monetary policies than to contractionary ones (see Tinsley and Krueger, 1997). Under this last
assumption, if prices are not fully flexible, the existence of different regional responses will
not always imply a higher value of the variance of regional inflation than the national one. In
figure 6 the relationship between the standard deviation of regional inflation and the regional
indicator of relative responses to monetary shocks is represented. As it can be seen, the sign
of the relationship is negative, the opposite of the sign predicted by classical theories.
However, no strong conclusion can be extracted from this. Something similar happens when
the relationship between the average regional inflation and the indicator of relative response
to monetary policy is considered (see figure 7).
Figure 6. Relationship between the inflation standard deviation and the indicator of monetary
policy relative responses
,020
|
PV ’ - _ BAL EXT " " - _ CAN " " _ ° MAD AST " ES |
) CAT T ARA f*P-CA,ST-L NAV |
|
CAN MUR GA |
. CAST-M ^ RIO " " - VAL |
,018
,016
^° ,014
ω
⊂
.2 ,012
ГО
⅛
-= ,010
-20 -10 0 10 20
Indicator
INFsd = 0,02 - 0,1∙10-3∙IND + e R2=0,05
(0.4∙10-3) (0.07∙10-3)
So, which are the implications of the obtained results for EMU? In spite of the simplicity of
the analysis and the possible deficiencies of the proposed indicator, some facts can be
remarked: First, the theoretical and empirical literature reviewed in the paper suggests the
existence of important differences between regions in terms of responses to common
monetary shocks. Second, after elaborating a regional indicator of relative responses to
common monetary shocks using direct data of the determinants of asymmetries related to the
different transmission mechanism of monetary policy, a clear relationship exists between a
higher response to monetary policy and a higher variance of output for the Spanish case. Also,
16