The Road Ahead - The Political Geography of Inequality - page 47

the requirements of a natural experiment, but allow me to capture the interplay
between the different elements of the argument over time. The contrast between
these two unions is also interesting because they vary in terms of the their
political and institutional status quo at the onset of the process. Finally, the
empirical strategy also includes a set of medium to large N statistical analyses
( Chapter 8 ) to transcend a handful of historical cases, and assess whether the
analysis applies more broadly to the universe of political unions.
This chapter motivates the selection of cases and explains how they serve the
purpose of testing the central contentions of the topic. The empirical evaluation
of my hypotheses requires the following conditions:
1. The formation of the union itself does not result from the distributive
conflicts of interest in this topic.
2. The changes in economic geography or in the patterns of mobility that
trigger the political conflict occur for reasons independent of the pre-
existing fiscal structures and system of representation.
3. The set of cases allows for sufficient variation in terms of the organization
of political representation.
4. Finally, the identification of the conditional relationship between eco-
nomic geography and political representation requires that, in the event
of an exogenous change in economic geography, the system of represen-
tation in place can be considered as given. Symmetrically, given a geog-
raphy of income inequality, causal identification demands that changes
in the system of representation follow from reasons independent of pre-
existing fiscal structures. The same requirement of exogenous variation
applies to the identification of the impact of mobility as well.
The remainder of this chapter discusses how the different cases stand in
relation to each of these conditions. The first section shows how all five unions
under consideration meet the first condition. The second focuses on the sources
of change in economic geography and their implications for political con-
tentions over the organization of redistribution (condition 2). Finally, the third
section analyzes the variation among the cases in terms of the centrifugal-
centripetal continuum, and the conditions under which political representation
can be considered exogenous (conditions 3 and 4).
UNION FORMATION AND FISCAL STRUCTURES
Political integration, understood as the decision to remain within the political
union or secede, conditions directly the size and shape of political unions. The
link between the choice to secede and conflicts over who gets what within the
union has been the focus of theoretical (Alesina and Spolaore 2003 ; Bolton and
Roland 1997 ) as well as empirical research (Bakke and Wibbels 2006 ). There
is little doubt that these two things are directly linked. In thinking about case