Information Technology Reference
In-Depth Information
However, even in this symplified framework, a propensity to consume decreas-
ing with wealth (
c
3
=0
.
5) modifies the household's wealth distribution. Indeed,
in this case we can observe a larger wealth inequality: wealth distribution not
only presents a left tail, but also a right tail, thus the richest agent has a higher
wealth and the skewness becomes about zero (with a mean slightly higher than
the median of households' wealth); the larger inequality is also signaled by an
increase of the standard deviation; moreover, average wealth increases given that
richest households save more. This analysis on wealth distribution is summarized
in table 1.
Tabl e 1.
Statistics about wealth distribution at time T=500 in two simulation with
different value of parameter
c
3
:
c
3
=1and
c
3
=0
.
5
Statistic
c
3
=1
c
3
=0
.
5
Mean
1.38
1.61
Standard deviation 0.38
0.54
Skewness
-0.72
-0.08
Maximum
2.13
3.28
These results are robust both at different time steps (for instance we check the
wealth distribution also at time t=150) and in different simulations. Indeed, we
perform a Monte Carlo with 100 simulations on a time horizon
T
= 500 (again
skipping the first 100 periods, then we analyse the last 400 time steps).
Table 2 reports some relevant macroeconomic features of the two Monte Carlo
simulations with
c
3
=1and
c
3
=0
.
5.
We can observe that in both cases the economy falls in a large crisis scenario,
that is with a mean unemployment rate above 20%, 2 times over 100 simula-
tions and the mean unemployment rate is the same. However, the unemployment
volatility is much higher when
c
3
=0
.
5, that is whith larger wealth inequality,
and the difference is statistically significant at 99% level. Therefore, the business
cycle is “larger” with a lower minumum and a higher maximum for the unem-
ployment rate. Indeed, while in the baseline case (
c
3
= 1) we never find a time
step with unemployment above 20%, but for the two large crisis scenario, in the
inequality case (
c
3
=0
.
5) we detect 10 simulations in which the unemployment
peaks above 20% (the two large crises plus other 8 simulations). If the policy
maker considers the business cycle volatility as a problem to be stabilised, than
the reduction of the wealth inequality seems to be an effective tool to reach
this target. Especially, the policy maker could avoid large unemployment crises
(e.g., with an unemployment rate above 20%) by reducing inequality. For in-
stance, in an agent based macroeconomic setting, [3] find that more inquality
leads to higher volatility, increasing the likelihood of unemployment crises; they
also show that fiscal policy is an effective countercyclical tool especially when
income distribution is skewed towards profits.
It is worth to note that the percentage of firm defaults increases in the case
of
c
3
=0
.
5, probably due to higher macroeconomic volatility. Indeed, the mean
