COMPARISONS OF REAL EXCHANGE RATE VOLATILITY ACROSS EXCHANGE RATE SYSTEMS
| Date | 01 November 1992 |
| Author | Michael Bleaney |
| Published date | 01 November 1992 |
| DOI | http://doi.org/10.1111/j.1468-0084.1992.mp54004005.x |
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 54,4(1992)
0305-9049 $3.00
COMPARISONS OF REAL EXCHANGE RATE
VOLATILITY ACROSS EXCHANGE RATE
SYSThMS
Michael Bleaney
I. INTRODUCTION
This paper uses Deutschmark real exchange rates against currencies inside
and outside the European Monetary System (EMS) to make some general
points about comparing real exchange rate behaviour across exchange rate
regimes. In particular it argues that short-run measures of real exchange rate
volatility (such as the variance of monthly or quarterly movements) need to
be supplemented by longer-run measures that take into account the persis-
tence and mean-reverting tendencies in such movements. Empirical studies
have shown that short-run real exchange rate volatility is much smaller under
the EMS than under floating rates (Artis and Taylor, 1988; Cobham, 1989;
Macdonald and Zis, 1989; Ungerer et al., 1986). However, this need not
prevent a steady but persistent trend in intra-EMS real exchange rates which
might ultimately engender serious misaligmnent. For example, if a high-infla-
tion country uses the EMS to benefit from the anti-inflation credibility of the
Bundesbank, as some authors have suggested (Giavazzi and Pagano, 1986;
Melitz, 1988; Wyplosz, 1989), then the high-inflation country faces the
problem that, in order to preserve the punishment for excess domestic infla-
tion, it needs to allow its real exchange rate to increase if such excess inflation
occurs. It is possible to envisage circumstances in which this process would
generate more serious real exchange rate misalignments than might arise
from apparently more volatile exchange rate regimes which are not charac-
terized by persistence in real exchange rate movements. Moreover, if the real
exchange rate follows a random walk, its variance increases steadily with
time, which is not the case if there is reversion towards the mean. Thus both
persisence and mean-reversion are important factors in the medium to long-
run variability of real exchange rates which are not captured by short-run
measures. Essentially this is the fable of the hare and the tortoise. The hare
moved fast, but with a high variance; the tortoise moved slowly but with great
consistency and won the race.
The purpose of the paper is to demonstrate the significance of this point
using relatively simple descriptive measures of real exchange rate volatility,
and to employ these measures to compare real exchange rate volatility
557
To continue reading
Request your trial