Geoscience Reference
In-Depth Information
3
The Road Ahead
The Empirical Strategy
This topic studies the nature of redistribution in political unions. As reflected
in my hypotheses, summarized in
Chapter 2
, the nuts and bolts behind the
selection of fiscal structures lie in the interplay between economic geography
and political representation on the one hand, and the scope of mobility on the
other. Varying combinations of these factors are expected to yield different fis-
cal structures, and thereby different patterns of association between economic
geography, the scope of fiscal redistribution, and overall levels of inequality in
society.
The chapters ahead approach these mechanisms in two ways: first, as causal
factors whose marginal effects ought to be identifiable empirically
1
; second, as
elements of a dynamic process in which economic geography, mobility, repre-
sentation, and fiscal structures condition each other over time. The endogenous
nature of this process is a double-edge sword. Mapping out the process itself
becomes necessary to understand how the different elements of the argument
interact over time to shape redistribution in political unions. At the same time,
however, the very nature of the process makes identifying marginal causal
effects especially challenging.
To test the argument of the topic, the empirical chapters (
Chapters 4
to
8
)
develop a dual strategy that combines both natural experiments to facilitate the
causal identification of marginal effects and attention to endogenous processes
in two political unions with very different starting conditions. The natural
experiments include the North American responses to the Great Depression,
and Germany's response to Reunification. These cases pin down whether the
causal logic posited in this topic bears any resemblance with actual historical
experiences and illustrate the workings of the main mechanisms in the model.
I complement this study with attention to the evolution of contentions about
fiscal structures in the European Union and Spain. These unions fail to meet
1
On the requirements for causal identification and its implications for research design see Fearon
(
1991
); and Manski (
1995
).
46
