
6-5
-------
TABLE 6-1. SAMPLE OF PHYSICAL OUTPUT DATA
AVAILABLE FROM THE SURVEY OF
CURRENT BUSINESS (continued)
Iron Ore - Operations in all U.S. Districts (thousands of long tons)
Mine production
Pig Iron
Production
Iron Products - castings (thousands of short tons)
Gray and ductile iron - shipments, total
Malleable iron - shipments, total
Steel, Raw and Semifinished (thousands of short tons)
Steel, raw - Production, total
Steel Products, Net Shipments - By Product (thousands of short tons)
Total (all grades)
Bars and tool steel, total
Sheets and strip, total
Aluminum (thousands of metric tons, except as noted)
Production, primary (from domestic and foreign areas)
Aluminum Products (millions of pounds)
Shipments
Copper (thousands of metric tons, except as noted)
Copper production
Copper-Base Mill and Foundry Products - Shipments (millions of pounds)
Brass mill (copper mill) products
Copper wire mill products (copper content)
Brass and bronze foundry products
Lead (thousands of metric tons, except as noted)
Production
Tin (metric tons)
Recovery from scrap (tin content), total
Zinc (thousands of metric tons, except as noted)
Mine production, recoverable zinc
Slab zinc
Production
Heating, Combustion, and Atmosphere Equipment - New orders
(domestic), net (millions of $), total
Industrial Supplies, Machinery, Equipment (1977 = 100)
Industrial suppliers distribution
Sales index, seas. adj.
Inflation index, not seas. adj.
Fluid Power Products, Shipments, (Index, 1985 = 100)
Hydraulic
Pneumatic
(continued)
6-6
-------
TABLE 6-1. SAMPLE OF PHYSICAL OUTPUT DATA
AVAILABLE FROM THE SURVEY OF
CURRENT BUSINESS (continued)
Tractors Used in Construction Industry, Shipments
Tracklaying
Units
Tractor shovel loaders (integral units), wheel and tracklaying
Electrical Equipment (thousands)
Batteries (automotive replacement type), shipments
Radio sets, total market, production
Television sets, total market production
Household major appliances, factory sales, total
Vacuum cleaners
Gas Equipment - Residential equipment sales
Furnaces (warm air)
Ranges
Water heaters (storage)
Pulp, Paper, and Paper Products
Pulpwood (thousands of cords - 128 cu. ft.)
Consumption
Waste Paper (thousands of short tons)
Consumption
Woodpulp (thousands of short tons)
Production, total
Paper and Board (thousands of short tons)
Production, All grades, total
Selected Types of Paper (thousands of short tons)
Groundwood paper, uncoated
Orders
Tissue paper - Production
Newsprint (thousands of metric tons, except as noted)
Production
Rubber and Rubber Products
Natural Rubber (thousands of metric tons)
Consumption
Synthetic Rubber (thousands of metric tons)
Production
Pneumatic Casings (thousands)
Production
Inner Tubes, Exports (thousands)
(continued)
6-7
-------
TABLE 6-1. SAMPLE OF PHYSICAL OUTPUT DATA
AVAILABLE FROM THE SURVEY OF
CURRENT BUSINESS (continued)
Stone, Clay, and Glass Products
Portland Cement - Shipments (thousands of barrels)
Clay Construction Products
Shipments
Brick (mil. of standard brick)
Structural tile, except facing (thousands of short tons)
Sewer pipe and fittings, vitrified (thousands of short tons)
Floor and wall tile and accessories (mil. of sq. ft.)
Glass Containers (thousands of gross)
Production
Gypsum and Products (thousands of short tons)
Production
Textile Products
Woven Fabrics, Finishing Plants (millions of linear yards)
Production, total
Cotton Fiber (thousands of running bales)
Production (ginnings)
Cotton Cloth
Broadwoven goods over 12 inches in width
Production (millions of sq. yards)
Manmade Fibers (millions of pounds)
Production
Acetate filament yarn
Staple, including tow (rayon)
Noncellulosic, exc. textile glass
Yarn and monofilaments
Staple, including tow
Textile glass fiber
Manmade Fiber Manufacturers
Production - Fabrics (broadwoven), manmade fiber (millions of square
yards)
Manmade fiber and silk fabrics, gray, total
Filament yarn (100%) fabrics, total
Spun yarn (100%) fabrics, total
Acetate filament and spun yarn fabrics
Wool and Manufactures
Production - Woolen and worsted woven goods (Mil. of sq. yds.)
Floor Coverings - Carpet, rugs, carpeting - shipments (mil. of sq. yds.)
Apparel (thousands of units, except as noted)
Women's
Men's
(continued)
6-8
-------
TABLE 6-1. SAMPLE OF PHYSICAL OUTPUT DATA
AVAILABLE FROM THE SURVEY OF
CURRENT BUSINESS (continued)
Transportation Equipment
Aircraft (complete) (millions of $)
Shipments
Value
„ Airframe weight (thous. of pounds)
Passenger Cars (new) (thousands of units, except as noted)
Trucks and Buses (new) (thousands of units)
Truck Trailers, New - Shipments (Number)
Trailers and chassis, Total complete units
Trailer bodies (containers) - Detachable, sold separately
Trailer chassis and running gear - Detachable, sold separately
Freight cars, new (excluding rebuilt), (Number)
6-9
-------
levels. The indexes have been updated to reflect the most recent available data.3
The FRB index uses direct physical measures where they are available. For industries
lacking comprehensive physical-product data, changes in physical output are estimated using
production indexes published every five years by the Census Bureau. These indexes are
constructed by converting dollar output values into constant dollar values, yielding comparable
indicators of physical production for 1977, 1982, and 1987. If, for example, the constant dollar
outputs for the chemical industry were 100 and 105 in 1982 and 1987, respectively, this means
that the amount of physical output was five percent greater in 1987 than in 1982.
In order to develop production estimates for 1978-81, 1983-86, and 1988-89, physical product
indexes were developed by FRB staff. These indexes are conceptually similar to the indexes of
production developed by the Census Bureau and use dollar output data from the Annual Survey
of Manufacturers and price deflators from the Bureau of Economic Analysis.3
In order to develop physical production indexes for the 250 individual manufacturing sectors, a
technique has been developed to estimate the proportion of value added that is attributable to
each of the individual sectors. This technique uses physical product data, industrial electricity
consumption, and total production-worker hours.3 Details of the technique are not discussed here.
For a detailed description of the technique see The Federal Reserve Bulletin, April, 1990.
6.2FORECASTING PHYSICAL OUTPUT
There are two general approaches that can be used to forecast physical output. The first method
correlates changes hi employment with changes in physical output. The second method correlates
changes in value added with changes in physical output. Both of these methods are discussed
in this section.
6.2.1 Forecasting Physical Output Using Employment Data
When physical output is forecast using employment data, three factors need to be
considered: (1) the number of workers, (2) the productivity of the labor force; and (3) the number
of direct physical units per dollar of material produced by an industry. The relationship between
physical output and these variables is given hi Equation 1:
6-10
-------
units. =
n
LABn x PRODn x
unit
n
LAB x PROD.
(1)
where:
unitsn = the physical output of industry n
LABn = the number of workers in industry n
PRODn = the productivity of workers in industry n, where
productivity is measured as dollar value added per
worker in industry n
(unit/LAB,, x PRODn) = units of physical output in industry n per dollar of
value added in industry n
The effect of the first of these factors, number of workers, is obvious: a change in the
amount of labor employed must lead to a change in the total value of an industry, otherwise
hiring decisions would not be rational. The second factor, productivity of the labor force,
accounts for the historical increase in the amount of output (typically measured in dollar value
added) per worker. The product of these factors, number of workers and value added per worker,
equals the value added of an industry. Both of these factors are relatively easy to forecast.
Forecasts of the number of workers in an industry are based on a number of identifiable factors;
most economic models produce forecasts of industrial employment. In the post-World War n
period, output per worker has increased, on average, about two percent per year, and this trend
is expected to continue.4 For the purposes of forecasting productivity increases, the forecasts by
the President's Council of Economic Advisors can be used. CEA estimates that annual
productivity will grow by 2.2 percent over the 1990-2030 time period, with component annual
increases of 2.8 and 1.7 percent for 1990-2010 and 2010-2030, respectively.4
The third factor affecting physical output is the number of physical units per dollar of
value added in an industry. Although physical output will be forecast at the three- and four-digit
SIC level where product classification is fairly narrow, there can still be considerable changes
in output over time which are not easily modeled. For example, SIC 357 is defined as office and
computing machines and includes computers, typewriters, calculators, and other office machines.
This sector experienced tremendous growth between 1977 and 1987 when value added more than
6-11
-------
doubled in the sector. However, the physical production index more than tripled during the same
period, which means that the unit output per dollar value added increased significantly in this
period. Rapid movements in output per dollar can be difficult to explain precisely, but there are
identifiable general factors which will cause such movements.
The number of physical units per dollar value added in SIC 357 will be affected by two
factors: the proportion of each product type to total production and the average price per product
type. In the example above, the proportion of product types is simply the percent that computers,
typewriters, calculators, and other machinery contribute to the value added of the sector. The
increase hi the units per dollar value (or, equivalently, the decrease in the price per unit) in the
1982-87 period could have been caused by an increasing percentage of less expensive items (e.g.,
calculators) being produced or a decrease in the price of one or more goods in the sector.
Changes in units per dollar value added are difficult to explain without a detailed examination
of the sector and may be very difficult to forecast. However, the general movement of units per
dollar value added may be captured with a time trend.
There are two types of physical output data. The first is the direct physical measure of
output (e.g., tons of steel) and the second is a physical production index, such as those
constructed by FRB. These two measures of physical output must be forecast using different
methods. The factors affecting physical measures of output were described in Equation 1. This
equation expresses physical output as a function of the number of workers in an industry, the
productivity of the workers, and units per dollar value added in an industry. Regression
equations can be developed from the relationship in Equation 2. This form is:
VA° Jl n\
: 1_ e n (2)
labor
6-12
-------
where:
laborn
VA,,
A
T
a,b
the number of workers in industry n
the value added in industry a
the intercept of the equation,
a time variable used to capture the change in physical output per dollar
value added
estimated parameters
Growth factors for physical output in industry "n" and year "t" can then be calculated
using Equation 3:
G'F
PROD,
(LAB - LAB, J X
,
PROD.
90
, b(T-1990)
(3)
where:
G-F.
A, a, and b
[(LABn,t - LABni90)*(PRODt / PROD*,)]
the physical output growth factor for
industry "n" in year "t"
the estimated parameters derived
from Equation 2
the change in productivity from 1990
to the year "t"
6.2.2 Forecasting Physical Output Using Value Added Data
To develop a physical output forecast using value added data, two factors are considered
directly: (1) value added; and (2) physical units per dollar value added. This approach differs
from forecasting physical output using employment data in one way. The use of employment
forecasts requires the use of productivity forecasts in order to develop value added forecasts.
This approach uses actual value added forecasts rather than constructing them from employment
and productivity forecasts. The relationship between physical output and value added is given
in Equation 4:
6-13
-------
units =
n
unit
VA x 1
VA
n
(4)
where:
unitsn = the physical output of industry n
VA,, = value added in industry n
The effects of changes in value added and units per value added were described in the
previous section which discussed forecasting value added with employment data.
Equation 4 expresses physical output as a function of value added and units of output
per unit of value added. In developing regression equations using value added output, an
additional variable, capacity utilization, was included. The addition of the capacity utilization
variable helps control for fluctuations in physical output and value which occurred in certain
industries in the 1970s and 1980s.5 The form of these regression equations can be described as:
"physical n
= (A x VAna x ebT x eBCu)
(5)
where:
A
VA.
T
CU
a,b,B
the intercept of the equation
value added in industry n
a time variable used to capture the change in physical output per dollar
value added
capacity utilization
estimated parameters
Growth factors for physical output in industry "n" and year "t" can then be calculated
using Equation 6:
(6)
6-14
-------
where:
G.F. (Optoma!,,,,') = the physical output growth factor for industry "n" in
year "t", where 1990 is the base year
a and b = the estimated parameters derived from Equation 5
(VA,,, / VA^)] = the growth in value added in industry n from 1990
to the year "t"
This growth factor does not include the capacity utilization factor. While capacity
utilization helps explain changes in physical output in the 1970s and 1980s, the REMI models
do not forecast capacity utilization. Therefore, the capacity utilization is assumed to remain
constant over the forecast period, i.e., the ratio of capacity utilization in any forecast year to
capacity utilization will always equal one and, therefore, will not affect the growth factor.
6.3 PHYSICAL OUTPUT IN E-GAS
6.3.1 Forecasting
The physical output module hi E-GAS uses value added to forecast physical output for
two reasons. First, EPA guidance suggests that value added be used to forecast physical output
when the data are available. For sectors for which physical output bridge equations are not
developed, growth in value added will be used as a proxy for growth in physical output. Thus,
the use of value added data in the bridge equations maintains consistency within the physical
output module. Second, the concurrent version of the REMI models, used when the E-GAS
model plan was being developed, only forecasts value added for 14 sectors. The structure of the
bridge equations which used employment data was developed during this time. However, the
generation of REMI models which are used in E-GAS include the capability of forecasting value
added for 210 sectors. For these reasons, E-GAS uses value added data to forecast physical
output.
Sectors for which physical output equations have been developed will also be forecast
using Equation (6). The development of equations for these sectors uses regression techniques
to define the parameters "a" and "b" for the specific sector. In the absence of sector-specific
parameters, a default value of one is used for "a" and a default factor of zero is used for "b."
6-15
-------
This results in physical output growth estimates which are proportional to growth in value added
in the sector.
6.3.2 Sources for Which Physical Output Equations Are Developed
To date, equations have been estimated for eleven VOC categories. The source used to
compile the series of physical production data is the Survey of Current Business.2 More sectors
will be estimated during the second generation E-GAS development phase. Problems were
encountered collecting appropriate capacity utilization data, so the results for the categories below
will undergo a rigorous review process.
Point sources for which physical output equations were developed include auto surface
coating, paper surface, and rubber and synthetic fibers. The area source categories for which
equations have been developed include petroleum refinery fugitives, surface coating of fabricated
metals, gasoline marketed, asphalt, auto surface coating, paper surface coating, rubber and
synthetic fibers, and general surface coating.
6-16
-------
6.4 REFERENCES
1. U.S. Environmental Protection Agency. Procedures for Preparing Emissions Projections.
EPA 450/4-91-019 (NTIS PB91-242404). Office of Air Quality Planning and Standards.
Research Triangle Park, NC. July, 1991.
2. U.S. Department of Commerce. Survey of Current Business. Washington, D.C. Various
years.
3. Federal Reserve Board. "Industrial Production: 1989 Development and Historical
Revision", Federal Reserve Bulletin. Washington, D.C. April, 1990.
4. Schmalensee, R, Council of Economic Advisors. "Long-Term Forecasts", memorandum
to Larry Jones, U.S. Environmental Protection Agency. June 4, 1991.
5. Boyd, G.A., E.G. Kokkelberg, and M.H. Ross. Sectoral Electricity and Fossil Fuel
Demand in U.S. Manufacturing: Development of the Industrial Regional Activity and
Energy Demand (INRAD) Model. Argonne National Laboratory. Argonne, IL. February,
1990.
6-17
-------
CHAPTER 7
METHODOLOGY USED TO FORECAST
VEHICLE MILES TRAVELED IN E-GAS
7.1 INTRODUCTION
A number of options for projecting activity for highway mobile sources were considered
for E-GAS. This chapter discusses these options and describes the approach used in E-GAS for
projecting mobile source activity. The primary activity measurement used for highway mobile
source inventory purposes is vehicle miles traveled (VMT) by road and vehicle type. Thus, the
projection methods studied during E-GAS development concentrated on estimating area- and
State-specific growth factors for VMT.
