Geoscience Reference
In-Depth Information
are no direct resource loss shocks after 2013). Examining the legacy of the event
year shocks, it is clear that while the enduring effects of the resource loss shocks
have largely dissipated by 2024, this is not the case for the behavioral shocks. In the
event year, the resource loss shocks depress GDP by
0.026 % relative to baseline.
By 2024 this contribution has fallen to
0.0009 % (1/28th of the 2013 resource
result). The event year behavioral shocks lower GDP in 2013 by
0.056 % relative
to baseline. Examining Fig.
16.10
, we see a tapering contribution by the 2013
behavioral shocks to the GDP deviations of future years. Nevertheless, by 2024, the
2013 behavioral shocks are contributing
0.016 % to the GDP deviation: just over
1/4th of the 2013 behavioral impact. Examining the time paths of the decomposi-
tion results for each of the post-event years, Fig.
16.10
shows a similar pattern of
gradual tapering of the enduring effects of the deviations from baseline in the year
1+
n
values for our risk premium, wage premium and willingness to pay variables.
The persistence of the behavioral contributions to the GDP deviations of future
years can be traced to our previous explanation of the dynamic interaction between
the deviations in capital and population. That is, the interaction between these two
variables, via their respective impacts on the marginal product of labor (and thus
wages and inter-regional immigration) and the marginal product of capital (and thus
rates of return and investment) damps the speed with which these variables return to
baseline.
The advantages of a dynamic model in elucidating the long-run consequences of
the behavioral responses arising from a terrorism event are demonstrated by a
comparison of Fig.
16.10
with the estimated long-run real GDP consequences of
an RDD attack as reported in Giesecke et al. (
2012
). As discussed in Sect.
16.3.1
,
the latter paper used a comparative static version of the dynamic model used in this
paper. Because their model was not dynamic, Giesecke et al. inferred annual results
between years
t
+ 1 and
t
+ 10 by assuming a linear transition of GDP impacts from
short-run (event-year) results, through to long-run results for the year
t
+ 5, and then
to an assumption of a return to baseline (i.e. zero GDP impacts) by year
t
+ 10.
23
However, their assumption of no enduring impacts by year
t
+ 10 appears, in the
light of Fig.
16.10
, to under-estimate the potential for long-run adverse regional
GDP consequences via behavioral effects.
Table
16.3
provides a comparison of event year and long-run real regional GDP
outcomes in terms of contributions by BI, other resource losses, and behavioral
effects. Consistent with Giesecke et al., Table
16.3
highlights the importance of
behavioral effects in both the event year and the long-run. The total real regional
GDP loss in the event year is $447 m., of which $311 m. arises from behavioral
effects alone. Much of the remainder is due to the direct and indirect flow-on
consequences of BI ($81 m. and $50 m., respectively). These BI losses are signifi-
cantly lower than those calculated for an RDD event in Giesecke et al., in large part
23
See Fig. 3 in Giesecke et al. (
2012
).
