equilibrium expected seat share given by (4.4) above, while each one of the small
parties in the opposition has an equilibrium expected seat share equal to (4.3)
above.
Equilibrium policies As for coalition governments, it is easy to verify that
targeted spending on public goods is not affected by the electoral rule, once we
control for the type of government. Specifically, the equilibrium expressions for
gSJ are identical to those in (3.7) for a single-party government under proportional
elections. Combining this with the earlier results for coalition governments, we
thus have a sharp and testable prediction: only the type of government, and not
the electoral rule shapes government spending on public-goods programs. Once
we hold constant the type of government, the electoral rule should have no inde-
pendent influence on this aspects of fiscal policy.
Political rents collected by single-party governments do depend directly on the
electoral rule, however, as in the case of coalition government. Repeating the steps
in the previous subsection, it is easy to verify that rSP = rCP, where the latter is
given by (4.5). Thus, a single party in government collects the same equilibrium
rents as each member of a coalition government, so that total political rents are
half of those collected under coalition government, just as it was the case under
proportional elections. The ambiguity over which electoral rule generates more or
less rents remains.
We summarize the foregoing discussion as follows:
Proposition 3
(i) Equilibrium public spending only depends on the type of government (coali-
tion vs. single party), as described by Proposition 1, and not on the electoral
rule or the number of parties in the opposition.
(ii) Equilibrium rents collected by each party in government only depend on
the electoral rule, not on the type of government or the number of parties in the
opposition. Under both electoral rules, total political rents collected by coalition
governments are higher than the rents collected by single-party governments. The
effect of the electoral rule on equilibrium rents is ambiguous, but greater within-
group voter mobility and smaller aggregate voter mobility make higher rents under
proportional elections more likely.
25