Geoscience Reference
In-Depth Information
(Krugman 1991 ). Therefore, to the extent that large levels of interindustry
labor mobility are in place, this will imply the existence of larger economic
externalities across subnational units.
The in-depth study of the relationship between mobility and interregional
redistribution in both the European Union and Germany revealed that these
two variables are jointly endogenous. In an effort to isolate the impact of
mobility on interregional redistribution I follow a similar econometric strategy
as in the previous two sections. In a first equation I instrument mobility using
the share of population living in ethnic minority regions. The intuition behind
this instrument is that the geography of ethnicity is unlikely to be a function
of previous levels of interregional transfers but it is bound to affect directly
the patterns of labor mobility. Other things being equal, I expect workers
of a particular ethnic background to be less likely to move to other regions
dominated by competing ethnic groups. As a result, to the extent that the
minority ethnic region(s) harbor(s) large shares of the population, I expect the
overall levels of mobility to be lower. 17 All other exogenous variables in the
system are used as controls in the first equation. In turn, the predicted values of
inter-industry labor mobility are used as an independent variable in the analysis
of the following relationship:
mobility it +
ineq ir
it
f
(
IRR
) = α +
bp
(
R
) it +
+ θ it + ε
(8.5)
) it stands for the measure of representation discussed above, ineq ir
where bp
(
R
it
captures the geography of income inequality within the union as defined above,
and
θ it incorporates the same set of controls included in earlier analyses. In esti-
mating (8.5), I use the same battery of econometric techniques of the previous
two sections. Table 8.6 reports the results of the analysis.
The estimates of the determinants of interregional transfers offer a number of
interesting insights. Concerning the control variables, the proxies for economic
development, international exposure (country size), and quality of democracy
fail to show consistent and robust effects across more than one econometric
specification. Fortunately, the estimates of other variables' effects offer more
assurance. Ethnic fractionalization shows a positive and significant effect on
the level of interregional transfers in political unions around the world. This
result is consistent with the argument that fiscal transfers contribute to the
fiscal appeasement of mobilized minorities in ethnically fragmented societies
(Treisman 1999 ).
In turn, total taxation shows a negative and consistent impact on the level
of redistributive transfers of resources between regions. Note that the direc-
tion of the effect of this variable is opposite to the one identified when pre-
dicting interpersonal redistribution. This contrast between the impact of total
17 Note that this is a measure of the geography of ethnicity and not of overall ethno-linguistic
fractionalization. The latter does not pay attention to the regional distribution of ethnic groups.
Nevertheless, I keep it as a control variable in equation (8).
Search WWH ::




Custom Search