Order Code RL31204
CRS Report for Congress
Received through the CRS Web
Fixed Exchange Rates,
Floating Exchange Rates,
and Currency Boards:
What Have We Learned?
Updated January 23, 2004
Marc Labonte
Analyst in Macroeconomics
Government and Finance Division
Congressional Research Service ˜ The Library of Congress

---

Fixed Exchange Rates, Floating Exchange Rates, and
Currency Boards: What Have We Learned?
Summary
Congress is generally interested in promoting a stable and prosperous world
economy. Stable currency exchange rate regimes are a key component to stable
economic growth. This report explains the difference between fixed exchange rates,
floating exchange rates, and currency boards/unions, and outlines the advantages and
disadvantages of each. Floating exchange rate regimes are market determined; values
fluctuate with market conditions. In fixed exchange rate regimes, the central bank
is dedicated to using monetary policy to maintain the exchange rate at a
predetermined price. In theory, under such an arrangement, a central bank would be
unable to use monetary policy to promote any other goal; in practice, there is limited
leeway to pursue other goals without disrupting the exchange rate. Currency boards
and currency unions, or “hard pegs,” are extreme examples of a fixed exchange rate
regime where the central bank is truly stripped of all its capabilities other than
converting any amount of domestic currency to a foreign currency at a predetermined
price.
The main economic advantages of floating exchange rates are that they leave the
monetary and fiscal authorities free to pursue internal goals — such as full
employment, stable growth, and price stability — and exchange rate adjustment often
works as an automatic stabilizer to promote those goals. The main economic
advantage of fixed exchange rates is that they promote international trade and
investment, which can be an important source of growth in the long run, particularly
for developing countries. The merits of floating compared to fixed exchange rates
for any given country depends on how interdependent that country is with its
neighbors. If a country’s economy is highly reliant on its neighbors for trade and
investment and experiences economic shocks similar to its neighbors’, there is little
benefit to monetary and fiscal independence, and the country is better off with a fixed
exchange rate. If a country experiences unique economic shocks and is economically
independent of its neighbors, a floating exchange rate can be a valuable way to
promote macroeconomic stability. A political advantage of a fixed exchange rate
regime, and a currency board particularly, in a country with a profligate past is that
it “ties the hands” of the monetary and fiscal authorities.
Recent experience with economic crisis in Mexico, East Asia, Russia, Brazil,
and Turkey suggests that fixed exchange rates can be prone to currency crises that
can spill over into wider economic crises. This is a factor not considered in the
earlier exchange rate literature, in part because international capital mobility plays
a greater role today than it did in the past. These experiences suggest that unless a
country has substantial economic interdependence with a neighbor to which it can fix
its exchange rate, floating exchange rates may be a better way to promote
macroeconomic stability, provided the country is willing to use its monetary and
fiscal policy in a disciplined fashion. The collapse of Argentina’s currency board in
2002 suggests that such arrangements do not get around the problems with fixed
exchange rates, as their proponents claimed.
This report will not be updated.

---

Contents
What Determines Exchange Rates? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Floating Exchange Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Hard Pegs and Soft Pegs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Currency Boards or Currency Unions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Economic Advantages to a Hard Peg . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Political Advantages to a Currency Board or Union . . . . . . . . . . . . . . . 8
Fixed Exchange Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Economic Advantages of a Fixed Exchange Rate . . . . . . . . . . . . . . . . 11
Political Advantages of a Fixed Exchange Rate . . . . . . . . . . . . . . . . . 11
What Have Recent Crises Taught Us About Exchange Rates? . . . . . . . . . . . . . . 12
Appendix: How Interdependent Are International Economies? . . . . . . . . . . . . . 20
List of Figures
Figure 1. Exchange Rates of Asian Crisis Countries . . . . . . . . . . . . . . . . . . . . . 14
List of Tables
Table 1. Differences in Types of Currency Arrangements . . . . . . . . . . . . . . . . . . 7
Table 2. Economic Interdependence of Selected Developed Countries
and Hong Kong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Table 3. Economic Interdependence of Selected Developing Countries . . . . . . 24