The options considered can be divided into two general groups: relatively simple trend-
based approaches; or more sophisticated models, which are comprised of detailed national-level
projections based on econometric methods, followed by allocation from the national to State
level.
7.2 TREND-BASED APPROACHES
The only available uniform national data source for VMT data that can be used in trend-
based approaches to VMT projections is the Highway Performance Monitoring System. Other
specific trend-based approaches can be used to develop area-specific VMT projections using
HPMS and other simple data sources. Approaches considered include: direct regression-based
projection of historic HPMS VMT for an area; bounding of historic VMT trend projections with
demographic and/or economic projections; and use of trends in indexes such as VMT per capita
and per unit of industrial output. The first option was included since it is the second method
(after local travel demand modeling) specified in the EPA projection guidance. The second
option represents an attempt to bound longer-term projections using the first option, since the
EPA guidance prescribes using only six years of HPMS data due to previous changes in HPMS
area coverage. The third option uses indices of VMT/population and VMT/economic activity to
7-1
-------
address concerns raised by observations that VMT growth rates have been increasing faster than
the growth in population or vehicle ownership over the past 40 years.
7.2.1 The Highway Performance Monitoring System
The HPMS is a large transportation data collection, analysis and reporting system which
is a cooperative effort involving State transportation agencies and the Federal Highway
Administration (FHWA). The purpose of the HPMS is to provide a procedure in which the
nation's functional system of highways is analyzed based on data annually sampled by all States.
HPMS is composed of two major components, data collection and analytical process, which are
described below. These descriptions are derived from two publications documenting HPMS
Version 2.1, the Technical Manual and the User's Guide.1
7.2.1.1 HPMS Data Collection
Three types of data are reported in the HPMS. First, universe mileage data include a
complete inventory of mileage classified by system, jurisdiction, and selected operational
characteristics. Second, sample data include specific inventory, condition, and operational data
obtained for the sample panels of highway sections. These data are expanded to represent the
universe of highway mileage, permitting evaluation of the performance of the various highway
systems. Finally, area-wide data are reported annually for rural, total small urban, and individual
urbanized areas. These are used in conjunction with universe and sample data. Area-wide data
consist of totals for mileage, daily vehicle miles of travel, accidents, local system data, land area,
population, and travel activity by vehicle type.
The HPMS is based on approximately 110,000 samples of functional system mileage.
Data collected from these samples or "sections" represent extensive information on pavement
attributes, geometries, traffic conditions, and operating characteristics. Seventy-eight attributes
are collected through HPMS. Some of the operating characteristics that are collected include:
functional system, type of facility, average annual daily traffic (AADT), future AADT (user
estimated), speed limit, peak capacity, K factor (design hour volume as a percentage of the
annual average daily traffic), percent commercial vehicles, signalization, green time, and peak
parking.
7-2
-------
Currently, States have two options in collecting HPMS data. Option one allows States
to aggregate or group data into three classifications: rural, small urban, and grouped urbanized
area. All urbanized areas are treated as an aggregate regardless of the number of individual
urbanized areas in the State. In this option, there are fewer total sections sampled; each
functional class and corresponding volume group is statistically represented at the statewide level
for each of the classifications. Data in option one are collected at an 80 percent confidence level
with a 10 percent estimate of allowable error for the following facility types: interstates, other
freeways and expressways, other principal arterials, minor arterials, and collectors.
The second option consists of sampling individual urbanized areas on a statistically valid
basis. Sampling individual urbanized areas requires that more sections be sampled. For States
with three or fewer urbanized areas, the design precision for all functional classes and volume
strata is an 80 percent confidence level at a 10 percent allowable error. For States with more
than three urbanized areas, the sampling precision is a 70 percent confidence level at a 15 percent
allowable error for minor arterials and collectors. For principal arterials and above, the sampling
precision represents an 80 percent confidence level with a 10 percent allowable error. For States
choosing this option, the following facility types are required to be sampled: interstates, other
freeways and expressways, other principal arterials, minor arterials, and collectors. Currently,
there are over 190 urbanized areas that are sampled on an individual basis under this approach.
Local roads are not sampled on a section-by-section basis. However, States are required
to submit aggregate summary area-wide tables for each individual urbanized area for mileage and
daily vehicle miles traveled (DVMT) for each functional class.
Requirements for sampling of individual urbanized areas are in the process of being
revised to accommodate needs for area-specific data. In addition, coordination of consistent
sampling approaches is being promoted for multi-State urbanized areas.
Estimates of DVMT by functional system are prepared for rural, small urban, and
individual urbanized areas of the State on an annual basis. These DVMT estimates are important
to the analyses of vehicle operating costs, travel time, fuel consumption, and emissions.
Development of HPMS estimates of highway travel by functional system are derived using
count-based traffic data. The procedures entail traffic counting one-third of the sample sections
and one-sixth of the non-sample interstate universe sections each year, and the application of
7-3
-------
correction factors, such as weekday/weekend and seasonal, to machine-generated counts, Growth
factors are applied to sections not counted in the current year.
7.2.1.2 HPMS Analytical Process
The HPMS analytical process provides information on future highway system conditions
based on the level of funding provided for capital improvements. The analytical process analyzes
data for each highway section and expands the results to represent each functional system. These
functional systems are as follows:
Rural Urban
Principal arterial - Interstate Principal arterial - Interstate
Other principal arterial Principal arterial - Other freeway
Minor arterial Other principal arterial
Major collector Minor arterial
Minor collector Collector
Local Local
The HPMS analytical process consists of six modules: (1) Needs, (2) Investments, (3)
Impact, (4) Composite Index, (5) Multiple Deficiency, and (6) Deferred Cost. All but the Impact
Module are concerned with transportation system, analysis and planning. The Impact analysis
module simulates the operation of vehicles on the highway and produces some key performance
measures, such as operating cost, fuel consumption, average overall travel speed, and emissions.
All can be calculated by functional system for rural and urban areas. The purpose of the impact
analysis is to provide comparison of vehicle performance measures under various scenarios.
These comparisons can be made among the target years for several scenarios or between a base
year and a target year for a particular scenario. The emissions component of this module is not
applicable to the contexts currently required by EPA SIP inventory guidance.
Thus, the major HPMS asset for emission inventory projection is the historical database
of VMT estimates available for all States and for approximately 400 individually reported areas
which include whole urban areas and fractions of multi-State urban areas. These data cover a
total of about 190 urban areas. Annual VMT data for these areas, broken down by functional
7-4
-------
road type and vehicle type, can be obtained in electronic media from the FHWA Office of
Highway Information Management.
7.3 TRENDS IN VMT INDEXES
With historic HPMS and Census data, it is possible to create regression-based
extrapolations of area-specific trends in car VMT per capita and truck VMT per total area
industrial or economic output. These indexes can then be used with area-specific population and
industrial/economic output from the REMI models to produce overall VMT projections. This
type of approach would have the advantage of directly considering the national trend illustrated
in Figures 7-1 and 7-2 while accounting for possible local variations. Existing detailed analyses
of these types of indexes have not been identified, so there may be some unanticipated aspects
or problems with applying this principle on a small scale which would not be revealed until VMT
indexes have actually been created for a variety of areas.
7.4 ECONOMETRIC APPROACHES
Many econometric and statistical or analytical approaches have been developed for the
projection of VMT at various levels and for different types of applications. These include a
number of computer-based modeling systems that are suitable for application in mobile source
emission inventory projection. Examples of such modeling systems include the ANL/NAPAP
developed Transportation Energy and Emissions Modeling System (TEEMS), the U.S. EPA
MOBILES Fuel Consumption Model, the Department of Energy's similar Highway Fuel
Consumption Model, the Alternative Motor Fuel Use Model developed by Oak Ridge National
Laboratory and the FHWA/Faucett VMT Forecasting Model.2'3'4-5
While it is beyond the scope of E-GAS to include a transportation demand model in the
system, many areas which use E-GAS may also use a computer-based VMT model. Descriptions
of some of these modeling systems were included in the E-GAS model plan.
7-5
-------
ON
-------
o
1950
1990
Figure 7-2. Historic VMT per capita.
-------
7.5 METHODOLOGY USED IN E-GAS
As described, EPA emission inventory guidance prefers using a travel demand model to
forecast VMT. This approach is not feasible in the E-GAS context, so the second option
mentioned by EPA is discussed here. This "historical area-wide VMT" method can be used if
a travel demand model cannot be made available, based on an ordinary least squares linear
regression extrapolation of the area's 1985 to 1990 HPMS-reported data. This method is only
valid for the years 1993 through 1995 in the CO nonattainment case to which this guidance
specifically applies. In addition, States may use "any reasonable methodology" to forecast VMT
for the portion of the VMT Tracking Area outside of the Federal Aid Urbanized Area.
The published draft EPA guidance does not directly address projection of VMT for
different vehicle types or road classes, but indicates that single overall VMT projections from
aggregate HPMS statistics are adequate for application across the entire mobile source inventory.
The MOBILE4.1 emission factor model contains a default set of vehicle-specific VMT mixes that
reflect the expected changes in the typical national distribution of VMT by vehicle type over
time. Areas using a locally-derived base year VMT mix would have to update this distribution
for future years. The vehicle types in HPMS and in MOBILE4.1-based inventories are not
directly compatible, as different vehicle descriptions are used in the truck and heavy-duty
categories and diesels are not separated in HPMS. It is possible to develop separate VMT
projection factors for light-duty and other vehicles by creating a simple split of HPMS VMT
totals into a light-duty category of passenger cars with other smaller vehicles and a second
category including buses, larger trucks and other vehicles. However, given the inclusion of a
default split for vehicle type in MOBILE4.1, it was determined that disaggregation of VMT was
not necessary.
EPA guidance requires that a travel demand model be used for projections submitted after
January 1, 1994, for projections beyond 1996.6 In determining an approach for E-GAS VMT
projections, the short period for which the projections would be used was considered. VMT
projections in E-GAS will be based on a bounded trend approach, with trends in VMT bounded
by population forecasts for nonattainment areas and attainment portions of States. The approach
used in E-GAS consists of bounding State-level VMT projections with population growth for the
nonattainment areas and States modeled. Specifically, State-level equations for VMT per capita
7-8
-------
were estimated as a function of time using State-level HPMS data for the years 1985-1990.
VMT estimates for an area are then developed using these equations and population data from
REMI.
The simple HPMS trend method is based on only six years of data and reliance on this
method beyond the cited five-year projection time frame could be problematic. However, EPA
guidance requires that areas which must forecast VMT beyond 1996 use a validated network
travel demand model.6 Areas which use the VMT projections from E-GAS may use the default
split of vehicle type in MOBILE4.1. MOBILE4.1 can be used to estimate mobile source
emissions using a single VMT projection. The model is designed to disaggregate the VMT
projection into VMT growth by vehicle type. In recognition that some areas which use E-GAS
will be using a travel demand model to estimate VMT growth and, therefore, will not use VMT
projections from E-GAS, the system is designed to allow the user to input VMT growth data by
vehicle type and by road type.
Use of ranges based on simple population growth as a bounding factor as described above
is problematic due to the documented continuing growth in rates of VMT per person.
Figures 7-1 and 7-2 illustrate national trends in population, car ownership, and car-related VMT.7
Figure 7-1 presents normalized trends of these three statistics, each indexed to a value of 100 in
1950. Relative growth hi numbers of vehicles and VMT has been consistently higher than
population growth through most of the period from 1950 to 1988, with some short-term
perturbations in VMT due to fuel and economic factors. Figure 7-2 illustrates the relatively
constant climb in VMT per capita, which began at 3,000 VMT/person in 1950 and is currently
above 8,000 VMT/person.
These trends indicate that bounding of HPMS-based projections for the longer term should
include allowance for the expected increases in VMT per capita. In E-GAS, this relationship
is based on State-specific indexes created from historic HPMS State-level VMT and State-level
population data. Projections are based on forecasting these indexes and using population growth
data from the REMI models in E-GAS. The VMT per capita equations developed for E-GAS
are presented in Table 7-1.
7-9
-------
TABLE 7-1. VMT EQUATIONS
VPAL = 8010.36 + 394.08 (TREND) VPMT
VPAR = 6758.61 + 343.59 (TREND) VPNE
VPAK = 7708.47 - 79.97 (TREND) VPNV
VPCA = 7662.96 + 187.04 (TREND) VPNH
VPCO = 8045.13 + 48.77 (TREND) VPNJ
VPCT = 7238 + 173.88 (TREND) VPNM
VPDE = 8610.79 + 220.31 (TREND) VPNY
VPDC = 5066.92 + 105.03 (TREND) VPNC
VPFL = 7319.77 + 224.1 (TREND) VPND
VPGA = 8401.56 + 445.65 (TREND) VPOH
VPID = 7080.29 + 346.65 (TREND) VPOK
VPIL = 5924.51 + 217.86 (TREND) VPOR
VPIN = 6574.36 + 585.26 (TREND) VPPA
VPIA = 662.34 + 261.27 (TREND) VPRI
VPKS = 7548.61 + 255.2 (TREND) VPSC
VPKY = 7304.94 + 287.89 (TREND) VPSD
VPLA = 6284.68 + 413.21 (TREND) VPTN
VPME = 7476.13-I-424.09 (TREND) VPTX
VPMD = 7496.68 + 163.31 (TREND) VPUT
VPMA = 6625.13 + 196.25 (TREND) VPVT
VPMI = 7317.22 + 256.35 (TREND) VPVA
VPMN = 7601.09 + 211.83 (TREND) WWV
VPMS = 6592.86 + 448.48 (TREND) VPWI
VPMO = 7337.42 + 410.59 (TREND) VPWY
8854.19 + 284.86 (TREND)
7351.04 + 251.63 (TREND)
8088.96 + 79.76 (TREND)
7386.43 + 291.46 (TREND)
6896.12 + 175.29 (TREND)
8684.87 + 342.40 (TREND)
4964.09 + 179.71 (TREND)
7693.71 + 301.62 (TREND)
7866.49 + 211.82 (TREND)
6974.83 + 161.90 (TREND)
9029.73 + 235.55 (TREND)
7791.85 + 281.04 (TREND)
6164.89 + 163.78 (TREND)
5390.5 + 237.13 (TREND)
7650.24 + 362.07 (TREND)
6326.21 + 224.56 (TREND)
7414.14 + 375.67 (TREND)
8569.06 + 164.86 (TREND)
6885.01 + 253.40 (TREND)
8267.79 + 366.16 (TREND)
8344.82 + 265.01 (TREND)
6054.05 + 398.97 (TREND)
7484.29 + 277.02 (TREND)
9831.15 + 473.34 (TREND)
Where: VP»» = VMT/capita for the state identified by **
e.g., VPAL = vmt for Alabama/population of Alabama
e.g., VPPA = vmt for Pennsylvania/population of Pennsylvania
TREND = time series variable, where 1 = 1985....6 = 1990
7-10
-------
7.6 RESULTS FROM THE NEW ENGLAND E-GAS MODEL
In order to test the methodology used in E-GAS to develop VMT estimates, the VMT
module was run using data from the New England REMI model. The New England REMf
model, which has a base year of 1990, was run using the REMI/BLS baseline national economic
forecast for 1991 to 2010. The model was run using the baseline assumptions for New England,
i.e., no policy changes were simulated. Population data from the baseline run were input to the
VMT module to estimate VMT growth for each of the eight areas in New England.
