-17-
appropriate in order to estimate a model which is parsimonious where possible. The LR
test for overidentified VARs suggests that our short-run restrictions cannot be rejected at
any conventional confidence level. The statistic is equal to 5.1 and the corresponding p-
value is 0.17.
The major result of our structural factor augmented vector autoregression
(SFVAR) model is that global liquidity shocks are driving forces of the global economy
and various national economies. Moreover, the outcomes of our empirical analysis are in
line with economic theory, since frequently emerging puzzles as, for example, the "price
puzzle" and the "liquidity puzzle" do not appear in our case. As can be seen in the charts
below, a global liquidity shock has a significant positive impact on global GDP after six
quarters. As always, the solid line in each chart represents the response to a one-standard
deviation shock, again measured in standard deviations. The dashed lines represent the
95% confidence intervals bootstrapped by ourselves based on a standard residual
bootstrap procedure with 500 draws (Enders, 2004).
Furthermore, global inflation responds significantly with a considerable time lag
of 11 quarters to a global liquidity shock. However, in contrast to our findings for global
GDP, the inflationary effect is far more persistent. Strong responses can also be found for
the common house price and the short-term interest rate factor. The global house price
factor rises strongly and persistently without any delay. This may indicate that excess
liquidity on a worldwide level has contributed to the phase of exceptionally high
increases in residential property prices across countries. Global liquidity shocks also lead
to a marked liquidity effect, driving short-term interest rates down by up to one standard
deviation.
All in all, our impulse response analysis seems to confirm the results found by
Rüffer and Stracca (2006) and Belke, Orth and Setzer (2008) on the basis of global
VARs. Common liquidity disturbances are influencing major macro variables on an
international level. However, with respect to asset prices, our results show some marked
differences versus previous empirical work. We are not able to find any significant
impact both on commodity and share prices, whereas Belke, Orth and Setzer (2008)
report a significant response of commodity prices after a global liquidity shock. A recent