---

Fixed Exchange Rates, Floating Exchange
Rates, and Currency Boards:
What Have We Learned?
A prosperous world economy is beneficial to the American economy, especially
given our robust international trade sector, and it is thought to bring political benefits
as well, through its salutary effect on the political stability of our allies. Congress
plays a role in promoting a stable and prosperous world economy. Congressional
interest in currency exchange rates is twofold. First, Congress has an interest in
determining the most appropriate exchange rate regime for the United States to
promote domestic economic stability. Second, it has an interest in understanding and
influencing the exchange rate regime choices of other nations. Stable exchange rate
regimes are a key element of a stable macroeconomic framework, and a stable
macroeconomic framework is a prerequisite to a country’s development prospects.
The collapse of a fixed exchange rate regime was central to every important
international economic crisis since the mid-1990s — the 1994 Mexican peso crisis,
the Asian economic crisis of 1997, the Russian debt default of 1998, the Brazilian
devaluation of 1999, the Turkish crisis of 2001, and the Argentine crisis of 2002.
This reports evaluates the benefits and drawbacks of different types of exchange
rate regimes from the perspective of their effects on macroeconomic stability. It
focuses on three major types of exchange rate regimes: a floating exchange rate, a
fixed exchange rate, and “hard pegs,” such as a currency board or a currency union.
While there are permutations on these regimes too numerous to mention, a thorough
understanding of these three will allow the reader to understand any permutation
equally well. In the case of exchange rate regimes “one size does not fit all”;
different countries have very different political and economic conditions that make
some regimes more suitable than others.
What Determines Exchange Rates?
At times, the exchange rate is erroneously imagined to be an incidental value
that can be sustained by the good intentions of government and undermined by the
malevolence of greedy speculators. Economic theory holds it to be a value that is far
more fundamental. It is the value at which two countries trade goods and services
and the value at which investors from one country purchase the assets of another
country. As such, it is dependent on the two countries’ fundamental macroeconomic
conditions, such as its inflation, growth, and saving rates. Thus, it is generally
accepted that the value of the exchange rate cannot be predictably altered (for long)
unless the country’s macroeconomic conditions are modified relative to those of its
trading partners.

---

CRS-2
Many view the volatility of floating exchange rates as proof that speculation and
irrational behavior, rather than economic fundamentals, drive exchange rate values.
Empirical evidence supports the view that changes in exchange rate values are not
well correlated with changes in economic data in the short run.1 But this evidence
does not prove that economic theory is wrong. Although floating exchange rate
values change frequently, and at times considerably, there are important economic
conditions that change frequently in ways that cannot be measured. Factors such as
investors’ perceptions of future profitability and riskiness cannot be accurately
measured, yet changes in these factors can have profound influence on exchange rate
values. Economists have had more success at correlating long run exchange rate
movements with changes in economic fundamentals.
A decision by a government to influence the value of its exchange rate,
therefore, is likely to succeed only if its overall macroeconomic conditions are
altered. Government does have tools at its disposal to alter aggregate demand in the
short run — fiscal and monetary policy. Fiscal policy refers to increasing or
decreasing the government’s budget surplus (or deficit) in order to increase or
decrease the amount of aggregate spending in the economy.2 Monetary policy refers
to increasing or decreasing short-term interest rates through manipulation of the
money supply in order to decrease or increase the amount of aggregate spending in
the economy.3 For example, other things being equal, lower interest rates lead to
more investment spending, one component of aggregate spending. Furthermore,
fiscal and monetary policy influence interest rates differently, and interest rates are
the key determinant of the exchange rate. Expansionary fiscal policy is likely to raise
interest rates and “crowd out” private investment while expansionary monetary
policy, or reducing short-term interest rates, is likely to temporarily lower interest
rates.
Intervening in foreign exchange markets directly is equivalent to changing
monetary policy if the intervention is “unsterilized.” When a central bank sells
foreign currency to boost the exchange rate, it takes the domestic currency it receives
in exchange out of circulation, decreasing the money supply. Often, it prints new
money to replace the domestic currency that has been removed from circulation —
referred to as sterilization — but economic theory suggests that when it does so, it
negates the intervention’s effect on the exchange rate.4
If a government wishes to alter a floating exchange rate or maintain a fixed
exchange rate, it may do so by altering fiscal and/or monetary policy but only if it is
1 For example, see Robert Flood and Andrew Rose, “Fixing exchange rates: A Virtual Quest
for Fundamentals,” Journal of Monetary Economics, v. 36, n. 1, December 1995, p. 1.
2 For more information, see CRS Report RL30583, The Economics of the Federal Budget
Surplus, by Brian Cashell.
3 For more information, see CRS Report RL30354, Monetary Policy: Current Policy and
Conditions, by Gail Makinen.
4 Similarly, if exchange rate intervention was undertaken by a government’s treasury, theory
suggests it would have no lasting effect on the exchange rate because the treasury cannot
alter the money supply.

