FRBSF Economic Letter
2001-16; May 25, 2001
Monetary Policy and Exchange Rates in Small Open Economies
Controlling inflation is a key concern of central bankers around the
world. But how best to control inflation differs across countries according
to their individual characteristics; for example, small open economies
tend to import more goods as a percentage of GDP than larger, more closed,
economies, such as the United States. Recognizing this, several recent
papers have sought to clarify how the ingredients for successful monetary
policy in small open economies differ from those in larger, more closed,
economies. Of particular interest are whether central banks in small open
economies should respond to exchange rate movements and how the presence
of policy channels operating through the exchange rate affects the policymaking
process. This Economic Letter will explore this literature and
summarize the role the exchange rate plays in small open economies. It
will also spell out the arguments for and against policymakers responding
to exchange rate movements when setting policy.
Effects of exchange rate movements
Exchange rates are key variables in small open economies. Movements in
exchange rates directly influence the prices of goods that are traded
in world markets. Imported goods are both purchased by consumers and used
as inputs into the production process. As the economy's exchange rate
depreciates, the prices of imported consumption goods increase, directly
raising the consumers' price index. The price of imported intermediate
inputs also rises, raising firms' production costs. Higher production
costs also tend to culminate in higher consumer prices, as firms attempt
to pass on their higher costs to consumers through higher prices for their
final product. On the real economy side, the level of the real exchange
rate influences world demand for domestically produced goods. An appreciation
of the real exchange rate leads to less demand for domestically produced
goods, as both foreign and domestic consumers substitute cheaper foreign
goods for more expensive domestic goods. Falling domestic demand leads
to less pressure on prices to rise and may even lead to falling prices.
To target domestic inflation or consumer
price inflation?
Where exchange rates are volatile, this volatility can feed into consumer
prices through the direct effect of higher prices for consumer imports.
Consequently, a strategy of targeting consumer price inflation may lead
to considerable interest rate volatility, as policymakers attempt to neutralize
the exchange rate's impact on consumer prices. Interest rate and exchange
rate volatility may, however, be damaging if they lead to swings in real
output. Consequently, it is sometimes argued that small open economies
should target a measure of inflation that excludes the prices of tradable
goods, or that, at the very least, policymakers should look beyond exchange
rate fluctuations when setting policy (Ball 1999). Of course targeting
a measure of domestic inflation that excludes exchange rate effects is
easier said than done: The prices of many goods are influenced by exchange
rates, not just those that are physically traded in world markets.
But distinguishing between tradable and nontradable goods is relatively
easy to do in simulation studies. Such studies do tend to find that targeting
consumer price inflation can create considerable interest rate and exchange
rate volatility (Svensson 2000, Dennis 2000), but so too can targeting
inflation in nontradable goods. Large swings in interest rates and exchange
rates are associated more with targeting inflation too closely than they
are with the measure of inflation targeted. If policymakers are concerned
about the real effects of policy and are consequently prepared to permit
inflation to return to its target rate more slowly, then interest rate
volatility and exchange rate volatility are naturally dampened. Only when
policymakers target inflation to the virtual exclusion of everything else
do important differences between targeting domestic inflation and targeting
consumer price inflation arise (Ball 1999).
Should policymakers respond to the exchange
rate?
Regardless of which measure of inflation is targeted, there is an important
issue of whether central banks in small open economies should respond
to exchange rate movements when setting short-term nominal interest rates.
That is, if a central bank sets interest rates using a rule, would a rule
that included some measure of the exchange rate produce better economic
outcomes? Responding to exchange rate movements is, of course, very different
from targeting the exchange rate. The latter makes the exchange rate a
policy goal, whereas the former treats the exchange rate as one further
piece of information to be weighed when setting interest rates. Expressed
in this way, the answer seems obvious. Policymakers should respond to
all information that helps them meet their policy goals and achieve better
economic outcomes. If the exchange rate provides information that is timely
and relevant, then policymakers should make full use of this information
when they set interest rates. So, if it can produce better economic outcomes,
why does responding to exchange rates raise any controversy?
The arguments against responding to exchange rate movements fall into
two broad categories. In the first category, policymakers set interest
rates using a set of variables, and these variables are also the driving
forces behind exchange rate movements. Provided the central bank is efficiently
using the information at its disposal, there is no additional value to
be had from also responding to the exchange rate itself; the exchange
rate contains no extra information. In the second category, policymakers'
incomplete understanding of how the economy operates is manifest in uncertainty
about how the economy responds to exchange rate movements. This uncertainty
can place policymakers in awkward situations in which they mistakenly
respond in precisely the wrong way to exchange rate movements: tightening
when they should be loosening and vice versa. Consequently, it is argued,
the information value in exchange rate movements is in fact very low and
a prudent approach would suggest that policymakers should ignore exchange
rate movements and set policy on the basis of more reliable indicators.
Into the first category fall Clarida, et al. (2001) and Batini, et al.
