Geoscience Reference
In-Depth Information
insurance because it is a policy that is fiscally and politically salient, and deals
with risks that vary across territories as much as across social groups. Other
programs, such as old-age insurance, deal with universal risks (Atkinson
1995
;
Varian
1980
), and hence are less prone to conflicts articulated along territorial
lines.
10
Thereafter,
Chapter 6
examines Germany's response to the challenge
of Reunification. Finally,
Chapter 7
analyzes the design of interpersonal and
interregional redistribution in the context of Spain's decentralization process.
As a complement to the case studies,
Chapter 8
provides a statistical evaluation
of the main implications of the argument.
The organization of the chapters around specific unions as opposed to
themes (interregional vs. interpersonal redistribution) or causal mechanisms
(representation, mobility) follows from the premise that it makes the material
more accessible to the non-specialized reader. However, the empirical evalua-
tion of the theory rests on the lessons to be drawn across the five unions under
study. By way of summarizing the research strategy outlined in this chapter,
I finish with a discussion of the way the different chapters address the main
hypotheses of the argument and conclude with an overview of how the different
chapters contribute to the evaluation of the topic's leading hypotheses.
Hypothesis 1: In the absence of high levels of regional economic specialization,
large levels of mobility contribute to the homogeneization of redistributive
preferences across regions and, through this route, to the centralization of
interpersonal redistribution.
Two chapters speak directly to this hypothesis. The inquiry into the rea-
sons behind the varying responses by Canada and the United States to the
Great Depression provides a good setup to evaluate this expectation: both
countries confronted the Depression from fairly similar starting points in terms
of both political representation and the geography of income inequality, and
yet Canada decided to centralize unemployment insurance whereas the United
States did not. Do the reasons behind this divergence reflect differences in
the patterns of crossregional mobility as predicted by the argument? In addi-
tion,
Chapter 8
offers a test of the claim that higher levels of mobility work
to homogenize distributive tensions across territories, thereby facilitating the
centralization of redistribution.
Hypothesis 2: The decentralization of interpersonal redistribution results from
the interaction between economic geography and political representation.
The empirical chapters of this topic provide insights on this hypothesis in
three different ways: (1) the natural experiments, in particular the contrast
10
Thus it is not surprising that old-age insurance is effectively the responsibility of the federal
government in all advances industrial democratic federations. Even in the highly fragmented US
welfare system, old-age became the responsibility of the federal government. Consistent with
this reasoning, it is equally noticeable that public health insurance in the United States is only
available to the very poor (Medicaid) and citizens over 65 (Medicare).