Table 7-2 summarizes the results of the model run. Results are presented for the Boston
nonattainment area (which includes Hillsborough County, NH); the sum of the Boston and
Springfield, MA nonattainment areas (which includes all counties in Massachusetts plus
Hillsborough County, NH); the attainment portion of New Hampshire (which includes all counties
in New Hampshire which are not part of the Boston or Portsmouth, NH nonattainment areas);
the States of Maine and Vermont; and the New England region (which includes every county in
New England except Fairfield and Litchfield Counties in Connecticut, which are part of the New
York City nonattainment area).
As the results indicate, VMT is expected to grow faster than population. While annual
population growth for each of the areas is less than.one percent, VMT is estimated to grow about
two to three percent annually over the 1990 to 2010 time period. Population forecasts for the
New England region indicate that population will grow, on average, less than one-half of one
percent between 1990 and 2010. Average annual VMT growth, however, is estimated at almost
three percent. These estimated increases in VMT per capita are consistent with the national trend
shown hi Figure 7-1.
The results presented are for five areas in New England plus the entire New England
region, which includes all New England counties except the two Connecticut counties which are
part of the New York City nonattainment area. It is interesting to note that VMT per capita is
expected to grow faster in the attainment areas, which are generally rural areas, than in the
Massachusetts nonattainment areas of Boston and Springfield. If these trends are validated, they
may need to be considered when developing long-term ozone control strategies.
7-11
-------
TABLE 7-2. RESULTS FROM THE NEW ENGLAND E-GAS MODEL
Area
Boston'
Massachusetts"
New
Hampshire0
Mained
Vermont11
Total New
England'
1990-2000
2.64
2.64
4.12
4.49
3.97
3.10
2000-2010
2.26
2.25
3.01
3.03
2.88
2.45
1990-2010
2.45
2.45
3.56
3.76
3.42
2.78
1990-2000
0.36
0.37
1.27
0.87
0.90
0.54
2000-2010
0.41
0.39
0.87
0.38
0.54
0.43
1990-2010
0.39
0.38
0.90
0.63
0.72
0.48
1990-2000
2.27
2.27
2.81
3.59
3.05
2.55
2000-2010
1.85
1.85
2.19
2.64
2.33
2.02
1990-2010
2.06
2.06
2.50
3.11
2.69
2.28
'Boston nonattainment area (includes Hillsborough County, NH)
'Boston and Springfield nonattainment areas (includes all counties in MA, plus Hillsborough County, NH)
'Attainment portion of New Hampshire (includes all counties not in Boston or Portsmouth, NH nonattainment areas)
"Entire state
'Includes all counties in New England, except Fairfield and Litchfield counties in Connecticut
-------
7.7 REFERENCES
1. Federal Highway Administration. Highway Performance Monitoring System Analytical
Process: Volume II, Version 2.1 Technical Manual, and Volume III, Version 2.1 Users
Guide. December 1987.
2. Saricks, C. L. The Transportation Energy and Emissions Modeling System: Selection
Process, Structure, and Capabilities. Argonne National Laboratory Report ANL/EES-
TM-295. November 1985.
3. Vyas, A.D. and Saricks, C. L. Implementation of the Transportation Energy and
Emissions Modeling System (TEEMS) in Forecasting Transportation Source Emissions for
the 1985 Assessment. Argonne National Laboratory Report ANL/EES-TM-321. October
1986.
4. National Acid Precipitation Assessment Program. Acidic Deposition: State of Science and
Technology — Report 26: Methods for Modeling Future Emissions and Control Costs
(Section 6.5). Washington, DC. December 1990.
5. Jack Faucett Associates. The FHWA/Faucett VMT Forecasting Model. Final Report
under Contract #DTFH61-86-C-00114 to the Federal Highway Administration, Office of
Policy Development. August 1988.
6. U.S. Environmental Protection Agency. Procedures for Preparing Emissions Projections.
EPA-450/4-91-019 (NTIS PB91-242404). Office of Air Quality Planning and Standards.
Research Triangle Park, NC. July 1991.
7. Wolcott, Mark. Personal communication on data compilation on population, vehicles and
VMT. U.S. Environmental Protection Agency Office of Mobile Sources. Ann Arbor, MI.
March 10, 1992.
7-13
-------
CHAPTER 8
E-GAS CROSSWALK
8.1 OVERVIEW
CROSSWALK, the final component in E-GAS, is a module which assigns growth factors
from other E-GAS modules to point, area, and mobile source SCCs for each county in each
E-GAS modeling region. The growth factors used by the CROSSWALK are those generated by
the models and modules in Tier 3, which were described in detail in Chapters 5, 6, and 7 of this
report.
E-GAS activity growth factors will ultimately be applied to inventories in the AIRS sub-
systems, AFS and AMS. The appropriate growth factor for each source in each inventory is
determined by information given in the SCC description. For example, SCC descriptions may
include the general economic sector associated with the emission (e.g., residential, commercial,
industrial); the specific economic sector associated with the emission (e.g., the chemical
industry); and the process associated with the emission (e.g., fuel combustion, solvent
application). The CROSSWALK was developed to associate each SCC hi AFS and AMS with
the appropriate growth factor from E-GAS based on similarities in E-GAS growth factors and
SCC definitions. Using CROSSWALK, energy, economic, and VMT growth factors are
converted to point, area, and mobile source SCCs growth factors for each county in each E-GAS
modeling region.
After growth factors have been calculated by models and modules in Tier 3, E-GAS
automatically calls CROSSWALK. CROSSWALK then connects each of the SCCs with its
matching E-GAS growth factor. A list of CROSSWALK matches and the justification for the
assignment is given in Appendix A. For example, the existing files for fossil fuels and electricity
will be converted to SCC growth factors based upon the fuel type and economic sector associated
with the emissions. After CROSSWALK has converted the growth factor files from E-GAS to
the appropriate SCC growth factors, the output is compiled in output files. CROSSWALK
creates separate output files for residential, commercial, industrial, and utility fuel consumption,
industrial physical output, and VMT. After the files are compiled, they are stored in the E-GAS
directory.
8-1
-------
8.2 SICS AND SCCS
SICs and SCCs are two distinct classification systems that were developed for different
purposes. SIC codes were developed by the U.S. Office of Management and Budget (OMB) for
use by the Department of Commerce (DOC). The DOC required a classification of
establishments by type of activity or primary business. The general activities include agriculture,
mining, construction, manufacturing, transportation, communication, electricity, gas, and sanitary
services.
Source Classification Codes were developed by the EPA Office of Air Quality Planning
and Standards (OAQPS) for use by EPA. EPA required a classification of processes within
establishments. OAQPS developed AIRS and its precursor, the National Emissions Database
System (NEDS). Since AIRS stores emissions inventories, it was necessary to classify process
information at the point-source level.
The following example illustrates the differences between SICs and SCCs. SICs are four-
digit codes which correspond to industrial categories.. The first digit corresponds to a division
of SICs. The first two digits together correspond to a major grouping, and each of the next two
digits further refine the SIC grouping.
2 = Manufacturing
28 = Chemicals and Allied Products
286 = Industrial Organic Chemicals
2865 = Includes over 100 different chemical processes including coal, tar, dyes
and organic pigments
SCC codes are either eight or ten digits in length. Point sources are classified using eight-
digit codes, while area and mobile sources are classified using ten-digit codes. The eight-digit
are hyphenated after the first, third, and sixth digits. The ten-digit codes are hyphenated after
the second, fourth, and seventh digits. Each set of digits between hyphens contains information
about the type of emission associated with the code. The following are examples of point and
area source codes and their format.
8-2
-------
Point source : 2 Internal combustion engine
2-01 Internal combustion electrical generation
2-01-002 Internal combustion electrical generation
using natural gas
2-01-002-01 Internal combustion electrical generation
using natural gas and a turbine engine
Area source : 21 Stationary source fuel combustion
21-01 Stationary source fuel combustion at electric
utilities
21-01-004 Stationary source fuel combustion at electric
utilities using distillate oil
21-01-004-002 Stationary source fuel combustion at electric
utilities using distillate oil and internal
combustion engine
The SCCs associated with the SIC above (SIC 28) are too numerous to list in this section.
However, some associations are possible as a brief illustration. A great number of the point
source SCCs beginning with a 3 are chemical manufacturing codes, and would correspond to SIC
28. Those SCCs for organic chemicals would correspond to SIC 286. A smaller number of
SCCs would correspond to SIC 2865.
8.3 FOSSIL FUELS
The growth factors for fossil fuel consumption are generated by HOMES, CSEMS, and
INRAD. The output files containing growth factor information are used as input files to
CROSSWALK. This section describes the specific information contained in the fuel consumption
input files to CROSSWALK.
8.3.1 Residential Fossil Fuels
Residential fossil fuel demands are generated by HOMES. Growth factors are developed
for each fuel type by year, State, and county. The fields in a HOMES input file to
CROSSWALK include:
8-3
-------
STATE
COUNTY
YEAR
FUEL TYPE
GROWTH FACTOR
The three-digit codes preceding the fuel type is the code used in E-GAS to identify the fuel. The
fuel type field can contain the fuels in the following list.
001 Coal
004 Distillate oil
004 Residual oil
004 Liquefied petroleum gas
006 Natural gas
009 Wood
CROSSWALK attaches the proper SCC to the records in the file by fuel type. The SCCs for
residential fossil fuels are the AMS codes 21-04-***-***.
8.3.2 Industrial Fossil Fuels
Industrial fossil fuel demands are generated by INRAD. The growth factors are by State,
county, year and fuel type. The fields in an INRAD input file to CROSSWALK are listed below:
STATE
COUNTY
YEAR
FUEL TYPE
INDUSTRY TYPE
GROWTH FACTOR
The fuel type field can contain the following fuels
001 Coal
004 Oil
006 Gas
009 Wood
099 Electricity
8-4
-------
The industry field contains one of the eight INRAD industrial categories. CROSSWALK
attaches the proper growth factor to each SCC using information on fuel and industry type
associated with the growth factor.
8.3.3 Commercial Fossil Fuels
Commercial fossil fuel demands are generated by CSEMS. The growth factors are
developed for each fuel type by State, county, and year. The fields in a CSEMS input file to
CROSSWALK are listed below:
STATE
COUNTY
YEAR
FUEL TYPE
GROWTH FACTOR
The fuel type field can contain the following fuels:
001 Anthracite coal
001 Bituminous/Subbituminous coal
001 Lignite coal
004 Distillate oil
004 Residual oil
004 Liquefied petroleum gas
006 Natural gas
CROSSWALK attaches the proper SCC to the records in the file by matching them to the fuel
type. The SCCs for commercial fossil fuels are the AMS codes 21-03-***-*** and AFS codes
1-03-***-** and 2-03-***-**.
8.3.4 Fossil Fuel Consumption at Utilities
Electric demands are generated by a two-step process. Initial residential, commercial, and
industrial demands are generated by HOMES, CSEMS, and INRAD respectively. The three
output files are read by an electric model preprocessor which prepares an input file for
8-5
-------
CROSSWALK. The growth factors are listed by fuel type, State, county, and year. The fields
in the input file to CROSSWALK are listed below:
STATE
COUNTY
YEAR
FUEL TYPE
GROWTH FACTOR
The fuel type field can contain one of the following fuels:
001 Anthracite coal
002 Bituminous coal
002 Subbituminous coal
003 Lignite
004 Residual oil
005 Distillate oil
006 Natural gas
007 Process gas
008 Coke
009 Wood/bark waste
010 Liquefied petroleum gas
Oil Bagasse
012 Solid waste
013 Liquid waste
014 Landfill gas
015 Kerosene/naphtha jet fuel
016 Geysers/geothermal
CROSSWALK attaches the proper SCC to the records in the file by matching them to the fuel
type.
8-6
-------
8.4 VMT ESTIMATES
Growth factors for vehicle miles traveled are generated by the VMT module. Growth
factors are listed by State, county, year, road, and vehicle type.
STATE
COUNTY
YEAR
ROAD TYPE
VEHICLE TYPE
GROWTH FACTOR
The vehicle type and road type fields correspond exactly to the numerous AMS highway mobile
source vehicle types and road types. CROSSWALK will attach SCC codes corresponding to the
proper road and vehicle type.
8.5 INDUSTRY-SPECIFIC PHYSICAL OUTPUT
The physical output module in E-GAS generates an input file for CROSSWALK which
contains physical output growth factors for 210 sectors. Each growth factor is listed by State,
county, BLS code, and year.
Below is the PHYSICAL OUTPUT input file to CROSSWALK:
STATE
COUNTY
YEAR
BLS CODE
GROWTH FACTOR
The BLS code field can contain any of the 210 BLS codes. CROSSWALK matches SCCs using
BLS codes.
8-7
-------
8.6 OTHER SCCs
In some instances, SCCs do not have a corresponding growth factor from E-GAS. For
example, area source codes 27-**-***-*** correspond to biogenic emissions; there are no outputs
from E-GAS which logically relate to future growth in emissions from these sources. For SCCs
for which a logical growth rate could not be determined from E-GAS outputs, the code will be
stored in the OTHER.SCC file and assigned a growth rate of 1.0, (i.e., no growth in emission-
generating activity) for all forecast years.
8.7 CROSSWALK FILES
CROSSWALK assigns SCCs as described above and creates a set of output files. These
files are in ASCII format and are automatically saved in the user directory. It is the user's
responsibility to maintain her/his file directory.
The output files all contain five fields : State, county, SCC, year, and growth factor.
Example output files are presented in Appendix B for the years 1993 through 1997. The format
of the files generated by CROSSWALK will be the same regardless of the sources included or
the years forecast. The example output files in Appendix B illustrate the format of the
CROSSWALK output files.
The output files are identified as follows:
RES_FUEL.SCC HOMES, residential fossil fuel
COM_FUEL.SCC CSEMS, commercial fossil fuel
IND_FUEL.SCC INRAD, industrial fossil fuel
ELECTRIC.SCC electric growth factors
VMT.SCC VMT, transportation
PHY.SCC PHYSICAL OUTPUT, industrial output
OTHER.SCC Growth for unclassified SCCs
CROSSWALK generates the seven files listed above. Examples of these files are
presented in Appendix A. After exiting E-GAS, the user can view and print these ASCII files
using any standard ASCII file reader. The files are stored in the E-GAS user's directory.
8-8
-------
The generic CROSSWALK output file format is as follows:
STATE numeric (2) State code
COUNTY numeric (3) county code
SCC numeric (10) SCC code
GROWTH FACTOR numeric (8)
The years corresponding to the growth factors will be listed at the top of the file. Figure 8-1
depicts CROSSWALK file handling capabilities and input and output file characteristics. The
fields hi the input and output files correspond to the fields described throughout this report.
8-9
-------
CROSSWALK
F6
COM_OTHR.DAT
COMMERCIAL FOSSIL
PJEL DEMAND
STATE
COUNTY
FUEL TYPE
YEAR
GROWTH FACTOR
F8
RES OTHR.DAT
RESIDENTIAL FOSSIL
FUEL DEMAND
STATE
COUNTY
FUEL TYPE
YEAR
GROWTH FACTOR
no
IND_OTHR.DAT
INDUSTRIAL FOSSIL
FUEL DEMAND
STATE
COUNTY
FUEL TYPE
INDUSTRY TYPE
YEAR
GROWTH FACTOR
F11
PHY_OUT.DAT
INDUSTRY-SPECIFIC
PHYSICAL OUTPUT
STATE
COUNTY
BLS CODE
YEAR
GROWTH FACTOR
F13
ELEC.DAT
ELECTRICAL GENE.WION
FOSSIL FUEL DEMAND
STATE
COUNTY
YEAR
FUEL TYPE
GROWTH FACTOR
VMT_OUT.DAT
SCC CROSS
REFERENCE LIS1
VMT ESTIMATES
STATE
COUNTY
YEAR
ROAD TYPE
VEHICLE TYPE
GROWTH FACTOR
F16
GROWTH FACTOR
FILES
STATE
COUNTY
SCC
YEAR
GROWTH FACTOR
Figure 8-1. CROSSWALK design.