---

CRS-3
willing to abandon other macroeconomic goals such as providing stable economic
growth, preventing recessions, and maintaining a moderate, stable inflation rate. The
magnitude of response of the exchange rate to changes in monetary or fiscal policy
is not likely to be constant or predictable over time, but under most circumstances
policy can eventually lead to the desired result if it is truly dedicated to achieving it.
As discussed later, problems with exchange rates usually arise when a government’s
heart is not truly wedded to achieving its stated goal.
Floating Exchange Rates
The exchange rate arrangement maintained between the United States and all
of its major trading partners is known as a floating exchange rate regime. In a
floating exchange rate regime, the exchange rate is a price freely determined in the
market by supply and demand. The dollar is purchased by foreigners in order to
purchase goods or assets from the United States. Likewise, U.S. citizens sell dollars
and buy foreign currencies when they wish to purchase goods or assets from foreign
countries.5 The exchange rate is determined by whatever rate clears these markets.
Monetary and fiscal policy are not regularly or systematically used to influence the
exchange rate.6
Thus, when the demand for U.S. goods and/or assets rises relative to the rest of
the world, the exchange rate value of the dollar will appreciate. This is necessary to
restore balance or equilibrium between the dollar value exported and the dollar value
imported. Dollar appreciation accomplishes this through two effects on the United
States economy, all else being equal. First, it makes foreign goods cheaper for
Americans, which increases the purchasing power of American income. This is
known as the terms-of-trade effect. Second, it tends to offset the changes in
aggregate demand that first altered the exchange rate. The offset in demand may not
be instantaneous or complete, but it helps to make macroeconomic adjustment
possible if wages and prices are not completely flexible.
When foreigners increase their demand for U.S. goods, aggregate demand in the
United States increases. If the United States is in a recession, this increase in
aggregate demand would boost growth in the short run. If economic growth in the
United States is already robust, it would be inflationary — there would be too many
buyers (domestic and foreign) seeking the goods that Americans can produce. Under
a floating exchange rate, a substantial part of this increase in U.S. aggregate demand
5 The dollar is also widely used as an international medium of exchange for transactions that
do not involve American goods or assets. These transactions have no effect on the exchange
value of the dollar, however.
6 From time to time, governments and central banks in countries with floating exchange rates
may enter the foreign exchange market in an attempt to influence the exchange rate value.
This is known as “managed floating” or “dirty floating.” Historically, such interventions
have had patchy success. When they have failed, it has frequently been due to the fact that
intervention was not coupled with a change in monetary policy. Managed floating is very
different from a fixed exchange rate regime, where monetary policy is devoted to
maintaining the exchange rate value on a continual basis as its primary goal.

