Geoscience Reference
In-Depth Information
Table 10.4
Estimation ratio of SECGE and traditional CGE
Truck
Air
Water
Transit
All modes
Agriculture
4%
2%
2%
34 %
6%
Manufacture
1%
2%
2%
7%
6%
Utility & Construction
2%
4%
1%
4%
2%
Trade
4%
16 %
2%
8%
5%
Truck
12 %
2%
6%
12 %
12 %
Rail
3 %
2 %
3%
2%
23 %
Air
35 %
2%
2 %
25 %
1 %
Transit
11 %
2%
4%
11 %
11 %
Water
75 %
2%
1%
57 %
1%
Pipeline
2%
5%
0%
10 %
2%
Warehouse
7 %
2 %
1%
13 %
2%
Information
7%
0%
4%
2%
3%
Service
8%
3%
3%
2%
5%
GDPVA
0 %
0 %
0 %
0 %
0 %
Household
0 %
0 %
0 %
0 %
0 %
Welfare
0 %
0 %
0 %
0 %
0 %
Note
: Variation over 10 percent is highlighted in bold
Source
: Authors' calculation
indicates there is no difference of simulation results; a positive value indicates that
SECGE provides higher values of simulation than traditional CGE based on OLS
estimation of substitution elasticities. In spite of the fact that the two estimation
routines do not show any differences in economic outputs at the aggregate level, the
ratio does reveal that the results vary among different sectors and by different
modes.
We indicate large difference (over
10 %) from the estimation ratio of SECGE
and CGE based on OLS in bold. Most changed impacts are among the transport-
ation sectors which indicate the spatial interdependence across these sectors and the
importance of a multimodal assessment. Next the largest changes outside the
transportation sectors are between air transportation and trade and between transit
and agriculture. The former is clear but the latter may reflect agglomeration or
urbanization.
Conclusion
This study develops a new method that integrates both spatial econometrics and
equilibrium modeling to improve the effectiveness of impact analysis on trans-
portation infrastructure. Findings of the study have three implications:
First, the economic impacts of public transportation infrastructure in the US
are confirmed to be positive under the general equilibrium framework. However,
the magnitude of impact is much smaller than that have been found in many
previous studies (Boarnet
1998
; Holtz-Eakin and Schwartz
1995
; Kelejian and
Robinson
1997
; Ozbay et al.
2007
; Cohen and Morrison Paul
2003
,
2004
; Cohen
2007
). There are two possible causes of this. First of all, the study differs from