8-10
-------
APPENDIX A
CROSSWALK FILES : BLS AND SCC MATCHES
A-l
-------
BLS
Code
SCC Code
Justification
Code
76
76
76
76
76
76
76
122
122
122
122
122
157
157
157
131
131
131
131
131
131
131
131
131
131
131
136
136
136
136
136
136
137
137
137
137
137
137
137
137
137
131
40201501
40201502
40201503
40201504
40201505
40201531
40201599
40201301
40201303
40201304
40201305
40201399
50190005
50190006
50190010
30100101
30100102
30100103
30100104
30100105
30100106
30100107
30100108
30100109
30100180
30100199
30100305
30100306
30100307
30100308
30100309
30100399
30100501
30100502
30100503
30100504
30100506
30100507
30100508
30100509
30100599
30100601
4
4
4
4
4
4
4
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-2
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30100603
30100604
30100605
30100699
30100701
30100702
30100703
30100704
30100705
30100706
30100707
30100708
30100709
30100799
30100801
30100802
30100803
30100804
30100805
30100899
30101101
30101198
30101199
30101202
30101203
30101204
30101205
30101206
30101299
30101901
30101902
30101904
30101905
30101906
30101907
30102101
30102102
30102103
30102104
30102105
30102106
30102107
30102108
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-3
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30102110
30102111
30102112
30102113
30102114
30102120
30102201
30102301
30102304
30102306
30102308
30102310
30102312
30102314
30102316
30102318
30102319
30102320
30102321
30102322
30102399
30103101
30103102
30103103
30103104
30103105
30103180
30103199
30103201
30103202
30103203
30103204
30103299
30103402
30103403
30103404
30103405
30103406
30103410
30103411
30103412
30103414
30103415
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-4
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30103420
30103425
30103430
30103435
30103499
30103501
30103502
30103503
30103506
30103507
30103510
30103515
30103520
30103550
30103551
30103552
30103553
30103554
30103599
30103801
30103901
30103902
30103903
30104201
30104202
30104203
30104204
30104301
30107001
30107002
30109101
30109105
30109110
30109151
30109152
30109153
30109154
30109180
30109199
30110002
30110003
30110004
30110005
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-5
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30110080
30110099
30111103
30111199
30111201
30111202
30111299
30112001
30112002
30112005
30112006
30112007
30112011
30112012
30112013
30112014
30112017
30112021
30112031
30112032
30112033
30112034
30112037
30112099
30112199
30112401
30112402
30112403
30112404
30112405
30112406
30112407
30112480
30112501
30112502
30112504
30112505
30112506
30112509
30112510
30112511
30112512
30112514
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-6
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
136
136
131
131
131
30112515
30112520
30112521
30112522
30112524
30112525
30112526
30112527
30112528
30112529
30112530
30112531
30112532
30112533
30112534
30112535
30112540
30112541
30112542
30112543
30112544
30112545
30112546
30112547
30112550
30112551
30112552
30112553
30112555
30112599
30112699
30112701
30112702
30112703
30112720
30112730
30112740
30112780
30113004
30113005
30113201
30113205
30113210
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-7
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30113221
30113222
30113223
30113224
30113227
30113299
30113301
30113302
30113303
30113380
30113701
30113710
30113799
30114001
30114002
30114003
30114004
30114005
30115201
30115301
30115310
30115311
30115312
30115320
30115321
30115322
30115380
30115601
30115602
30115603
30115604
30115605
30115606
30115607
30115680
30115701
30115702
30115703
30115704
30115780
30115801
30115802
30115803
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC conclation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-8
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30115821
30115822
30115880
30116701
30116702
30116703
30116704
30116780
30116799
30116901
30116902
30116903
30116904
30116905
30116906
30116980
30117401
30117402
30117410
30117411
30117421
30117480
30117601
30117610
30117611
30117612
30117613
30117614
30117615
30117616
30117617
30117618
30117630
30117631
30117632
30117633
30117634
30117680
30118101
30118102
30118103
30118104
30118105
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-9
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30118106
30118107
30118108
30118109
30118110
30118180
30119001
30119002
30119003
30119004
30119010
30119011
30119012
30119013
30119014
30119080
30119501
30119502
30119503
30119504
30119505
30119506
30119580
30119701
30119705
30119706
30119707
30119708
30119709
30119710
30119741
30119742
30119743
30119744
30119745
30119749
30119799
30120201
30120202
30120203
30120204
30120205
30120206
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-10
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30120210
30120211
30120280
30120501
30120502
30120503
30120504
30120505
30120506
30120507
30120508
30120509
30120520
30120521
30120522
30120523
30120524
30120525
30120526
30120527
30120528
30120529
30120530
30120531
30120532
30120540
30120541
30120542
30120543
30120544
30120545
30120546
30120547
30120548
30120549
30120550
30120551
30120552
30120553
30120554
30120555
30120580
30120601
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-ll
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30120602
30120603
30120680
30121001
30121002
30121003
30121004
30121005
30121006
30121007
30121008
30121009
30121010
30121080
30121101
30121102
30121103
30121104
30121121
30121122
30121123
30121124
30121125
30121180
30125001
30125002
30125003
30125004
30125005
30125010
30125015
30125020
30125025
30125099
30125101
30125102
30125103
30125104
30125180
30125201
30125301
30125302
30125305
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-12
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30125306
30125315
30125316
30125325
30125326
30125380
30125401
30125405
30125406
30125407
30125408
30125409
30125410
30125411
30125412
30125413
30125415
30125416
30125417
30125418
30125420
30125499
30125801
30125802
30125803
30125805
30125806
30125807
30125810
30125815
30125816
30125817
30125880
30125899
30130101
30130102
30130103
30130104
30130105
30130106
30130107
30130108
30130110
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-13
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
30130180
30130201
30130202
30130203
30130280
30130301
30130302
30130303
30130304
30130305
30130380
30130401
30130402
30130403
30130404
30130405
30130480
30130501
30130502
30130503
30130504
30130505
30130580
30180001
30181001
30182001
30182002
30182003
30183001
30184001
30187001
30187002
30187003
30187004
30187005
30187006
30187007
30187008
30187009
30187010
30187097
30187098
30187501
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC pan of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-14
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
30187502
30187597
30187598
30188501
30188502
30188503
30188504
30188505
30188599
30188801
30188802
30188803
30188804
30188805
30101801
30101802
30101803
30101805
30101807
30101808
30101809
30101810
30101811
30101812
30101813
30101814
30101815
30101816
30101817
30101818
30101819
30101820
30101821
30101822
30101827
30101832
30101837
30101838
30101839
30101840
30101842
30101847
30101849
4
4
4
4
4
4
4
4
4
4
4
4
4
4
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-15
-------
BLS
Code
SCC Code
Justification
Code
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
30101852
30101860
30101861
30101863
30101864
30101865
30101866
30101870
30101871
30101872
30101880
30101881
30101882
30101883
30101884
30101885
30101890
30101891
30101892
30101893
30101894
30101899
30102401
30102402
30102403
30102404
30102405
30102406
30102407
30102408
30102409
30102410
30102411
30102412
30102413
30102414
30102415
30102416
30102417
30102418
30102419
30102421
30102422
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-16
-------
BLS
Code
SCC Code
Justification
Code
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
132
133
133
133
133
133
133
133
133
133
133
133
133
133
134
134
134
134
135
30102423
30102424
30102425
30102426
30102427
30102499
30102501
30102505
30102506
30102599
30102601
30102602
30102608
30102609
30102610
30102611
30102612
30102613
30102614
30102615
30102616
30102617
30102625
30102630
30102699
30106001
30106002
30106003
30106004
30106005
30106006
30106007
30106008
30106009
30106010
30106011
30106012
30106099
30100901
30100902
30100910
30100999
30101401
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-17
-------
BLS
Code
SCC Code
Justification
Code
135
135
135
135
135
135
135
135
135
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
30101402
30101403
30101404
30101499
30101501
30101502
30101503
30101505
30101599
30103301
30103311
30103312
30103399
30101301
30101302
30101303
30101304
30101399
30101601
30101602
30101603
30101699
30101702
30101703
30101704
30101705
30101706
30101707
30101708
30101799
30102701
30102704
30102705
30102706
30102707
30102708
30102709
30102710
30102711
30102712
30102713
30102714
30102717
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-18
-------
BLS
Code
SCC Code
Justification
Code
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
30102718
30102720
30102721
30102722
30102723
30102724
30102725
30102727
30102728
30102729
30102730
30102801
30102803
30102804
30102805
30102806
30102807
30102820
30102821
30102822
30102823
30102824
30102825
30102903
30102904
30102905
30102906
30102907
30102908
30102920
30102921
30102922
30102923
30102924
30102925
30103001
30103002
30103003
30103004
30103020
30103021
30103022
30103023
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-19
-------
BLS
Code
SCC Code
Justification
Code
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
136
137
137
137
137
137
137
137
137
137
137
137
137
137
137
137
137
137
137
137
137
137
137
104
104
104
104
30103024
30103025
30103099
30104001
30104002
30104003
30104004
30104005
30104006
30104007
30104008
30104009
30104010
30104011
30104012
30104013
30104501
30101011
30101012
30101013
30101014
30101015
30101021
30101022
30101023
30101030
30101099
30102001
30102002
30102003
30102004
30102005
30102099
30104101
30104102
30104103
30104104
30104199
30105001
30202001
30202002
30202101
30202102
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
3
3
3
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-20
-------
BLS
Code
SCC Code
Justification
Code
104
104
105
105
104
104
104
104
104
104
104
104
104
2
2
2
2
2
2
2
2
2
106
107
2
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
30202105
30202106
30203001
30203099
30203801
2805000000
2805001000
2805001001
2805005000
2805005001
2805010001
2805015000
2805015001
2801000001
2801000002
2801000003
2801000004
2801000005
2801000006
2801000007
2801000008
2801500000
2801520000
2801600000
30202801
30200401
30200402
30200403
30200404
30200410
30200499
30200501
30200502
30200503
30200504
30200505
30200506
30200507
30200508
30200509
30200510
30200511
30200512
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
2
2
2
2
4
4
4
5
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-21
-------
BLS
Code
SCC Code
Justification
Code
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
2
2
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
30200513
30200601
30200602
30200603
30200604
30200605
30200606
30200607
30200608
30200609
30200610
30200611
30200699
30202201
30202601
30203103
30203104
30203105
30203106
30203107
30203108
30203109
30203110
30203111
30299998
30299999
30200101
30200102
30200103
30200104
30200199
30200712
30200713
30200714
30200721
30200722
30200723
30200724
30200730
30200731
30200732
30200733
30200734
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
4
4
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-22
-------
BLS
Code
SCC Code
Justification
Code
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
160
2
2
2
2
30200741
30200742
30200743
30200744
30200745
30200751
30200752
30200753
30200754
30200755
30200756
30200760
30200771
30200772
30200773
30200774
30200781
30200782
30200783
30200784
30200785
30200786
30200787
30200788
30200789
30200790
30200791
30200799
30200801
30200802
30200803
30200804
30200805
30200806
30200815
30200816
30200899
30201999
30203601
30288801
30288802
30288803
30288804
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-23
-------
BLS
Code
SCC Code
Justification
Code
2
112
112
112
112
112
112
112
104
107
109
109
109
109
109
109
109
109
109
109
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
112
112
112
112
112
108
30288805
30201201
30201202
30201203
30201204
30201205
30201206
30201299
30201301
30201401
30201501
30201599
30201601
30201699
30201701
30201702
30201703
30201704
30201799
30201899
30200901
30200902
30200903
30200904
30200905
30200998
30200999
30201001
30201002
30201003
30201004
30201099
30201103
30201104
30201105
30201106
30201199
30200201
30200202
30200203
30200299
30200301
30203201
4
2
2
2
2
2
2
2
1
2
1
1
1
1
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-24
-------
BLS
Code
SCC Code
Justification
Code
108
108
113
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
30203202
30203299
30203399
30501001
30501002
30501003
30501004
30501005
30501006
30501007
30501008
30501009
30501010
30501011
30501012
30501013
30501014
30501015
30501016
30501017
30501021
30501022
30501023
30501024
30501030
30501031
30501032
30501033
30501034
30501035
30501036
30501037
30501038
30501039
30501040
30501041
30501042
30501043
30501044
30501045
30501046
30501047
30501048
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-25
-------
BLS
Code
SCC Code
Justification
Code
7
7
7
7
10
10
10
10
10
10
41
41
41
41
41
41
41
41
41
41
41
41
41
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
30501049
30501050
30501090
30501099
2325000000
2325010000
2325020000
2325030000
2325040000
2325050000
30501101
30501106
30501107
30501108
30501109
30501110
30501111
30501112
30501113
30501114
30501115
30501120
30501199
30501901
30501902
30501903
30501904
30501905
30501906
30501907
30501999
30502001
30502002
30502003
30502004
30502005
30502006
30502007
30502008
30502009
30502010
30502011
30502012
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-26
-------
BLS
Code
SCC Code
Justification
Code
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
42
42
42
42
42
42
42
42
42
42
42
42
42
30502013
30502014
30502015
30502016
30502020
30502099
30502101
30502102
30502103
30502104
30502105
30502106
30502201
30502299
30502401
30502499
30502501
30502502
30502503
30502504
30502505
30502506
30502507
30502508
30502509
30502510
30502511
30502599
30502601
30502699
30503099
30503101
30503102
30503103
30503104
30503105
30503106
30503107
30503108
30503109
30503110
30503111
30503199
1
1
1
1
1
1
2
2
2
2
2
2
1
1
2
2
1
1
1
1
1
1
1
1
1
1
1
1
2
2
1
3
3
3
3
3
3
3
3
3
3
3
3
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-27
-------
BLS
Code
SCC Code
Justification
Code
42
42
42
42
42
42
42
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
44
44
44
44
44
44
44
42
42
42
42
42
42
42
41
41
41
41
30503201
30503202
30503203
30503204
30503205
30503206
30503299
30503301
30504001
30504002
30504003
30504010
30504020
30504021
30504022
30504023
30504024
30504025
30504030
30504031
30504032
30504033
30504034
30504036
30504099
30500401
30500402
30500403
30500404
30500405
30500406
30500499
30500501
30500502
30500503
30500504
30500505
30500598
30500599
30500606
30500607
30500608
30500609
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-28
-------
BLS
Code
SCC Code
Justification
Code
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
30500610
30500611
30500612
30500613
30500614
30500615
30500616
30500617
30500618
30500619
30500699
30500706
30500707
30500708
30500709
30500710
30500711
30500712
30500714
30500715
30500716
30500717
30500718
30500719
30500799
30501501
30501502
30501503
30501504
30501505
30501506
30501507
30501508
30501509
30501510
30501511
30501512
30501513
30501514
30501515