---

CRS-4
would be offset by the appreciation in the dollar, which would push U.S. exports and
the production of U.S. import-competing goods back towards an equilibrium level.
By reducing aggregate demand, an appreciating dollar reduces inflationary pressures
that might otherwise result.
Likewise, if the foreign demand for U.S. assets increased, foreign capital would
flow into the United States, lowering interest rates and increasing investment
spending and interest-sensitive consumption spending (e.g., automobiles). Absent
exchange rate adjustment, this would boost U.S. aggregate demand. But since the
greater demand for U.S. assets causes the dollar to appreciate, the demand for U.S.
exports and U.S. import-competing goods declines, offsetting the increase in demand
caused by the foreign capital inflow.7
Since floating exchange rates allow for automatic adjustment, they buffer the
domestic economy from external changes in international supply and demand. A
floating exchange rate also becomes another automatic outlet for internal adjustment.
If the economy is growing too rapidly, the exchange rate is likely to appreciate, which
helps slow aggregate spending by slowing export growth. While this is unfortunate
for exporters, overall it may be preferable to the alternative — higher inflation or a
sharp contraction in fiscal or monetary policy to stamp out inflationary pressures. If
the economy is in recession with falling income, the exchange rate is likely to
depreciate, which will help boost overall growth through export growth even in the
absence of domestic recovery.8
The maintenance of a floating exchange rate does not require support from
monetary and fiscal policy. This frees the government to focus monetary and fiscal
policy on stabilizing the economy in response to domestic changes in supply and
demand. Fiscal and monetary policy usually can be focused on domestic goals, such
as maintaining price and output stability, without being constrained by the policy’s
effect on the exchange rate.9 The drawback to fiscal and monetary autonomy, of
course, is that governments are free to pursue ill-conceived policies if they desire, a
particular problem for developing countries historically. Many times, a floating
7 This discussion assumes that changes in exchange rates are driven by changes in economic
fundamentals. To the extent that they are, floating exchange rates are an equilibrating force.
But if exchange rates are dominated by non-economic speculation — a proposal that
economists have not been able to rule out empirically — then movements in floating
exchange rates could be a destabilizing, rather than equilibrating, force. If this were true,
it would weaken the primary argument in favor of floating exchange rates. To the extent
that exchange rates may be driven by both non-economic speculation and economic
fundamentals, a fixed exchange rate could be superior, but only if governments could
promptly, correctly, and calmly adjust exchange rates when fundamentals changed.
8 Two seminal papers in favor of floating exchange rates are Milton Friedman, “The Case
for Flexible Exchange Rates, in Essays in Positive Economics, The University of Chicago
Press, (Chicago: 1953); and Harry Johnson, “The Case for Flexible Exchange Rates, 1969”
in Further Essays in Monetary Economics, Harvard University Press, (Cambridge: 1973).
9 The Treasury is often asked to explain its “dollar policy.” The most accurate explanation
would be that its policy is to use its macroeconomic tools to maintain domestic stability
rather than exchange rate stability.

