Optimal fiscal policy in a DSGE model with heterogeneous agents

This paper solves optimal scal policy in a simple real business cycle model with
agent heterogeneity. Introducing rule-of-thumb behaviour in some agents, I show
that the steady state optimal tax on capital in the long run should always be
zero regardless of the governments favouritism towards particular agents. Over
the business cycle, the inclusion of rule-of-thumb behaviour had a signicant ef-
fect on optimal stabilisation policy. The tax rate changes implemented to stabilise
the economy after shocks were contrary to those recommended by a representa-
tive agent version of the model . Following a productivity shock, the government
should nance a short term subsidy on capital with additional labour taxes. This
is at odds with the representative agent model that recommended a labour tax
reduction nanced by a short term tax on capital. It was also found that tempo-
rary increases to government expenditure should be nanced by short term taxes
on capital. Again this was contrary to the representative agent model which rec-
ommended that additional spending should be nanced with labour tax increases.
Whilst forming an insightful analysis, the study is not sucient for making policy
recommendations. Endogenising the agent heterogeneity and introducing imper-
fections such as nominal rigidities would go some way to developing a suitable
policy tool