30501516
30501517
30501518
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-29
-------
BLS
Code
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
41
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
42
42
42
42
42
SCC Code
30501519
30501520
30501521
30501522
30501599
30501601
30501602
30501603
30501604
30501605
30501606
30501607
30501608
30501609
30501610
30501611
30501612
30501613
30501614
30501615
30501616
30501617
30501699
30501401
30501402
30501403
30501404
30501406
30501407
30501408
30501410
30501411
30501412
30501413
30501414
30501415
30501416
30501499
30500301
30500302
30500303
30500304
30500307
Justification
Code
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-30
-------
BLS
Code
SCC Code
Justification
Code
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
30500308
30500309
30500310
30500311
30500312
30500313
30500314
30500315
30500316
30500318
30500321
30500398
30500399
30500801
30500802
30500803
30500899
30500901
30500902
30500903
30500904
30500905
30500906
30500907
30500908
30500909
30500910
30500915
30500916
30500917
30500999
30501201
30501202
30501203
30501204
30501205
30501206
30501207
30501208
30501209
30501211
30501212
30501213
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
4
4
4
4
4
4
4
4
4
4
4
4
4
4
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-31
-------
BLS
Code
SCC Code
Justification
Code
42
42
132
132
132
132
132
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
136
41
7
7
10
10
42
10
10
10
42
42
136
41
7
7
10
10
42
30501214
30501215
30501221
30501222
30501223
30501224
30501299
30501301
30501399
30501701
30501702
30501703
30501704
30501705
30501799
30501801
30501899
30510001
30510002
30510003
30510004
30510005
30510006
30510007
30510101
30510102
30510103
30510104
30510105
30510106
30510107
30510108
30510196
30510197
30510198
30510199
30510201
30510202
30510203
30510204
30510205
30510206
30510207
2
2
2
2
2
2
2
3
3
1
1
1
1
1
1
2
2
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-32
-------
BLS
Code
SCC Code
Justification
Code
10
10
10
42
42
136
41
7
7
10
10
42
10
10
10
42
42
136
41
7
7
10
10
42
10
10
10
42
42
136
41
7
7
10
10
42
10
10
10
42
42
42
42
30510208
30510296
30510297
30510298
30510299
30510301
30510302
30510303
30510304
30510305
30510306
30510307
30510308
30510396
30510397
30510398
30510399
30510401
30510402
30510403
30510404
30510405
30510406
30510407
30510408
30510496
30510497
30510498
30510499
30510501
30510502
30510503
30510504
30510505
30510506
30510507
30510508
30510596
30510597
30510598
30510599
30515001
30515002
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-33
-------
BLS
Code
42
42
42
42
42
42
42
42
42
139
139
139
139
139
139
139
139
139
139
139
139
139
139
139
139
139
139
139
139
139
138
138
138
138
138
138
138
138
138
138
138
138
138
SCC Code
30515003
30515004
30515005
30588801
30588802
30588803
30588804
30588805
30599999
30500101
30500102
30500103
30500104
30500105
30500110
30500111
30500112
30500113
30500198
30500201
30500202
30500203
30500204
30500205
30500206
30500207
30500208
30500209
30500211
30500299
30600101
30600102
30600103
30600104
30600105
30600106
30600107
30600108
30600111
30600199
30600201
30600202
30600301
Justification
Code
2
2
2
4
4
4
4
4
4
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-34
-------
BLS
Code
SCC Code
Justification
Code
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
30600401
30600402
30600503
30600504
30600505
30600506
30600602
30600603
30600701
30600702
30600801
30600802
30600803
30600804
30600805
30600806
30600807
30600811
30600812
30600813
30600814
30600815
30600816
30600817
30600818
30600819
30600820
30600821
30600822
30600901
30600902
30600903
30600904
30600905
30600999
30601001
30601101
30601201
30601401
30601402
30601599
30609901
30609902
1
1
4
4
4
4
1
1
1
1
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC pan of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-35
-------
BLS
Code
138
138
138
138
138
138
138
138
138
138
138
34
34
34
34
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
31
31
31
31
31
31
31
31
31
31
31
SCC Code
30609903
30609904
30609905
30610001
30688801
30688802
30688803
30688804
30688805
30699998
30699999
30700501
30700597
30700598
30700599
30700701
30700702
30700703
30700704
30700705
30700706
30700707
30700708
30700709
30700711
30700712
30700713
30700714
30700715
30700716
30700798
30700799
30700801
30700802
30700803
30700804
30700805
30700806
30700807
30700808
30700896
30700897
30700898
Justification
Code
4
4
4
4
4
4
4
4
4
4
4
2
2
2
2
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-36
-------
BLS
Code
SCC Code
Justification
Code
31
34
34
34
34
34
34
39
37
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
120
122
122
30700899
30703001
30703002
30703096
30703097
30703098
30703099
30702098
30702099
30700101
30700102
30700103
30700104
30700105
30700106
30700107
30700108
30700109
30700110
30700199
30700203
30700211
30700212
30700213
30700214
30700215
30700221
30700222
30700223
30700231
30700232
30700233
30700234
30700299
30700301
30700302
30700303
30700304
30700401
30700402
30700499
30701199
30701301
1
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-37
-------
BLS
Code
SCC Code
Justification
Code
122
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
141
142
142
142
142
142
142
142
142
142
142
142
49
49
52
52
52
52
52
52
52
52
52
54
30701399
30800101
30800102
30800103
30800104
30800105
30800106
30800107
30800108
30800109
30800110
30800120
30800121
30800122
30800123
30800197
30800198
30800199
30800501
30800699
30800701
30800702
30800703
30800704
30800705
30800720
30800721
30800722
30800723
30800724
30800799
30902099
30902501
30903004
30903005
30903006
30903099
30988801
30988802
30988803
30988804
30988805
30900198
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
1
1
4
4
4
4
4
4
4
4
4
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-38
-------
BLS
Code
SCC Code
Justification
Code
54
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
8
8
8
8
8
8
30900199
30900201
30900202
30900203
30900204
30900205
30900206
30900207
30900208
30900298
30900299
30900301
30900302
30900303
30900304
30901001
30901097
30901098
30901099
30901101
30901102
30901103
30901104
30901199
30901501
30901601
30901602
30901603
30901604
30901605
30901606
30901607
30904001
30904010
30904020
30906001
30906099
31000101
31000102
31000103
31000104
31000105
31000199
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-39
-------
BLS
Code
SCC Code
Justification
Code
8
8
8
8
8
8
8
8
8
8
8
74
74
74
74
74
74
84
182
182
399
399
100
100
100
194
194
144
144
144
114
114
114
114
114
114
114
141
141
141
141
141
141
31000201
31000202
31000203
31000204
31000205
31000206
31000207
31000299
31088801
31088802
31088803
31307001
31307002
31390001
31390002
31390003
31399999
31400901
31401001
31401002
31401101
31401102
31501001
31501002
31501003
31502001
31502002
32099997
32099998
32099999
33000101
33000102
33000103
33000104
33000105
33000198
33000199
33000201
33000202
33000203
33000211
33000212
33000213
4
4
4
4
4
4
4
4
4
4
4
2
2
2
2
2
2
4
5
5
6
6
2
2
2
1
1
1
1
1
2
2
2
2
2
2
2
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-40
-------
BLS
Code
SCC Code
Justification
Code
141
114
118
141
141
141
116
130
169
169
169
169
169
169
169
169
117
117
117
117
117
117
117
117
117
117
117
117
117
34
34
34
34
34
34
34
34
34
34
34
34
37
37
33000214
33000499
33000599
33000297
33000298
33000299
33000399
36000101
40100101
40100102
40100103
40100104
40100105
40100106
40100198
40100199
40201101
40201103
40201104
40201105
40201111
40201112
40201113
40201114
40201115
40201116
40201199
40201201
40201210
40202101
40202103
40202104
40202105
40202106
40202107
40202108
40202109
40202131
40202132
40202133
40202199
40201901
40201903
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-41
-------
BLS
Code
SCC Code
Justification
Code
37
37
127
127
127
127
127
127
127
127
127
127
127
127
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
40201904
40201999
40500501
40500502
40500503
40500506
40500507
40500510
40500511
40500512
40500513
40500514
40500598
40500599
40200901
40200902
40200903
40200904
40200905
40200906
40200907
40200908
40200909
40200910
40200911
40200912
40200913
40200914
40200915
40200916
40200917
40200918
40200919
40200920
40200921
40200922
40200923
40200924
40200925
40200926
40200927
40200928
40200929
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
Direct BLS/SCC correlation
SCC part of a larger BLS group
BLS represents SCC end use
SCC represents an ancillary process occurring within BLS industry
SCC assigned to BLS industry responsible for its manufacture
General growth indicator
A-42
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
138
138
138
138
138
138
138
138
138
138
138
138
138
138
40200930
40200931
40200998
40703601
40703602
40703603
40703604
40703605
40703606
40703607
40703608
40703609
40703610
40703611
40703612
40703613
40703614
40703615
40703616
40703617
40703618
40703619
40703620
40703621
40703622
40703623
40703624
40703697
40703698
40301001
40301002
40301003
40301004
40301005
40301006
40301007
40301008
40301009
40301010
40301011
40301012
40301013
40301014
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-43
-------
BLS
Code
SCC Code
Justification
Code
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
40301015
40301016
40301017
40301018
40301019
40301020
40301021
40301097
40301098
40301099
40301101
40301102
40301103
40301104
40301105
40301106
40301107
40301108
40301109
40301110
40301111
40301112
40301113
40301114
40301115
40301116
40301117
40301118
40301119
40301120
40301130
40301131
40301132
40301133
40301134
40301135
40301140
40301141
40301142
40301143
40301144
40301145
40301150
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-44
-------
BLS
Code
SCC Code
Justification
Code
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
131
131
57
57
57
57
57
57
49
49
49
49
49
49
49
49
49
49
49
49
49
49
49
49
49
49
49
40301151
40301152
40301153
40301154
40301155
40301197
40301198
40301199
40301201
40301202
40301203
40301204
40301205
40301206
40301207
40301299
49099998
49099999
40201801
40201803
40201804
40201805
40201806
40201899
40201702
40201703
40201704
40201705
40201721
40201722
40201723
40201724
40201725
40201726
40201727
40201728
40201731
40201732
40201733
40201734
40201735
40201736
40201799
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-45
-------
BLS
Code
SCC Code
Justification
Code
49
49
49
49
49
49
49
49
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
127
84
84
84
84
84
84
84
84
84
84
84
84
84
84
86
86
86
86
86
40202601
40202602
40202603
40202604
40202605
40202606
40202607
40202699
40201401
40201402
40201403
40201404
40201405
40201406
40201431
40201432
40201433
40201434
40201435
40201436
40201437
40201438
40201499
40588801
40201601
40201602
40201603
40201604
40201605
40201606
40201619
40201620
40201621
40201622
40201623
40201624
40201625
40201626
40201627
40201628
40201629
40201630
40201631
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
4
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-46
-------
BLS
Code
SCC Code
Justification
Code
86
86
57
57
57
57
57
57
57
57
57
57
57
57
57
91
91
91
91
91
91
91
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
40201632
40201699
40202501
40202502
40202503
40202504
40202505
40202531
40202532
40202533
40202534
40202535
40202536
40202537
40202599
40202301
40202302
40202303
40202304
40202305
40202306
40202399
40400101
40400102
40400103
40400104
40400105
40400106
40400107
40400108
40400109
40400110
40400111
40400112
40400113
40400114
40400115
40400116
40400117
40400118
40400119
40400120
40400130
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
Direct BLS/SCC correlation
SCC part of a larger BLS group
BLS represents SCC end use
SCC represents an ancillary process occurring within BLS industry
SCC assigned to BLS industry responsible for its manufacture
General growth indicator
A-47
-------
BLS
Code
SCC Code
Justification
Code
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
40400131
40400140
40400141
40400151
40400152
40400153
40400154
40400160
40400161
40400170
40400171
40400199
40400201
40400202
40400203
40400204
40400205
40400206
40400207
40400208
40400209
40400210
40400211
40400212
40400213
40400230
40400231
40400240
40400241
40400250
40400251
40400254
40400260
40400261
40400270
40400271
40400401
40400402
40400403
40400404
40400405
40400406
40400407
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-48
-------
BLS
Code
SCC Code
Justification
Code
138
138
138
138
138
138
138
138
138
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
148
138
138
138
138
138
138
40400408
40400409
40400410
40400411
40400412
40400413
40400414
40400497
40400498
40600231
40600232
40600233
40600234
40600235
40600236
40600237
40600238
40600239
40600240
40600241
40600242
40600243
40600244
40600245
40600246
40600248
40600249
40600250
40600251
40600253
40600254
40600255
40600256
40600257
40600259
40600298
40600299
40600130
40600131
40600132
40600133
40600134
40600135
5
5
5
5
5
5
5
5
5
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-49
-------
BLS
Code
SCC Code
Justification
Code
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
159
159
159
159
159
159
159
159
159
159
131
131
131
131
131
131
131
131
131
131
131
131
40600136
40600137
40600138
40600139
40600140
40600141
40600142
40600143
40600144
40600145
40600146
40600147
40600148
40600149
40600160
40600161
40600162
40600163
40600197
40600198
40600199
40600301
40600302
40600305
40600306
40600307
40600399
40600401
40600402
40600403
40600499
40700401
40700402
40700497
40700498
40700801
40700802
40700803
40700804
40700805
40700806
40700807
40700808
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-50
-------
BLS Justification
Code SCC Code Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
159
40700809
40700810
40700811
40700812
40700813
40700814
40700815
40700816
40700817
40700818
40700897
40700898
40701601
40701602
40701603
40701604
40701605
40701606
40701607
40701608
40701609
40701610
40701611
40701612
40701613
40701614
40701697
40701698
40702001
40702002
40702003
40702004
40702097
40702098
40703201
40703202
40703203
40703204
40703205
40703206
40703297
40703298
40688801
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-51
-------
BLS
Code
SCC Code
Justification
Code
159
159
159
159
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
40688802
40688803
40688804
40688805
40100201
40100202
40100203
40100204
40100205
40100206
40100207
40100208
40100209
40100215
40100216
40100217
40100221
40100222
40100223
40100224
40100225
40100235
40100236
40100251
40100252
40100253
40100254
40100255
40100256
40100257
40100258
40100259
40100295
40100296
40100297
40100298
40100299
40100301
40100302
40100303
40100304
40100305
40100306
4
4
4
4
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-52
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
114
114
135
135
135
135
135
135
135
135
135
135
135
135
137
137
137
137
135
135
135
135
135
135
142
142
142
142
142
142
142
142
142
142
142
40100307
40100308
40100309
40100310
40100335
40100336
40100398
40100399
40100401
40100499
40200101
40200110
40200201
40200210
40200301
40200310
40200401
40200410
40200501
40200510
40200601
40200610
40200701
40200706
40200707
40200710
40200801
40200802
40200803
40200810
40200898
40200899
40202201
40202202
40202203
40202204
40202205
40202206
40202207
40202208
40202209
40202210
40202211
5
5
5
5
5
5
5
5
4
4
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-53
-------
BLS
Code
SCC Code
Justification
Code
142
142
142
142
142
87
87
87
87
87
87
87
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
40202212
40202213
40202214
40202215
40202299
40202401
40202402
40202403
40202404
40202405
40202406
40202499
40300101
40300102
40300103
40300104
40300105
40300106
40300107
40300108
40300109
40300110
40300111
40300112
40300113
40300114
40300115
40300116
40300150
40300151
40300152
40300153
40300154
40300155
40300156
40300157
40300158