---

CRS-5
exchange rate is forced to act as an outlet for internal adjustment because poor fiscal
and monetary policy have made adjustment necessary, causing stress on the trade
sector of the economy. This can be thought of as a political, rather than an economic,
drawback to floating exchange rates.
How valuable the macroeconomic adjustment mechanism that floating exchange
rates provide depends on the economic independence of the country. For countries
that are closely tied to others through trade and investment links, the ability to adjust
policy independently has little value — whatever is affecting one economy is
probably affecting its neighbors as well. For countries like the United States, whose
economy is arguably more affected by internal factors than external factors, flexible
exchange rates allow significant internal adjustment. Trade is still a relatively small
portion of American GDP: exports are equivalent to about 10% of GDP, in
comparison to a country like Malaysia or Singapore where exports exceed 100% of
GDP.
The economic drawback to floating exchange rates is that exchange rate
volatility and uncertainty may discourage the growth of trade and international
investment. Uncertainty can be thought of as placing a cost on trade and investment,
and this cost discourages trade. For example, after an international sale has been
negotiated, one party to the transaction will not know what price he will ultimately
receive in his currency because upon payment the exchange rate may be higher or
lower than when he made the trade. If the exchange rate has depreciated, he will
receive lower compensation than he had expected. The cost of this uncertainty can
be measured precisely — it is the cost of hedging, that is the cost to the exporter of
buying an exchange rate forward contract or futures contract to lock in a future
exchange rate today.10 If trade and foreign investment are important sources of
growth — especially for developing countries — as many believe it to be, floating
exchange rates may impose a real cost not just to exporters and investors, but to
society as a whole.
Hard Pegs and Soft Pegs
The alternative to floating exchange rates are exchange rate regimes that fix the
value of the exchange rate to that of another country or countries. There are two
broad types of fixed exchange rates. “Hard pegs,” currency boards and currency
unions, are considered first because they are the most stark example of a fixed
exchange rate arrangement. The second category considered is fixed exchange rates,
in which the link to the other currency or currencies is less direct, making them “soft
pegs.”
Currency Boards or Currency Unions
At the opposite end of the spectrum from floating exchange rates are
arrangements where a country gives up its exchange rate and monetary freedom
10 The cost of hedging may be higher in countries with small, undeveloped financial markets,
another reason why floating exchange rates may be less advantageous in small countries.

---

CRS-6
entirely by tying itself to a foreign country’s currency, what former IMF Deputy
Director Stanley Fischer calls “hard pegs.” This can be done through a currency
board or a currency union.11 A currency board is a monetary arrangement where a
country keeps its own currency, but the central bank cedes all of its power to alter
interest rates, and monetary policy is tied to the policy of a foreign country.12 For
example, Hong Kong has a currency board linked to the U.S. dollar. Argentina had
a similar arrangement which it abandoned in 2002, during its economic crisis. In
Argentina, for every peso of currency in circulation the Argentine currency board
held one dollar-denominated asset, and was forbidden from buying and selling
domestic assets. Thus, the amount of pesos in circulation could only increase if there
was a balance of payment surplus. In effect, the exchange rate at which Argentina
competed with foreign goods is set by the United States. Since exchange rate
adjustment was not possible, adjustment had to come through prices (i.e., inflation
or deflation) instead. Domestically, since the central bank can no longer alter the
money supply to change interest rates, the economy can only recover from peaks and
valleys of the business cycle through price adjustment.
From an economic perspective, a currency union is very similar to a currency
board. An example of a currency union is the euro, which has been adopted by 12
members of the European Union. The individual nations in the euro zone have no
control over the money supply in their countries. Instead, it is determined by two
factors. First, the European Central Bank (ECB) determines the money supply for
the entire euro area by targeting short-term interest rates for the euro area as a whole.
Second, how much of the euro area’s money supply flows to, say, Ireland depends
upon Ireland’s net monetary transactions with the rest of the euro area. For this
second reason, different countries in the euro area have different inflation rates
despite the fact that they share a common monetary policy.
In a currency union such as the euro arrangement, each member of the euro has
a vote in determining monetary policy for the overall euro area.13 This is the primary
difference from a currency board — the country that has adopted a currency board
has no say in the setting of monetary policy by the country to which its currency
board is tied. The countries of the euro also share in the earnings of the ECB, known
as seigniorage, just as they would if they had their own currency.
Not all currency unions give all members a say in the determination of monetary
policy, however. For instance, when Ecuador, El Salvador, and Panama unilaterally
11 For more information, see CRS Report RL31093, A Currency Board As an Alternative to
A Central Bank, by Marc Labonte and Gail Makinen,.
12 While perhaps theoretically feasible, it would be practically impossible to operate a timely
or precise enough fiscal policy to maintain a currency board or fixed exchange rate as long
as fiscal policy must be legislated. Thus, maintaining a fixed exchange rate has been
delegated to the monetary authority in practice.
13 Specifically, each country is represented on the ECB’s Governing Council, which
determines monetary policy. The ECB has operational independence from the European
Commission, EU Council of Ministers, and the national governments of the euro area, just
as the Federal Reserve has operational independence from the U.S. Congress and executive
branch.

