The European Debt Crisis and the Euro - Page 106

4. Conclusion14
Germany is now the world’s second largest exporter. Its exports have soared since it
joined the common currency in 1999. This paper seeks to understand the relationship between
Germany’s exports and its exchange rate.
The results indicate that there is a long run equilibrium relationship between Germany’s
aggregate exports, its real exchange rate, and income in importing countries. In every
specification the evidence implies that an appreciation of the German reer would reduce exports.
The results using the unit labor cost-deflated reer are precisely and robustly estimated and
indicate that a 10 percent appreciation would reduce exports by 6 percent. German exports are
also sensitive to income in the rest of the world.
Results using a panel data set indicate that consumption exports are much more sensitive
than capital goods exports are to exchange rate changes. German capital goods exports tend to
be high quality goods that compete more on quality than on price. Thus the fact that the price
elasticity is small is not surprising. The findings also indicate that German exports to the
eurozone are much more price elastic than German exports outside the eurozone.
Since Germany experienced a large internal devaluation against other eurozone countries
after 2000, one would expect exports from Germany to the eurozone to have increased. Figure 2
shows that there was indeed a surge in exports to the eurozone. This in turn contributed to the
huge eurozone imbalances that are evident in Figure 3.
Viewed from the perspective of capital flows, eurozone imbalances developed partly
because German banks became more willing to invest in peripheral eurozone countries after the
countries joined the common currency (see Gros and Mayer, 2012). However, following the
eurozone crisis, German banks and savers have become less willing to invest in Greece, Spain, 15
Portugal, Italy, and other deficit countries. Current account imbalances between European
countries may thus prove unsustainable.
In a flexible exchange rate system, part of the resulting adjustment would come through a
nominal appreciation of the German currency relative to the currencies of the deficit countries.
This is precluded as long as these countries share a common currency. Exchange rate adjustment
must come instead through a decline in wages and prices or an increase in productivity in the
peripheral countries vis-à-vis Germany. The evidence presented here indicates that these
adjustments will have to be large. Restoring equilibrium in the eurozone is thus likely to be a
long and painful process.

Conclusions
The results of this study suggest that exchange rate
variability reduces the volume of international trade in
manufactured goods. This conclusion differs from the
findings of previous empirical research,'2 which has often failed to uncover any significant impact of exchange risk
on trade. Admittedly, our conclusion is based on the
floating rate experience of only two countries, Germany
and the United States. Further empirical research on the
experience of a broader group of countries would be
necessary to reach more general conclusions on the
significance of exchange rate uncertainty.
Why do the findings in this study differ from those in
earlier studies? One obvious explanation would seem
to be our choice of an investigation period which covers
the more recent experience with floating exchange rates.
Including recent data is important for our results
because exchange rate volatility has shown no con-
sistent downward tendency over time and because it
provides a sufficiently long sample period with floating
exchange rates. This impression is confirmed by the
results obtained with data through 1978. Earlier
research has not investigated the period since 1977-78
and generally has mixed observations from the first few
years of floating with those from the fixed rate period
before 1973. Even the recent IMF study, mentioned
above, does not update previous econometric tests
dealing with effects of nominal exchange rate variability
on trade flows of individual countries; it does update,
however, one earlier investigation based on real
exchange rate variability. But we have argued that real
exchange rate variability is not an appropriate proxy for
exchange rate uncertainty.
Another reason for the differences in findings may be
that our measure of average quarterly variability, based
on a daily effective exchange rate index, provides a
better proxy for uncertainty than those in earlier studies
that were based on a very small number of observa-
tions, e.g., average quarterly variability calculated by
using three monthly observations. Finally, by explicitly
16 FRBNY Quarterly Review/Spring 1984
considering the impact of risk on volume through prices,
our study probably provides a better reading of the full
effect of exchange rate variability on trade.
We have argued that our estimates are likely to
underestimate the effects of exchange rate uncertainty
on trade for two reasons. First, measured exchange rate
variability may itself understate the extent of true
uncertainty and second, some indirect exchange risk
effects on trade cannot be separated from those of
exchange rate changes themselves. The indirect effects
are particularly important when long-range investment
decisions and choices of input sources or output mar-
kets must be made under the shadow of potentially
large future exchange rate changes. Our use of long
lags on the variability index may capture a part of these
long-term effects. But this procedure is not adequate for
fully isolating and measuring those effects. In any case,.
the main point of our theoreticaI arguments on uncer-
tainty is that the results in this study are best interpreted
as providing a lower bound on the effects of exchange
uncertainty on international trade.
One important policy implication of our study is that,
from the perspective of international trade, it is desirable
to reduce exchange rate uncertainty or variability.
Broadly speaking, variability may be reduced either by
changes in macroeconomic policies, by exchange
market intervention strategies, or by moving to a sub-
stantially different exchange rate system. A discussion
of such a complex and broad issue is obviously beyond
the scope of this study. But it should be noted that the
possible adverse effect of exchange rate uncertainty on
international trade is only one of several considerations
in the choice of an exchange rate system, and on other
grounds one may still favor the present exchange rate
arrangements.