Page 155 - Special Topic Session (STS) - Volume 2

STS474 Gigih F. et al.

Fig 8. The spatial correlation before (left) and after (right) breakpoint in
Japan.

4. Discussion and Conclusion
Several statistical methods for structural change, such as F-test and Wald-
test based on broken time-trend model are used to analyze the effect of
Lehman’s crisis towards regional economic growth in US and Japan.
We observed that the F-test perform better compared to the Wald test for
spatially-independent model. The Wald test produced better results for
spatially-dependent model. We confirmed, the endogeneity in spatially-
dependent model might imply the F-test become non-robust for
autocorrelated disturbances (Krӓmer, 2003).
In the US, it was revealed that the negative impact of crisis clustered in
West Coast, Southeastern, and Great Lake region. States in those regions with
good manufacturing, construction, insurance, and finance as the main
contributor for their GRP. On the other hand, states that relied on agriculture,
forestry, fishing, hunting, and mining sectors had resilience.
In Japan, we observed a negative impact spread-out across all the regions.
After the crisis, most prefectures showed recoveries, except several other
prefectures, such as Nagasaki. The negative trend in Nagasaki mainly affected
by decline in the ship building industry, fisheries industry, and small enterprise
(Miyao, 2014).

References
1. Ahn, J.B., Amiti, M., & Weinstein, D.E. (2011). "Trade finance and the great
trade collapse." American Economic Review, 101 (3), 298-302.
2. Aloui, R., Aissa, M.S.B., & Nguyen, D.K. (2011). “Global financial crisis,
extreme interdependences, and contagion effects: the role of economic
structure?” Journal of Banking & Finance, Vol. 35 (1), 130-141.
th
3. Bank for International Settlements (BIS) Report. (2009). 79 Annual
Report: 1 April 2008-31 March 2009. BIS Publications.
144 | I S I W S C 2 0 1 9