The interesting di¤erence with respect to the benchmark is that when logrolling
is present, the high e¢ ciency party, may be induced into opportunistic behavior
by agreeing to an undesirable …scal policy in exchange for the electoral bene…ts
from reform.
Another point to discuss is the e¤ect of informational asymmetries.
3.3
Informational Issues
In contrast with models that require informational asymmetries to justify ine¢ -
cient policies, even under perfect information, politicians still block good policies
for political reasons. Uncertainty gives more credence to the story: Suppose that
parties have conducted research on the potential bene…ts and costs of a given
reform. If voters believe that one party has better chances of successfully imple-
menting reform, asymmetrical political gains arise. Since information is private,
the low e¢ ciency party might underestimate the value of the reform, while the
high e¢ ciency party might overestimate them.
In order to study the e¤ects of uncertainty and informational asymmetry,
a simple extension is presented in which uncertainty and informational asym-
metry are added to a …xed tax speci…cation.
15
In this extension I just allow
exogenous probabilities to exist for the two di¤erent parties and make those
private information. The standard approach would be to make the probabilities
and endogenous process which may depend on a policy choice by the parties,
but those are complications that do not add to the explanation.
3.3.1
Extension 3: Informational Asymmetries
The tax rate is …xed as in the …rst extension of the model. There is an institu-
tional reform with uncertain outcomes: if reform is successful then the level of
institutional e¢ ciency increases to Z
S
> 1. If it fails, the level of institutional
e¢ ciency becomes Z
F
< 1. The probability of success depends on which party
gets elected, as they are in charge of implementation. The high e¢ ciency party
has a probability of successful implementation equal to q
H
while the low e¢ -
ciency party has a probability q
L
. Voters are risk neutral. Voters can correctly
observe which party is the high e¢ ciency party, but not the actual values of q
H
and q
L
.
The timing of events is as follows:
1) Nature determines the identities and abilities of parties. Parties observe
q
H
, q
L
, Z
S
and Z
F
. Voters observe Z
S
and Z
F
and know that q
H
> q
L
. 2)
Parties support or block reform, reform gets enacted if both parties support it.
3) Voters observe whether reform is enacted and elect a party to government.
4) The winner implements reform if it was enacted.
Since there is no class advantage, claim 12 holds: successful enactment of re-
form leads to electoral success and reform implementation by the high e¢ ciency
party.
1 5
A previous version studied informational asymmetries in the context of class advantage.
Results are similar but require additional restrictions.
20
Convention