New Evidence on the Puzzles. Results from Agnostic Identification on Monetary Policy and Exchange Rates.

how results vary when applying the methodology to the analysis of monetary
policy shocks in US data. There, output seems to rise rather than fall, if the
restriction horizon is extended to one or even two years. This turns out to be
even more forcefully true here. It might be interesting to understand more
deeply, why this might be the case, but that discussion would be beyond
the scope of this paper. We therefore have chosen to omit results for other
choices for K .

4 Results

4.1 US Monetary Policy Shocks

For a US monetary policy shock, figure 4 contains the key results, while
figures 5 to 13 contain further details.

We have plotted all the impulse response functions for our benchmark
identification in figure 5 (two pages). The vertical lines in some of the impulse
response diagrams denote the sign restrictions we have imposed. Note that
- by construction - there is no liquidity puzzle and no price puzzle. In line
with Uhlig (2005), we find no significant effect on real output. We believe
that these are reasonable looking results, and shall thus turn to a discussion
of the results regarding the exchange rate response.

The first line of figure 4 shows the impulse response of the real exchange
rate to a US monetary policy contraction and should be compared to figure 1.
The second line shows the posterior distribution of the peak appreciation,
i.e. the distribution for the month containing the lowest point of an impulse
response drawn from the posterior and shown in the first line. The posterior
distribution for e.g. the peak of nominal exchange rate appreciation is very
similar to the posterior for the peak of the real exchange rate appreciation:
we therefore concentrate on the real exchange rate only.

The results show that the US-German and US-UK bilateral real exchange

16

<<   16   17   18   19   20   21   22   >>

More intriguing information