Geoscience Reference
In-Depth Information
Statistics and the Census Bureau using a sample survey of industries through the
Economic Census (USDOT, RITA and BTS
2005
). Although the CFS is useful due
to the wide range of commodity shipments and multimodal movement data, some
user groups are not satisfied with the CFS details which have the incomplete
coverage by commodity sectors and regions and the inability to fully capture
foreign imports and exports (Southworth
2005
; Park et al.
2009
; Giuliano
et al.
2010
). Furthermore, the CFS estimates only cover less than 75 % of all the
annual freight tons delivered in the U.S. This is because the CFS survey drops many
establishments classified as farms, forestry, fisheries, construction, transportation,
governments, foreign establishments, services, and most retail activities (ORNL
2000
).
Nevertheless, Park et al. (
2009
) managed to estimate interstate trade flows. They
developed two computation modules: an Adjusted Flow Model (AFM) and a
Doubly-constrained Fratar Model (DFM). The systematic approach relies on the
secondary data of CFS and IMPLAN. To reconcile different definitions and
classifications of the commodities among multiple data sources, a new commodity
sector scheme of 47 sectors (29 commodity, 18 service) was developed, referred to
as the “USC Sectors.” With various applications applying this initial version of
NIEMO that excluded interstate services trade, the NIEMO's compatibility of the
state-to-state trade flows and the flows between the states and the rest of world for
the 29 USC commodity sectors were all readily testable (Park and Gordon
2005
).
We use NIEMO to estimate total economic impacts including indirect impacts.
As many of our case studies show, we have developed and applied NIEMO to
estimate the impacts of various natural and man-made disasters. While many types
of economic approaches including benefit-cost analysis, single-region I-O
modeling, social accounting models, partial equilibrium models, and computable
general equilibrium (CGE) models, are available for estimating economic impacts
(Rich et al.
2005a
,
b
), none of these tools provide national economic impacts at the
level of individual states. For example, our studies, estimating the indirect and
induced effects of impacts associated with capacity losses at the twin ports of Los
Angeles-Long Beach showed that two-thirds of the impacts leaked outside the
Southern California region. Without an interstate model such as NIEMO, we
would have had no idea where these leakage effects might occur.
While a mathematical method to estimate interregional trade flows by Canning
and Wang (
2005
) permits an empirical performance test, our foundation of
estimating interregional trade flows is an approach based on secondary data
sources. It uses CFS and IMPLAN, but adopts a different approach to estimating
missing CFS data, a doubly-constrained Fratar model (DFM) to update the MRIO
matrix to a base year.
Constructing NIEMO requires two basic kinds of tables: industrial trade
coefficients tables and regional interindustry coefficients tables. While trade tables
by industry are difficult to construct because of incomplete information in the CFS
data, the interindustry tables present fewer problems because reliable data are
available from IMPLAN at the state and industry levels. For details of the procedure
used to estimate values for the empty cells in the trade flow matrix, see Park
