Geoscience Reference
In-Depth Information
judicial review, and political decentralization (Falleti
2005
; Hooghe, Marks
and Schakel
2008
; Rodden
2004
). All of these are dimensions of the territorial
distribution of power. Despite the fact that the causal arrows between them
may flow in multiple directions over time, not all of them necessarily co-vary
across space and time. This is partly because some aspects of decentralization
are part of the constitution and others not.
Constitutional rules and procedures regulate strategic interaction between
different levels of government. Independent judicial review systems and spe-
cial requirements governing constitutional reform tend to protect constitutions
from the instability of normal politics. In turn, the distribution of political
authority across levels of government reflects the outcome of such a bargaining
process. Ultimately, it reflects the equilibrium of forces within the federation
at a particular moment in time.
In between, extra-constitutional institutions, such as the party system, shape
the incentives of political elites and connect constitutional constraints to polit-
ical preferences and outcomes. Loosely following Filippov, Ordeshook and
Shvetsova (
2004
), these three layers correspond to three levels of institutional
analysis. In principle, constitutional rules and procedures are more resistant to
change, whereas political decentralization in different policy fields, including
redistributive policy, appears more malleable.
The possibility of change in the characteristics of the party system is likely
to fall between the two. As a result, the different dimensions of federations are
unlikely to
move
at the same time. They are not equally fluid. For this reason the
multidimensionality of political institutions becomes an asset worth exploiting.
The relationship between different elements of federalism and decentralization
can be approached by assuming that some of the other factors/dimensions are
given. The possibility that institutions can, over time, be endogenous as well
as the source of independent effects only becomes meaningful on the basis
of this assumption. Otherwise, partial equilibrium relationships could not be
identified. Hence the question: when can political representation be taken as
given?
The answer is simple, not very often: (1) when history provides natural
experiments in which unions with different systems of representation are
suddenly
exposed to
exogenous
shocks that require a change in the design
of their fiscal structures. Among the cases in this topic, these conditions apply
only to the situation created after the Great Depression in North America
and to the challenge of Reunification in Germany; (2) when the constitutional
design is independent of and prior to distributive conflicts over fiscal structures
that originate endogenously. These two situations differ in that they allow us
to capture empirically different aspects of the interaction between economic
geography and political representation.
Table
3.1
classifies the cases according to the nature of the system, or rep-
resentation, either when the exogenous shock occurs (
natural experiments
)or
at the onset of processes in which the different elements of the argument are
jointly endogenous over time (
endogenous processes
).
Table 3.1
conveys one
