8 Level of Disaster Resilience and Migration …
147
more people move-out from and move-in to municipalities with low resilience, while
less people move-out from and move-in to municipalities with high resilience.
Table 8.3 presents the results of multivariate regressions of in- and out-migration
rate (specification 1). In the first column, the dependent variable is out-migration rate,
and the explanatory variables include prefecture fixed effects as well as a dummy
variable for high resilience, a dummy variable for low resilience, logged per capita
income, logged school density and logged population density. A dummy variable
for high resilience is associated with a 0.22% reduction in out-migration rate, while
a dummy variable for low resilience is associated with a 0.45% increase in outmigration rate. Column two shows the same regression in the sample of foreigners.
Unlike Japanese nationals, foreigners do not move-out more from municipalities
with low resilience. Similarly, they do not move-out less from municipalities with
high resilience.
The next two columns show the results of the analysis of in-migration among the
Japanese and foreign population. The in-migration rate among Japanese nationals
is 0.16% lower in municipalities with high resilience, compared to municipalities
with middle resilience. This decrease in the in-migration rate is smaller than the
decrease in out-migration (0.22%). Thus, higher resilience is associated with social
increase. On the other hand, the in-migration rate is 0.38% higher in municipalities
with low resilience, compared to municipalities with middle resilience. However,
this increase in in-migration is smaller than the increase in out-migration (0.45%).
Thus, low resilience is associated with social decrease. However, these associations
are not observed in the sample of foreigners.
From columns five to eight, I disaggregated resilience in RCI’s three sub-indices:
regional economic capacity, socio-demographic capacity and community connectivity capacity. Column five replicates the analysis of the out-migration rate with
Japanese nationals using the disaggregated RCI. High community connectivity is
associated with a 0.33% decrease in out-migration, while low community connectivity is associated with a 0.99% increase in out-migration. In addition, low economic
capacity is associated with a 0.21% increase in out-migration. Column seven replicates an analysis of the in-migration rate with Japanese nationals using disaggregated RCI. High community connectivity is associated with a 0.29% decrease in inmigration, while low community connectivity is associated with a 0.85% increase in
in-migration. In addition, low economic capacity is associated with a 0.26% increase
in in-migration, while low socio-demographic capacity is associated with a 0.21%
decrease in in-migration. The results show that the main reason for low migration in
municipalities with high resilience comes from high community connectivity, while
the main reason for high migration in municipalities with low resilience is both low
economic capacity and low community connectivity.
In columns six and eight, I replicate the same regressions by using the sample of
foreigners. The coefficient of socio-demographic capacity, economic capacity and
community connectivity are no longer significant in the sample foreign population.
Until this point, I examined the direct association of the Resilience Capacity
Index with the migration behaviours of Japanese nationals and foreigners. Table 8.4
shows the result of estimating specification two and examines whether the presence
147
more people move-out from and move-in to municipalities with low resilience, while
less people move-out from and move-in to municipalities with high resilience.
Table 8.3 presents the results of multivariate regressions of in- and out-migration
rate (specification 1). In the first column, the dependent variable is out-migration rate,
and the explanatory variables include prefecture fixed effects as well as a dummy
variable for high resilience, a dummy variable for low resilience, logged per capita
income, logged school density and logged population density. A dummy variable
for high resilience is associated with a 0.22% reduction in out-migration rate, while
a dummy variable for low resilience is associated with a 0.45% increase in outmigration rate. Column two shows the same regression in the sample of foreigners.
Unlike Japanese nationals, foreigners do not move-out more from municipalities
with low resilience. Similarly, they do not move-out less from municipalities with
high resilience.
The next two columns show the results of the analysis of in-migration among the
Japanese and foreign population. The in-migration rate among Japanese nationals
is 0.16% lower in municipalities with high resilience, compared to municipalities
with middle resilience. This decrease in the in-migration rate is smaller than the
decrease in out-migration (0.22%). Thus, higher resilience is associated with social
increase. On the other hand, the in-migration rate is 0.38% higher in municipalities
with low resilience, compared to municipalities with middle resilience. However,
this increase in in-migration is smaller than the increase in out-migration (0.45%).
Thus, low resilience is associated with social decrease. However, these associations
are not observed in the sample of foreigners.
From columns five to eight, I disaggregated resilience in RCI’s three sub-indices:
regional economic capacity, socio-demographic capacity and community connectivity capacity. Column five replicates the analysis of the out-migration rate with
Japanese nationals using the disaggregated RCI. High community connectivity is
associated with a 0.33% decrease in out-migration, while low community connectivity is associated with a 0.99% increase in out-migration. In addition, low economic
capacity is associated with a 0.21% increase in out-migration. Column seven replicates an analysis of the in-migration rate with Japanese nationals using disaggregated RCI. High community connectivity is associated with a 0.29% decrease in inmigration, while low community connectivity is associated with a 0.85% increase in
in-migration. In addition, low economic capacity is associated with a 0.26% increase
in in-migration, while low socio-demographic capacity is associated with a 0.21%
decrease in in-migration. The results show that the main reason for low migration in
municipalities with high resilience comes from high community connectivity, while
the main reason for high migration in municipalities with low resilience is both low
economic capacity and low community connectivity.
In columns six and eight, I replicate the same regressions by using the sample of
foreigners. The coefficient of socio-demographic capacity, economic capacity and
community connectivity are no longer significant in the sample foreign population.
Until this point, I examined the direct association of the Resilience Capacity
Index with the migration behaviours of Japanese nationals and foreigners. Table 8.4
shows the result of estimating specification two and examines whether the presence