40300159
40300160
40300161
40300198
40300199
40300201
5
5
5
5
5
5
5
5
5
5
5
5
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-54
-------
BLS
Code
SCC Code
Justification
Code
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
9
9
9
9
9
127
127
127
127
127
127
127
127
127
127
127
127
127
127
127
127
127
40300202
40300203
40300204
40300205
40300207
40300208
40399209
40300210
40300211
40300212
40300213
40300214
40300215
40300216
40300299
40300302
40388801
40388802
40388803
40388804
40388805
40400301
40400302
40400303
40400304
40400305
40500201
40500202
40500203
40500211
40500212
40500301
40500302
40500303
40500304
40500305
40500306
40500307
40500311
40500312
40500314
40500401
40500411
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-55
-------
BLS
Code
SCC Code
Justification
Code
127
127
127
127
127
127
127
127
127
127
127
127
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
40500412
40500413
40500414
40500601
40500701
40500801
40500802
40500812
40588802
40588803
40588804
40588805
40704001
40704002
40704003
40704004
40704005
40704006
40704007
40704008
40704009
40704010
40704097
40704098
40704401
40704402
40704403
40704404
40704405
40704406
40704407
40704408
40704409
40704410
40704411
40704412
40704413
40704414
40704415
40704416
40704417
40704418
40704419
2
2
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-56
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
40704420
40704421
40704422
40704423
40704424
40704497
40704498
40704801
40704802
40704897
40704898
40705201
40705202
40705203
40705204
40705205
40705206
40705207
40705208
40705209
40705210
40705211
40705212
40705213
40705214
40705215
40705216
40705217
40705218
40705297
40705298
40705601
40705602
40705603
40705604
40705605
40705606
40705607
40705608
40705609
40705610
40705697
40705698
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-57
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
40706001
40706002
40706003
40706004
40706005
40706006
40706007
40706008
40706009
40706010
40706011
40706012
40706013
40706014
40706015
40706016
40706017
40706018
40706019
40706020
40706021
40706022
40706023
40706024
40706097
40706098
40706401
40706402
40706403
40706404
40706497
40706498
40706801
40706802
40706813
40706814
40706897
40706898
40707601
40707602
40707697
40707698
40708001
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-58
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
40708002
40708097
40708098
40708401
40708402
40708403
40708404
40708497
40708498
40717201
40717202
40717203
40717204
40717205
40717206
40717207
40717208
40717209
40717210
40717211
40717212
40717297
40717298
40717601
40717602
40717603
40717604
40717605
40717606
40717697
40717698
40718001
40718002
40718003
40718004
40718005
40718006
40718007
40718008
40718009
40718010
40718097
40718098
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-59
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
40720801
40720802
40720803
40720804
40720897
40720898
40722001
40722002
40722003
40722004
40722005
40722006
40722007
40722008
40722009
40722010
40722097
40722098
40722801
40722802
40722803
40722804
40722805
40722806
40722897
40722898
40723201
40723202
40723297
40723298
40781201
40781202
40781601
40781602
40781603
40781604
40781605
40781606
40781607
40781699
40782001
40782002
40782003
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-60
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
40782004
40782005
40782006
40782007
40782008
40782009
40782010
40782011
40782099
40782401
40782499
40783201
40783202
40783203
40783299
40784801
40784899
40786001
40786002
40786003
40786004
40786099
40786401
40786499
40787201
40787299
40799997
40799998
40899995
40899997
40899999
49000101
49000102
49000103
49000104
49000105
49000199
49000201
49000202
49000203
49000204
49000205
49000206
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-61
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
157
157
157
157
157
220
220
220
157
157
157
157
157
157
157
157
157
157
157
6
6
6
45
45
45
49000299
49000301
49000302
49000303
49000304
49000399
49000401
49000402
49000403
49000404
49000405
49000499
49090011
49090012
49090013
49090021
49090022
49090023
50100101
50100102
50100103
50100201
50100202
50100601
50100602
50100603
50100505
50100506
50100507
50100508
50100510
50100511
50100512
50100701
50100702
50100703
50100704
30300001
30300002
30300003
30300101
30300102
30300103
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-62
-------
BLS
Code
SCC Code
Justification
Code
45
45
45
45
45
45
45
45
45
45
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
45
45
30300104
30300105
30300106
30300107
30300108
30300109
30300110
30300111
30300199
30300201
30300302
30300303
30300304
30300305
30300306
30300307
30300308
30300309
30300310
30300311
30300312
30300313
30300314
30300315
30300316
30300331
30300332
30300333
30300334
30300335
30300336
30300341
30300342
30300343
30300344
30300351
30300352
30300353
30300361
30300399
30300401
30300502
30300503
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
3
4
5
6
Direct BLS/SCC correlation
SCC part of a larger BLS group
BLS represents SCC end use
SCC represents an ancillary process occurring within BLS industry
SCC assigned to BLS industry responsible for its manufacture
General growth indicator
A-63
-------
BLS
Code
SCC Code
Justification
Code
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
43
43
43
43
43
43
43
43
43
43
43
30300504
30300505
30300506
30300507
30300508
30300509
30300510
30300511
30300512
30300513
30300514
30300515
30300516
30300517
30300518
30300519
30300521
30300522
30300523
30300524
30300525
30300526
30300527
30300528
30300529
30300530
30300531
30300532
30300533
30300534
30300535
30300599
30300601
30300602
30300603
30300604
30300605
30300606
30300607
30300610
30300611
30300613
30300614
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-64
-------
BLS Justification
Code SCC Code Code
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
30300615
30300616
30300617
30300699
30300701
30300702
30300703
30300704
30300801
30300802
30300804
30300805
30300808
30300809
30300811
30300812
30300813
30300814
30300815
30300816
30300817
30300818
30300819
30300820
30300821
30300822
30300823
30300824
30300825
30300826
30300827
30300831
30300832
30300833
30300834
30300841
30300842
30300899
30300901
30300904
30300906
30300907
30300908
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-65
-------
BLS
Code
SCC Code
Justification
Code
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
43
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
30300910
30300911
30300912
30300913
30300914
30300915
30300916
30300917
30300918
30300919
30300920
30300921
30300922
30300923
30300924
30300925
30300931
30300932
30300933
30300934
30300935
30300936
30300998
30300999
30301001
30301002
30301003
30301004
30301005
30301006
30301007
30301008
30301009
30301010
30301011
30301012
30301013
30301014
30301015
30301016
30301017
30301018
30301019
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-66
-------
BLS
Code
SCC Code
Justification
Code
45
45
45
45
45
45
45
45
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
30301020
30301021
30301022
30301023
30301024
30301025
30301026
30301099
30301101
30301102
30301199
30301201
30301202
30301299
30301301
30301401
30301402
30301403
30301499
30302301
30302302
30302303
30302304
30302305
30302306
30302307
30302308
30302309
30302310
30302311
30302312
30302313
30302314
30302315
30302316
30302321
30302322
30302401
30302402
30302403
30302404
30302405
30302406
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-67
-------
BLS
Code
SCC Code
Justification
Code
6
6
6
6
6
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
46
46
46
46
46
46
30302407
30302408
30302409
30302410
30302411
30303002
30303003
30303005
30303006
30303007
30303008
30303009
30303010
30303011
30303012
30303014
30303015
30303016
30303099
30388801
30388802
30388803
30388804
30388805
30390001
30390002
30390003
30390004
30390011
30390012
30390013
30390014
30390021
30390022
30390023
30390024
30399999
30400101
30400102
30400103
30400104
30400105
30400106
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-68
-------
BLS
Code
SCC Code
Justification
Code
46
46
46
47
47
47
47
48
49
47
47
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
48
46
44
44
44
44
44
44
30400107
30400108
30400109
30400110
30400111
30400112
30400113
30400114
30400120
30400150
30400199
30400204
30400207
30400208
30400209
30400210
30400211
30400212
30400214
30400215
30400217
30400219
30400220
30400221
30400223
30400224
30400230
30400231
30400232
30400233
30400234
30400235
30400236
30400237
30400238
30400239
30400299
30400301
30400302
30400303
30400304
30400305
30400310
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-69
-------
BLS
Code
SCC Code
Justification
Code
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
46
46
46
46
46
46
46
46
48
46
46
46
46
46
46
82
82
82
82
30400315
30400320
30400325
30400330
30400331
30400332
30400333
30400340
30400341
30400342
30400350
30400351
30400352
30400353
30400354
30400355
30400356
30400357
30400358
30400360
30400370
30400371
30400398
30400399
30400401
30400402
30400403
30400404
30400405
30400406
30400407
30400408
30400409
30400410
30400411
30400412
30400413
30400414
30400499
30400501
30500502
30500503
30400504
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-70
-------
BLS
Code
SCC Code
Justification
Code
82
82
82
82
82
82
82
82
82
82
82
82
82
82
82
82
82
82
82
82
82
45
45
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
44
30400505
30400506
30400507
30400508
30400509
30400510
30400511
30400512
30400513
30400514
30400521
30400522
30400523
30400524
30400525
30400526
30400527
30400528
30400529
30400530
30400599
30400601
30400699
30400701
30400702
30400703
30400704
30400705
30400706
30400707
30400708
30400709
30400710
30400711
30400712
30400713
30400714
30400715
30400716
30400717
30400718
30400720
30400721
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-71
-------
BLS
Code
SCC Code
Justification
Code
44
44
44
44
44
44
44
44
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
46
30400722
30400723
30400724
30400725
30400726
30400730
30400731
30400799
30400801
30400802
30400803
30400805
30400806
30400807
30400809
30400810
30400811
30400812
30400814
30400818
30400824
30400828
30400834
30400838
30400840
30400841
30400842
30400843
30400851
30400852
30400853
30400854
30400855
30400861
30400862
30400863
30400864
30400865
30400866
30400867
30400868
30400869
30400870
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-72
-------
BLS
Code
SCC Code
Justification
Code
46
46
46
46
46
46
46
46
44
44
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
74
74
74
74
74
46
46
46
57
48
48
48
48
48
47
30400871
30400872
30400873
30400874
30400875
30400876
30400877
30400899
30400901
30400999
30401001
30401002
30401004
30401005
30401006
30401007
30401008
30401010
30401011
30401015
30401016
30401017
30401018
30401019
30401061
30401062
30401063
30401099
30402001
30402002
30402003
30402004
30402099
30402201
30402210
30402211
30404001
30404901
30404902
30404999
30405001
30405099
30488801
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
4
4
4
2
1
1
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-73
-------
BLS
Code
SCC Code
Justification
Code
47
47
47
47
149
149
149
149
149
149
149
149
149
149
149
149
149
145
145
145
141
141
141
141
141
141
194
194
194
131
131
104
107
108
110
112
44
46
48
41
42
139
30
30488802
30488803
30488804
30488805
2275000000
2275001000
2275020000
2275050000
2275060000
2275070000
2275085000
2275900000
2275900101
2275900102
2275900103
2275900201
2275900202
2285002000
2285002005
2285002010
2430000000
2430000170
2430000340
2430000350
2430000370
2430000999
2850000000
2850000010
2850000030
2301010000
2301010010
2302010000
2302040000
2302050000
2302070000
2302080000
2303020000
2304000000
2304050000
2305070000
2305080000
2306010000
2307010000
4
4
4
4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
4
4
4
4
4
4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-74
-------
BLS
Code
SCC Code
Justification
Code
31
32
34
141
57
57
57
57
57
57
57
57
57
57
57
8
8
8
8
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
2307020000
2307030000
2307060000
2308000000
2309100000
2309100010
2309100030
2309100050
2309100080
2309100110
2309100140
2309100170
2309100200
2309100230
2309100260
2310000000
2310010000
2310020000
2310030000
2311000000
2311000010
2311000020
2311000030
2311000040
2311000050
2311000060
2311000070
2311000080
2311000100
2311010000
2311010010
2311010020
2311010030
2311010040
2311010050
2311010060
2311010070
2311010080
2311010100
2311020000
2311020010
2311020020
2311020030
1
1
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
Direct BLS/SCC correlation
SCC part of a larger BLS group
BLS represents SCC end use
SCC represents an ancillary process occurring within BLS industry
SCC assigned to BLS industry responsible for its manufacture
General growth indicator
A-75
-------
BLS
Code
SCC Code
Justification
Code
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
66
10
10
10
10
10
10
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
2311020040
2311020050
2311020060
2311020070
2311020080
2311020100
2311030000
2311030010
2311030020
2311030030
2311030040
2311030050
2311030060
2311030070
2311030080
2311030100
2311040000
2311040080
2311040100
2312050000
2325000000
2325010000
2325020000
2325030000
2325040000
2325050000
2401001000
2401001030
2401001055
2401001060
2401001065
2401001070
2401001125
2401001130
2401001135
2401001170
2401001200
2401001210
2401001215
2401001235
2401001250
2401001275
2401001285
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
2
2
1
1
1
1
1
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-76
-------
BLS
Code
SCC Code
Justification
Code
399
399
182
182
182
182
182
182
182
182
182
182
182
182
182
182
182
182
182
182
182
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
114
114
114
2401001370
2401001999
2401005000
2401005030
2401005055
2401005060
2401005065
2401005070
2401005125
2401005130
2401005135
2401005170
2401005200
2401005210
2401005215
2401005235
2401005250
2401005275
2401005285
2401005370
2401005999
2401008000
2401008030
2401008055
2401008060
2401008065
2401008070
2401008125
2401008130
2401008135
2401008170
2401008200
2401008210
2401008215
2401008235
2401008250
2401008275
2401008285
2401008370
2401008999
2401010000
2401010030
2401010055
6
6
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-77
-------
BLS
Code
SCC Code
Justification
Code
114
114
114
114
114
114
114
114
114
114
114
114
114
114
114
114
34
34
34
34
34
34
34
34
34
34
34
34
34
34
34
34
34
34
34
39
39
39
39
39
39
39
39
2401010060
2401010065
2401010070
2401010125
2401010130
2401010135
2401010170
2401010200
2401010210
2401010215
2401010235
2401010250
2401010275
2401010285
2401010370
2401010999
2401015000
2401015030
2401015055
2401015060
2401015065
2401015070
2401015125
2401015130
2401015135
2401015170
2401015200
2401015210
2401015215
2401015235
2401015250
2401015275
2401015285
2401015370
2401015999
2401020000
2401020030
2401020055
2401020060
2401020065
2401020070
2401020125
2401020130
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC pan of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-78
-------
BLS
Code
SCC Code
Justification
Code
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
120
120
120
120
120
120
120
120
120
120
120
120
120
2401020135
2401020170
2401020200
2401020210
2401020215
2401020235
2401020250
2401020275
2401020285
2401020370
2401020999
2401025000
2401025030
2401025055
2401025060
2401025065
2401025070
2401025125
2401025130
2401025135
2401025170
2401025200
2401025210
2401025215
2401025235
2401025250
2401025275
2401025285
2401025370
2401025999
2401030000
2401030030
2401030055
2401030060
2401030065
2401030070
2401030125
2401030130
2401030135
2401030170
2401030200
2401030210
2401030215
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-79
-------
BLS
Code
