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

tions used by Eichenbaum and Evans (1995) as well as Grilli and Roubini
(1995, 1996). Both specifications investigate countries pairwise, e.g. the
US and Germany or the US and the UK. Typical monetary policy variables
like short term interest rates and price levels are included, as are data on
industrial production and the exchange rate.

The advantage of the Eichenbaum-Evans or EE specification is the in-
clusion of the ratio of nonborrowed reserves to total reserves as a monetary
aggregate, which a number of researchers have argued to be closely related to
monetary policy choices, see e.g. Christiano and Eichenbaum (1992), Stron-
gin (1995). The advantage of the Grilli-Roubini or GR specification is that
it treats the US and the “foreign” country in a more symmetric manner,
since typically data on reserves are not available to the same extent for the
“foreign” countries.

When using conventional identification methods including more variables
implies a rising number of assumptions which become increasingly difficult
to justify. The advantage of identification via sign restrictions is that we do
not need many more assumptions. Hence, in a second step, we study the
robustness of our results and extend both benchmark VAR specifications by
adding possibly important variables, and analyze the resulting BIG VAR.

We study three country pairs: the US and Germany, the US and the
UK, and the US and Japan. We employ monthly data from 1975.07 to
2002.07. For a detailed description of the variables and the specifications,
see appendix 6.1.

To identify monetary policy shocks we shall impose that domestic price
variables as well as monetary aggregates like the the ratio of nonborrowed
to total reserves or the money stock do not rise. Thus, “price puzzles” are
avoided by construction. We do not impose a sign restriction on the reaction
of industrial production, since the results in Uhlig (2005) suggest that there
is little evidence that GDP will fall in reaction to a contractionary monetary

13

<<   13   14   15   16   17   18   19   >>

More intriguing information