(2001). While these two papers differ greatly in their level of disaggregation
and overall complexity, they arrive at similar conclusions. Provided policymakers
implement a rule in which interest rates respond to expected future inflation
there is no additional need (Clarida, et al. 2001), or very little additional
need (Batini, et al. 2001), for interest rates also to respond to exchange
rate movements. In effect, the expectation of future inflation encompasses
the exchange rate, leaving little or no need for an additional response
to exchange rate movements. Despite their results, neither paper suggests
that policymakers should not respond to exchange rate movements in some
way, shape, or form. In fact, in both models policymakers are responding
to exchange rate movements. This response is not direct, but through the
exchange rate's effect on expected future inflation.
The papers by Leitemo and Söderström (2001) and Guender (2001)
fall into the second category. Leitemo and Söderström (2001)
argue that because we are uncertain about how exchange rates are determined,
policy rules developed in the context of one exchange rate model may perform
poorly if that model proves incorrect. For some exchange rate models it
may be appropriate for policymakers to tighten in response to a depreciating
exchange rate, whereas for others it may be appropriate to loosen. Not
knowing the correct exchange rate equation can result in policymakers
loosening when they should be tightening and vice versa. The consequence
can be excessive interest rate and exchange rate volatility that can easily
be avoided through simply not responding to direct movements in exchange
rates. Guender (2001), in a closely related result, shows that uncertainty
about whether to tighten or loosen in response to a depreciating exchange
rate can occur even with the correct exchange rate equation. If there
is uncertainty about other aspects of the economy, then policymakers may
respond to exchange rate movements inappropriately, with the economy suffering
as a result.
Wrapping
up
So where does our discussion of these four papers take us? Collectively
these studies underscore that whether the exchange rate provides important
channels for monetary policy and whether it should feature in the policy
rule are conceptually different issues. However, in the absence of model
uncertainty, each of the four papers discussed suggests that including
a measure of the exchange rate in the policy rule leads to economic outcomes
that are better, or at least no worse, than if the exchange rate is ignored.
In this respect these four papers are in agreement with Ball (1999), Svensson
(2000), and Dennis (2000). However, once model uncertainty is introduced
the issue becomes clouded and it may indeed be genuinely disadvantageous
for policymakers to employ rules that contain the exchange rate. What
is unclear, however, is whether the forms of uncertainty that lead to
poor outcomes with rules based on exchange rates are themselves reasonable.
In summary then, small open economies that target inflation tend to target
consumer price inflation. In principle, trying to set interest rates to
offset the imported inflation component of the consumers' price index
can lead to considerable volatility in interest rates and exchange rates.
In practice this volatility can be avoided by not targeting inflation
too closely, that is, by eliminating deviations between inflation and
its target rate gradually rather than trying to keep inflation at target
period by period. Whether central banks in small open economies should
apply rules that include some measure of the exchange rate remains open
to debate. Exploiting information contained in exchange rate movements
may lead to better economic outcomes. But these better outcomes may be
hard to realize, and trying to do so may prove destructive, unless policymakers
are confident of the economic environment in which they operate.
Richard Dennis
Economist
References
Ball, Laurence. 1999. "Policy Rules for Open Economies." In
Monetary Policy Rules, ed. J. Taylor. Chicago: University of Chicago
Press.
Batini, Nicoletta, Richard Harrison, and Steven Millard. 2001. "Monetary
Policy Rules for an Open Economy." Bank of England Working Paper.
http://www.frbsf.org/economics/conferences/0103/index.html
Clarida, Richard, Jordi Gal', and Mark Gertler. 2001. "Optimal
Monetary Policy in Open Versus Closed Economies: An Integrated Approach."
Universitat Pompeu Fabra Working Paper.
Dennis, Richard. 2000. "Optimal Simple Targeting Rules for Small
Open Economies." Federal Reserve Bank of San Francisco Working Paper
No. 2000-20.
http://www.frbsf.org/econrsrch/workingp/2000/wp00-20.pdf
Guender, Alfred. 2001. "On Optimal Monetary Policy Rules and the
Role of MCIs in the Open Economy." University of Canterbury Discussion
Paper 2001-03.
http://www.econ.canterbury.ac.nz/dpapers/dp2001.htm
(accessed May 14, 2001).
Leitemo, Kai, and Ulf Söderström. 2001. "Simple Monetary
Policy Rules and Exchange Rate Uncertainty." Sveriges Riksbank Working
Paper.
http://hem.passagen.se/ulfsoder/
(accessed May 14, 2001).
Svensson, Lars. 2000. "Open-Economy Inflation Targeting." Journal
of International Economics 50, pp. 155-183.
Opinions expressed in this newsletter do not necessarily reflect
the views of the management of the Federal Reserve Bank of San Francisco
or of the Board of Governors of the Federal Reserve System. Editorial
comments may be addressed to the editor or to the author. Mail comments
to:
Research Department
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120
|