SCC Code
Justification
Code
120
120
120
120
120
120
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
49
49
49
49
49
49
49
49
49
49
49
49
49
49
49
49
49
49
2401030235
2401030250
2401030275
2401030285
2401030370
2401030999
2401035000
2401035030
2401035055
2401035060
2401035065
2401035070
2401035125
2401035130
2401035135
2401035170
2401035200
2401035210
2401035215
2401035235
2401035250
2401035275
2401035285
2401035370
2401035999
2401040000
2401040030
2401040055
2401040060
2401040065
2401040070
2401040125
2401040130
2401040135
2401040170
2401040200
2401040210
2401040215
2401040235
2401040250
2401040275
2401040285
2401040370
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-80
-------
BLS
Code
SCC Code
Justification
Code
49
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
68
68
68
68
2401040999
2401045000
2401045030
2401045055
2401045060
2401045065
2401045070
2401045125
2401045130
2401045135
2401045170
2401045200
2401045210
2401045215
2401045235
2401045250
2401045275
2401045285
2401045370
2401045999
2401050000
2401050030
2401050055
2401050060
2401050065
2401050070
2401050125
2401050130
2401050135
2401050170
2401050200
2401050210
2401050215
2401050235
2401050250
2401050275
2401050285
2401050370
2401050999
2401055000
2401055030
2401055055
2401055060
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-81
-------
BLS
Code
SCC Code
Justification
Code
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
81
81
81
81
81
81
81
81
81
2401055065
2401055070
2401055125
2401055130
2401055135
2401055170
2401055200
2401055210
2401055215
2401055235
2401055250
2401055275
2401055285
2401055370
2401055999
2401060000
2401060030
2401060055
2401060060
2401060065
2401060070
2401060125
2401060130
2401060135
2401060170
2401060200
2401060210
2401060215
2401060235
2401060250
2401060275
2401060285
2401060370
2401060999
2401065000
2401065030
2401065055
2401065060
2401065065
2401065070
2401065125
2401065130
2401065135
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-82
-------
BLS
Code
SCC Code
Justification
Code
81
81
81
81
81
81
81
81
81
81
84
84
84
84
84
84
84
84
84
84
84
84
84
84
84
84
84
84
84
87
87
87
87
87
87
87
87
87
87
87
87
87
87
2401065170
2401065200
2401065210
2401065215
2401065235
2401065250
2401065275
2401065285
2401065370
2401065999
2401070000
2401070030
2401070055
2401070060
2401070065
2401070070
2401070125
2401070130
2401070135
2401070170
2401070200
2401070210
2401070215
2401070235
2401070250
2401070275
2401070285
2401070370
2401070999
2401075000
2401075030
2401075055
2401075060
2401075065
2401075070
2401075125
2401075130
2401075135
2401075170
2401075200
2401075210
2401075215
2401075235
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-83
-------
BLS
Code
SCC Code
Justification
Code
87
87
87
87
87
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
2401075250
2401075275
2401075285
2401075370
2401075999
2401080000
2401080030
2401080055
2401080060
2401080065
2401080070
2401080125
2401080130
2401080135
2401080170
2401080200
2401080210
2401080215
2401080235
2401080250
2401080275
2401080285
2401080370
2401080999
2401085000
2401085030
2401085055
2401085060
2401085065
2401085070
2401085125
2401085130
2401085135
2401085170
2401085200
2401085210
2401085215
2401085235
2401085250
2401085275
2401085285
2401085370
2401085999
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-84
-------
BLS
Code
SCC Code
Justification
Code
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
321
321
321
321
321
2401090000
2401090030
2401090055
2401090060
2401090065
2401090070
2401090125
2401090130
2401090135
2401090170
2401090200
2401090210
2401090215
2401090235
2401090250
2401090275
2401090285
2401090370
2401090999
2401100000
2401100030
2401100055
2401100060
2401100065
2401100070
2401100125
2401100130
2401100135
2401100170
2401100200
2401100210
2401100215
2401100235
2401100250
2401100275
2401100285
2401100370
2401100999
2401200000
2401200030
2401200055
2401200060
2401200065
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-85
-------
BLS
Code
SCC Code
Justification
Code
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
301
301
301
301
301
301
39
39
39
39
2401200070
2401200125
2401200130
2401200135
2401200170
2401200200
2401200210
2401200215
2401200235
2401200250
2401200275
2401200285
2401200370
2401200999
2401990000
2401990030
2401990055
2401990060
2401990065
2401990070
2401990125
2401990130
2401990135
2401990170
2401990200
2401990210
2401990215
2401990235
2401990250
2401990275
2401990285
2401990370
2401990999
2415000000
2415000300
2415000350
2415000370
2415000385
2415000999
2415005000
2415005300
2415005350
2415005370
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-86
-------
BLS
Code
SCC Code
Justification
Code
39
39
43
43
43
43
43
43
46
46
46
46
46
46
52
52
52
52
52
52
67
67
67
67
67
67
75
75
75
75
75
75
84
84
84
84
84
84
96
96
96
96
96
2415005385
2415005999
2415010000
2415010300
2415010350
2415010370
2415010385
2415010999
2415015000
2415015300
2415015350
2415015370
2415015385
2415015999
2415020000
2415020300
2415020350
2415020370
2415020385
2415020999
2415025000
2415025300
2415025350
2415025370
2415025385
2415025999
2415030000
2415030300
2415030350
2415030370
2415030385
2415030999
2415035000
2415035300
2415035350
2415035370
2415035385
2415035999
2415040000
2415040300
2415040350
2415040370
2415040385
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-87
-------
BLS
Code
SCC Code
Justification
Code
96
103
103
103
103
103
103
146
146
146
146
146
146
159
159
159
159
159
159
183
183
183
183
183
183
182
182
182
182
182
182
321
321
321
321
321
321
37
37
37
37
37
37
2415040999
2415045000
2415045300
2415045350
2415045370
2415045385
2415045999
2415050000
2415050300
2415050350
2415050370
2415050385
2415050999
2415055000
2415055300
2415055350
2415055370
2415055385
2415055999
2415060000
2415060300
2415060350
2415060370
2415060385
2415060999
2415065000
2415065300
2415065350
2415065370
2415065385
2415065999
2415100000
2415100300
2415100350
2415100370
2415100385
2415100999
2415105000
2415105300
2415105350
2415105370
2415105385
2415105999
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
6
6
6
6
6
6
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-88
-------
BLS
Code
SCC Code
Justification
Code
43
43
43
43
43
43
46
46
46
46
46
46
52
52
52
52
52
52
67
67
67
67
67
67
75
75
75
75
75
75
84
84
84
84
84
84
96
96
96
96
96
96
103
2415110000
2415110300
2415110350
2415110370
2415110385
2415110999
2415115000
2415115300
2415115350
2415115370
2415115385
2415115999
2415120000
2415120300
2415120350
2415120370
2415120385
2415120999
2415125000
2415125300
2415125350
2415125370
2415125385
2415125999
2415130000
2415130300
2415130350
2415130370
2415130385
2415130999
2415135000
2415135300
2415135350
2415135370
2415135385
2415135999
2415140000
2415140300
2415140350
2415140370
2415140385
2415140999
2415145000
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-89
-------
BLS
Code
SCC Code
Justification
Code
103
103
103
103
103
146
146
146
146
146
146
159
159
159
159
159
159
182
182
182
182
182
182
183
183
183
183
183
183
321
321
321
321
321
321
39
39
39
39
39
39
44
44
2415145300
2415145350
2415145370
2415145385
2415145999
2415150000
2415150300
2415150350
2415150370
2415150385
2415150999
2415155000
2415155300
2415155350
2415155370
2415155385
2415155999
2415160000
2415160300
2415160350
2415160370
2415160385
2415160999
2415165000
2415165300
2415165350
2415165370
2415165385
2415165999
2415200000
2415200300
2415200350
2415200370
2415200385
2415200999
2415205000
2415205300
2415205350
2415205370
2415205385
2415205999
2415210000
2415210300
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
6
6
6
6
6
6
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-90
-------
BLS
Code
SCC Code
Justification
Code
44
44
44
44
46
46
46
46
46
46
57
57
57
57
57
57
67
67
67
67
67
67
75
75
75
75
75
75
84
84
84
84
84
84
96
96
96
96
96
96
103
103
103
2415210350
2415210370
2415210385
2415210999
2415215000
2415215300
2415215350
2415215370
2415215385
2415215999
2415220000
2415220300
2415220350
2415220370
2415220385
2415220999
2415225000
2415225300
2415225350
2415225370
2415225385
2415225999
2415230000
2415230300
2415230350
2415230370
2415230385
2415230999
2415235000
2415235300
2415235350
2415235370
2415235385
2415235999
2415240000
2415240300
2415240350
2415240370
2415240385
2415240999
2415245000
2415245300
2415245350
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-91
-------
BLS
Code
SCC Code
Justification
Code
103
103
103
146
146
146
146
146
146
159
159
159
159
159
159
181
181
181
181
181
181
183
183
183
183
183
183
321
321
321
321
321
321
37
37
37
37
37
37
43
43
43
43
2415245370
2415245385
2415245999
2415250000
2415250300
2415250350
2415250370
2415250385
2415250999
2415255000
2415255300
2415255350
2415255370
2415255385
2415255999
2415260000
2415260300
2415260350
2415260370
2415260385
2415260999
2415265000
2415265300
2415265350
2415265370
2415265385
2415265999
2415300000
2415300300
2415300350
2415300370
2415300385
2415300999
2415305000
2415305300
2415305350
2415305370
2415305385
2415305999
2415310000
2415310300
2415310350
2415310370
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
6
6
6
6
6
6
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-92
-------
BLS Justification
Code SCC Code Code
43
43
46
46
46
46
46
46
52
52
52
52
52
52
67
67
67
67
67
67
75
75
75
75
75
75
84
84
84
84
84
84
96
96
96
96
96
96
103
103
103
103
103
2415310385
2415310999
2415315000
2415315300
2415315350
2415315370
2415315385
2415315999
2415320000
2415320300
2415320350
2415320370
2415320385
2415320999
2415325000
2415325300
2415325350
2415325370
2415325385
2415325999
2415330000
2415330300
2415330350
2415330370
2415330385
2415330999
2415335000
2415335300
2415335350
2415335370
2415335385
2415335999
2415340000
2415340300
2415340350
2415340370
2415340385
2415340999
2415345000
2415345300
2415345350
2415345370
2415345385
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-93
-------
BLS
Code
SCC Code
Justification
Code
103
146
146
146
146
146
146
159
159
159
159
159
159
181
181
181
181
181
181
183
183
183
183
183
183
169
169
169
169
169
169
169
169
169
169
169
169
127
127
127
127
127
127
2415345999
2415350000
2415350300
2415350350
2415350370
2415350385
2415350999
2415355000
2415355300
2415355350
2415355370
2415355385
2415355999
2415360000
2415360300
2415360350
2415360370
2415360385
2415360999
2415365000
2415365300
2415365350
2415365370
2415365385
2415365999
2420000000
2420000055
2420000370
2420000999
2420010000
2420010055
2420010370
2420010999
2420020000
2420020055
2420020370
2420020999
2425000000
2425000055
2425000370
2425000999
2425010000
2425010055
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-94
-------
BLS
Code
SCC Code
Justification
Code
127
127
127
127
127
127
127
127
127
127
127
127
127
127
142
142
142
142
142
142
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
2425010370
2425010999
2425020000
2425020055
2425020370
2425020999
2425030000
2425030055
2425030370
2425030999
2425040000
2425040055
2425040370
2425040999
2430000000
2430000170
2430000340
2430000350
2430000370
2430000999
2440000000
2440000060
2440000065
2440000070
2440000100
2440000125
2440000130
2440000135
2440000165
2440000200
2440000210
2440000215
2440000235
2440000250
2440000260
2440000275
2440000285
2440000300
2440000330
2440000350
2440000370
2440000999
2440020000
1
1
2
2
2
2
1
1
1
1
2
2
2
2
4
4
4
4
4
4
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-95
-------
BLS
Code
SCC Code
Justification
Code
322
322
322
322
322
322
322
322
322
322
322
322
322
322
322
322
322
322
322
322
309
304
304
304
139
139
139
139
139
139
139
139
139
139
139
139
309
309
309
309
309
309
309
2460000000
2460000030
2460000055
2460000060
2460000065
2460000070
2460000165
2460000170
2460000185
2460000250
2460000260
2460000285
2460000300
2460000330
2460000340
2460000345
2460000350
2460000370
2460000385
2460000999
2461000000
2461020000
2461020370
2461020999
2461021000
2461021370
2461021999
2461022000
2461022370
2461022999
2461023000
2461023370
2461023999
2461024000
2461024370
2461024999
2461050000
2461100000
2461160000
2461600000
2461800000
2461800999
2461900000
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
4
4
4
4
4
4
4
4
4
4
4
4
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-96
-------
BLS
Code
SCC Code
Justification
Code
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
2465000000
2465000030
2465000055
2465000060
2465000065
2465000070
2465000165
2465000170
2465000185
2465000250
2465000260
2465000285
2465000300
2465000330
2465000340
2465000345
2465000350
2465000370
2465000385
2465000999
2465100000
2465200000
2465400000
2465600000
2465800000
2465900000
2495000000
2495000001
2495000005
2495000010
2495000015
2495000020
2495000025
2495000030
2495000035
2495000040
2495000045
2495000050
2495000055
2495000060
2495000065
2495000070
2495000075
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-97
-------
BLS
Code
SCC Code
Justification
Code
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
2495000080
2495000085
2495000090
2495000095
2495000100
2495000105
2495000110
2495000115
2495000120
2495000125
2495000130
2495000135
2495000140
2495000145
2495000150
2495000155
2495000160
2495000165
2495000170
2495000175
2495000180
2495000185
2495000190
2495000195
2495000200
2495000205
2495000210
2495000215
2495000220
2495000225
2495000230
2495000235
2495000240
2495000245
2495000250
2495000255
2495000260
2495000265
2495000270
2495000275
2495000280
2495000285
2495000290
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC pan of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-98
-------
BLS
Code
SCC Code
Justification
Code
399
399
399
399
399
399
399
399
399
138
138
138
138
138
138
138
138
158
158
158
158
158
158
158
158
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
2495000295
2495000300
2495000305
2495000310
2495000315
2495000320
2495000325
2495000330
2495000335
2501000000
2501000030
2501000060
2501000090
2501000120
2501000150
2501000180
2501000900
2501010000
2501010030
2501010060
2501010090
2501010120
2501010150
2501010180
2501010900
2501050000
2501050030
2501050060
2501050090
2501050120
2501050150
2501050180
2501050900
2501060000
2501060050
2501060051
2501060052
2501060053
2501060100
2501060101
2501060102
2501060103
2501060200
6
6
6
6
6
6
6
6
6
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-99
-------
BLS
Code
SCC Code
Justification
Code
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
138
2501060201
2501070000
2501070050
2501070051
2501070052
2501070053
2501070100
2501070101
2501070102
2501070103
2501070200
2501070201
2501995000
2501995030
2501995060
2501995090
2501995120
2501995150
2501995180
2505000000
2505000030
2505000060
2505000090
2505000120
2505000150
2505000180
2505000900
2505010000
2505010030
2505010060
2505010090
2505010120
2505010150
2505010180
2505010900
2505020000
2505020030
2505020060
2505020090
2505020120
2505020150
2505020180
2505020900
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-100
-------
BLS
Code
SCC Code
Justification
Code
138
138
138
138
138
138
138
138
150
150
150
150
150
150
150
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
2505030000
2505030030
2505030060
2505030090
2505030120
2505030150
2505030180
2505030900
2505040000
2505040030
2505040060
2505040090
2505040120
2505040150
2505040180
2510000000
2510000030
2510000060
2510000065
2510000070
2510000100
2510000165
2510000185
2510000195
2510000220
2510000235
2510000240
2510000250
2510000260
2510000265
2510000270
2510000275
2510000285
2510000295
2510000310
2510000320
2510000345
2510000350
2510000370
2510000380
2510000385
2510000405
2510000900
5
5
5
5
5
5
5
5
5
1
1
1
1
1
1
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-101
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
2510010000
2510010030
2510010060
2510010065
2510010070
2510010100
2510010165
2510010185
2510010195
2510010220
2510010235
2510010240
2510010250
2510010260
2510010265
2510010270
2510010275
2510010285
2510010295
2510010310
2510010320
2510010345
2510010350
2510010370
2510010380
2510010385
2510010405
2510010900
2510050000
2510050030
2510050060
2510050065
2510050070
2510050100
2510050165
2510050185
2510050195
2510050220
2510050235
2510050240
2510050250
2510050260
2510050265
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-102
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
2510050270
2510050275
2510050285
2510050295
2510050310
2510050320
2510050345
2510050350
2510050370
2510050380
2510050385
2510050405
2510050900
2510050000
2510050030
2510050060
2510050065
2510050070
2510050100
2510050165
2510050185
2510050195
2510050220
2510050235
2510050240
2510050250
2510050260
2510050265
2510050270
2510050275
2510050285
2510050295
2510050310
2510050320
2510050345
2510050350
2510050370
2510050380
2510050385
2510050405
2510050900
2510995000
2510995030
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-103
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
2510995060
2510995065
2510995070
2510995100
2510995165
2510995185
2510995195
2510995220
2510995235
2510995240
2510995250
2510995260
2510995265
2510995270
2510995275
2510995285
2510995295
2510995310
2510995320
2510995345
2510995350
2510995370
2510995380
2510995385
2510995405
2515000000
2515000030
2515000060
2515000065
2515000070
2515000100
2515000165
2515000185
2515000195
2515000220
2515000235
2515000240
2515000250
2515000260
2515000265
2515000270
2515000275
2515000285
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC pan of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-104
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
2515000295
2515000310
2515000320
2515000345
2515000350
2515000370
2515000380
2515000385
2515000405
2515000900
2515010000
2515010030
2515010060
2515010065
2515010070
2515010100
2515010165
2515010185
2515010195
2515010220
2515010235
2515010240
2515010250
2515010260
2515010265
2515010270
2515010275
2515010285
2515010295
2515010310
2515010320
2515010345
2515010350
2515010370
2515010380
2515010385
2515010405
2515010900
2515020000
2515020030
2515020060
2515020065
2515020070
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-105
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
2515020100
2515020165
2515020185
2515020195
2515020220
2515020235
2515020240
2515020250
2515020260
2515020265
2515020270
2515020275
2515020285
2515020295
2515020310
2515020320
2515020345
2515020350
2515020370
2515020380
2515020385
2515020405
2515020900
2515030000
2515030030
2515030060
2515030065
2515030070
2515030100
2515030165
2515030185
2515030195
2515030220
2515030235
2515030240
2515030250
2515030260
2515030265
2515030270
2515030275
2515030285
2515030295
2515030310
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-106
-------
BLS
Code
SCC Code
Justification
Code
131
131
131
131
131
131
131
131
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
131
131
131
131
131
131
131
131
2515030320
2515030345
2515030350
2515030370
2515030380
2515030385
2515030405
2515030900
2515040000
2515040030
2515040060
2515040065
2515040070
2515040100
2515040165
2515040185
2515040195
2515040220
2515040235
2515040240
2515040250
2515040260
2515040265
2515040270
2515040275
2515040285
2515040295
2515040310
2515040320
2515040345
2515040350
2515040370
2515040380
2515040385
2515040405
2520000000
2520000010
2520000020
2520000030
2520000040
2520000900
2520010000
2520010010
5
5
5
5
5
5
5
5
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-107
-------
BLS
Code
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
131
150
150
150
150
SCC Code
2520010020
2520010030
2520010040
2520010900
2520050000
2520050010
2520050020
2520050030
2520050040
2520050900
2520995000
2520995010
2520995020
2520995030
2520995040
2525000000
2525000010
2525000020
2525000030
2525000040
2525.000900
2525010000
2525010010
2525010020
2525010030
2525010040
2525010900
2525020000
2525020010
2525020020
2525020030
2525020040
2525020900
2525030000
2525030010
2525030020
2525030030
2525030040
2525030900
2525040000
2525040010
2525040020
2525040030
Justification
Code
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1
1
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-108
-------
BLS
Code
SCC Code
Justification
Code
150
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42 »
2525040040
2530000000
2530000020
2530000040
2530000060
2530000080
2530000100
2530000120
2530010000
2530010020
2530010040
2530010060
2530010080
2530010100
2530010120
2530050000
2530050020
2530050040
2530050060
2530050080
2530050100
2530050120
2535000000
2535000020
2535000040
2535000060
2535000080
2535000100
2535000120
2535000140
2535010000
2535010020
2535010040
2535010060
2535010080
2535010100
2535010120
2535010140
2535020000
2535020020
2535020040
2535020060
2535020080
1
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-109
-------
BLS
Code
SCC Code
Justification
Code
42
42
42
42
42
42
42
42
42
42
42
324
321
322
399
324
321
322
399
324
321
322
399
399
321
157
157
399
399
399
399
399
321
321
321
321
321
322
322
322
322
322
324
2535020100
2535020120
2535020140
2535030000
2535030020
2535030040
2535030060
2535030080
2535030100
2535030120
2535030140
2601000000
2601010000
2601020000
2601030000
2610000000
2610010000
2610020000
2610030000
2620000000
2620010000
2620020000
2620030000
2630000000
2630010000
2630020000
2630030000
2640000000
2640000001
2640000002
2640000003
2640000004
2640010000
2640010001
2640010002
2640010003
2640010004
2640020000
2640020001
2640020002
2640020003
2640020004
2650000000
5
5
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
6
6
6
6
4
4
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-110
-------
BLS
Code
SCC Code
Justification
Code
324
324
324
324
324
324
131
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
107
304
304
68
84
37
37
37
37
37
37
37
2650000001
2650000002
2650000003
2650000004
2650000005
2660000000
30102199
30200701
30200702
30200703
30200704
30200705
30200711
30201901
30201902
30201903
30201904
30201905
30201906
30201907
30201908
30201909
30201911
30201912
30201913
30201914
30201915
30201916
30201917
30201918
30201919
30201920
31100199
31100299
31299999
31499999
40202001
40202002
40202003
40202004
40202005
40202031
40202032
6
6
6
6
6
6
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
6
6
2
2
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-lll
-------
BLS
Code
SCC Code
Justification
Code
37
37
37
127
127
220
322
322
322
322
322
322
322
322
322
322
322
322
322
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
321
40202033
40202034
40202099
40500101
40500199
50100604
50200101
50200102
50200103
50200104
50200105
50200201
50200202
50200301
50200302
50200505
50200506
50200601
50200602
50200901
50300101
50300102
50300103
50300104
50300105
50300106
50300107
50300108
50300109
50300201
50300202
50300203
50300204
50300205
50300501
50300506
50300601
50300602
50300603
50300701
50300801
50300810
50300820
4
4
4
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-112
-------
BLS
Code
SCC Code
Justification
Code
321
321
399
399
399
399
399
399
399
399
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
321
321
321
321
321
50300830
50300901
2260000000
2260001000
2260001010
2260001020
2260001030
2260001040
2260001050
2260001060
2260002000
2260002003
2260002006
2260002009
2260002012
2260002015
2260002018
2260002021
2260002024
2260002027
2260002030
2260002033
2260002036
2260002039
2260002042
2260002045
2260002048
2260002051
2260002054
2260002057
2260002060
2260002063
2260002066
2260002069
2260002072
2260002075
2260002078
2260002081
2260003000
2260003010
2260003020
2260003030
2260003040
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-113
-------
BLS
Code
SCC Code
Justification
Code
321
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
322
322
322
322
322
322
322
30
30
30
30
30
149
149
149
399
399
399
2260003050
2260004000
2260004010
2260004015
2260004020
2260004030
2260004035
2260004040
2260004045
2260004050
2260004055
2260004060
2260004065
2260004070
2260004075
2260005000
2260005010
2260005015
2260005020
2260005030
2260005035
2260005040
2260005045
2260005050
2260005055
2260006000
2260006005
2260006010
2260006015
2260006020
2260006025
2260006030
2260007000
2260007005
2260007010
2260007015
2260007020
2260008000
2260008005
2260008010
2265000000
2265001000
2265001010
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1
1
1
1
1
4
4
4
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-114
-------
BLS
Code
SCC Code
Justification
Code
399
399
399
399
399
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
321
321
321
321
399
399
399
399
399
399
2265001020
2265001030
2265001040
2265001050
2265001060
2265002000
2265002003
2265002006
2265002009
2265002012
2265002015
2265002018
2265002021
2265002024
2265002027
2265002030
2265002033
2265002036
2265002039
2265002042
2265002045
2265002048
2265002051
2265002054
2265002057
2265002060
2265002063
2265002066
2265002069
2265002072
2265002075
2265002078
2265002081
2265003000
2265003030
2265003040
2265003050
2265004000
2265004010
2265004015
2265004020
2265004025
2265004030
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-115
-------
BLS
Code
SCC Code
Justification
Code
399
399
399
399
399
399
399
399
399
314
314
314
314
314
314
314
314
314
314
314
322
322
322
322
322
322
322
30
30
30
30
30
149
149
149
399
399
399
399
399
399
399
399
2265004035
2265004040
2265004045
2265004050
2265004055
2265004060
2265004065
2265004070
2265004075
2265005000
2265005010
2265005015
2265005020
2265005025
2265005030
2265005035
2265005040
2265005045
2265005050
2265005055
2265006000
2265006005
2265006010
2265006015
2265006020
2265006025
2265006030
2265007000
2265007005
2265007010
2265007015
2265007020
2265008000
2265008005
2265008010
2270000000
2270001000
2270001010
2270001020
2270001030
2270001040
2270001050
2270001060
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1
1
1
1
1
4
4
4
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-116
-------
BLS
Code
SCC Code
Justification
Code
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
304
321
321
321
321
321
321
399
399
399
399
399
399
399
399
399
2270002000
2270002003
2270002006
2270002009
2270002012
2270002015
2270002018
2270002021
2270002024
2270002027
2270002030
2270002033
2270002036
2270002039
2270002042
2270002045
2270002048
2270002051
2270002054
2270002057
2270002060
2270002063
2270002066
2270002069
2270002072
2270002075
2270002078
2270002081
2270003000
2270003010
2270003020
2270003030
2270003040
2270003050
2270004000
2270004010
2270004015
2270004020
2270004025
2270004030
2270004035
2270004040
2270004045
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
S = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-117
-------
BLS
Code
SCC Code
Justification
Code
399
399
399
399
399
399
314
314
314
314
314
314
314
314
314
314
314
322
322
322
322
322
322
322
30
30
30
30
30
149
149
149
148
148
148
148
148
148
148
148
148
148
148
2270004050
2270004055
2270004060
2270004065
2270004070
2270004075
2270005000
2270005010
2270005015
2270005020
2270005025
2270005030
2270005035
2270005040
2270005045
2270005050
2270005055
2270006000
2270006005
2270006010
2270006015
2270006020
2270006025
2270006030
2270007000
2270007005
2270007010
2270007015
2270007020
2270008000
2270008005
2270008010
2280001000
2280001010
2280001020
2280001030
2280001040
2280002000
2280002010
2280002020
2280002030
2280002040
2280003000
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1
1
1
1
1
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-118
-------
BLS
Code
SCC Code
Justification
Code
148
148
148
148
148
148
148
148
148
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
2
1
2280003010
2280003020
2280003030
2280003040
2280004000
2280004010
2280004020
2280004030
2280004040
2282005000
2282005005
2282005010
2282005015
2282005020
2282005025
2282010000
2282010005
2282010010
2282010015
2282010020
2282010025
2282010000
2282020005
2282020010
2282020015
2282020020
2282020025
2801000000
2805010000
4
4
4
4
4
4
4
4
4
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
1
1
1 = Direct BLS/SCC correlation
2 = SCC part of a larger BLS group
3 = BLS represents SCC end use
4 = SCC represents an ancillary process occurring within BLS industry
5 = SCC assigned to BLS industry responsible for its manufacture
6 = General growth indicator
A-119
-------
APPENDIX B
EXAMPLE CROSSWALK OUTPUT FILES
B-l
-------
1) RES-FUEL.SCC HOMES Residential Fossil Fuels
State County
01 004
01 005
Years:
sec
2104001000
2104001000
1993 - 1997
Growth Factor
1.1 1.3
1.0 1.1
1.4 1.4 1.1
1.1 1.0 1.5
2) COM-FUEL.SCC CSEMS Commercial Fossil Fuels
Years: 1993 - 1997
State
01
01
County
004
005
sec
2103005000
2103005000
Growth Factor
1.1 1.3 1.4
1.0 1.1 1.1
1.4 1.1
1.0 1.5
3) IND-FUEL.SCC INRAD Industrial Fossil Fuels
State
01
01
Years:
County SCC
004 2102003000
005 2102003000
1993 - 1997
Growth Factor
1.1 1.3
1.0 1.1
1.4 1.4 1.1
1.1 1.0 1.5
B-2
-------
4) ELECTRIC.SCC Electric Demands
5)
6)
Years: 1993 - 1997
State County SCC Growth Factor
01 004 2101004000 1.1 1.3 1.4
01 005 2101004000 1.0 1.1 1.1
VMT.SCC VMT Transportation
1.4 1.1
1.0 1.5
Demands
Years: 1993 - 1997
State County SCC Growth Factor
01 004 2201001191 1.1 1.3 1.4
01 004 2201001000 1.0 1.1 1.1
PHY.SCC PHYSICAL OUTPUT
1.4 1.1
1.0 1.5
Demands
Years: 1993 - 1997
State County SCC Growth Factor
01 004 2304000000 1.1 1.3 1.4
01 005 2304000000 1.0 1.1 1.1
1.4 1.1
1.0 1.5
B-3
-------
7) OTHER.SCC Unmatched SSCs
Years: 1993 - 1997
State County SCC Growth Factor
34 123 2304000000 1.0 1.0 1.0 1.0 1.0
34 124 2304000000 1.0 1.0 1.0 1.0 1.0
B-4
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/R-93-067a
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Economic Growth Analysis System: Reference Manual
5. REPORT DATE
April 1993
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S) T> Lynch> T. Young, K.Johnson, D. Bowman,
J. Vitas, T. Wilson, A. Chadha, and L. Alpern
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
TRC Environmental Corporation
100 Europa Drive, Suite 150
Chapel Hill, North Carolina 27514
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D9-0173, Task 3/302
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 10/92 - 3/93
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES AEEHL project officer is E. Sue Kimbrough, Mail Drop 62, 919/
541-2612. "b" of this series is the User's Guide.
16. ABSTRACT.
The two-volume report describes the development of, and provides informa-
tion needed to operate, a prototype Economic Growth Analysis System (E-GAS)
modeling system. The model will be Used to project emissions inventories of vola-
tile organic compounds (VOCs), oxides of nitrogen (NOx), and carbon monoxide (CO)
for ozone nonattainment areas and Regional Oxidation Model (ROM) modeling regions.
The report details the design and development of the E-GAS computer modeling soft-
ware, and its relationships with internal modeling software components and external
software. The system is an economic and activity forecast model which translates
users' assumptions regarding regional economic policies and resource prices into
Source Code Classification (SCC) level growth factors for VOCs, NOx, and CO. The
report provides E-GAS users with sufficient background information to define and
calibrate the E-GAS model, as well as the procedures and syntax necessary to oper-
ate the system. The organization of the document is determined by the process used
in operating the system. The guide provides images of sample screens as well as
text.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution Volatility
Mathematical Models Emission
Analyzing Inventories
Economic Development Organic Corn-
Nitrogen Oxides pounds
Carbon Monoxide
Pollution Control
Stationary Sources
Economic Growth Analy-
sis System (E-GAS)
Volatile Organic Com-
pounds
13 B
12 A
05C
07B
20M
14G
15E
07C
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
257
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
B-5
------- |