---

CRS-7
adopted the U.S. dollar as their currency, they gained no influence over the actions
and decisions of the Federal Reserve. From a macroeconomic perspective, a
unilateral currency adoption and a currency board are indistinguishable. Between
these two arrangements, there are only two minor differences of note. First, currency
boards earn income on the dollar-denominated assets that they hold (another example
of seigniorage) while currency adopters do not.14 Second, investors may view a
currency union as a more permanent commitment than a currency board. If this were
the case, they would view the risks associated with investment in the former to be
lower.
Table 1. Differences in Types of Currency Arrangements
Independent
Role in
Circulation of
National
Setting
Seigniorage
National
Monetary
Monetary
Earnings
Currency
Policy
Policy
Currency
No
No
Yes
Yes
Board
Joint Currency
No
Yes
No
Yes
Union
Unilateral
Currency
No
No
No
No
Adoption
Economic Advantages to a Hard Peg. The primary economic advantage
of a hard peg comes through greater trade with other members of the exchange rate
arrangement. The volatility of floating exchange rates places a cost on the export and
import-competing sectors of the economy. Greater trade is widely seen to be an
engine of growth, particularly among developing countries. In a perfectly
competitive world economy without transaction costs, the cost of exchange rate
volatility could be very large indeed. For instance, since 1995 U.S. exporters and
domestic firms that compete with importers would have faced one-third higher prices
as a result of the (floating) dollar’s one-third appreciation against its main trading
partners. Until the domestic price level fell by one-third, U.S. producers would be
uncompetitive, all else being equal. Under a system of fixed exchange rates, U.S.
exporters would not have been placed at this price disadvantage, all else being equal.
In reality, for reasons not entirely clear to economists, the prices of tradeable goods
do not change as much or as quickly as the ideal would suggest, making the negative
effect of a floating exchange rate on trade smaller than expected.15 Between small
14 There have been congressional proposals to transfer seigniorage earnings to countries that
dollarize in order to encourage dollarization. For example, see H.R. 2617.
15 An overview of this extensive literature is given in Maurice Obstfeld, “International
Macroeconomics: Beyond the Mundell-Fleming Model,” National Bureau of Economic
Research working paper 8369, pp. 12-18, July 2001.

---

CRS-8
countries, a hard peg is also thought to promote more efficient and competitive
markets through lower barriers to entry and greater economies of scale.
Hard pegs also encourage international capital flows.16 The encouragement of
international capital flows can enhance a country’s welfare in a couple of ways.
First, it allows more investment to take place in areas where saving is relatively
scarce and rates of return are high, and investment is key to sustainable growth. This
makes both the borrower and the investor better off; the former because more
investment, and hence growth, is possible than otherwise would be, the latter because
they can now enjoy higher rates of return on their investment for a given amount of
risk than if limited to home investment. For developing countries, these investment
gains can be quite large. Because these countries have much lower capital-labor
ratios than the developed world, capital investment can yield relatively high returns
for some time if a friendly economic environment is constructed. On the other hand,
international capital flows can change rapidly in ways that can be destabilizing to
developing countries, as will be discussed below.
Weighed against the gains of higher trade and international investment is the
loss of the use of fiscal and monetary policy to stabilize the economy. For countries
highly integrated with their exchange rate partners, this loss is small. For example,
in the euro area, the business cycle of many of the “core” economies (e.g., Germany,
Netherlands, Belgium) have been highly correlated. As long as Belgium does not
face separate shocks from Germany, it does not lose any stabilization capabilities by
giving up the ability to set policy independently of Germany. By sharing a currency,
their fiscal and monetary policy can still be adjusted to respond jointly to shared
shocks to their economies, even if these shocks are not shared by the rest of the world
— the euro is free to adjust against the rest of the world’s currencies. Troubles only
arise if shocks harm one of these countries, but not its partners in the euro. In that
case, there cannot be policy adjustment for that country to compensate for the
shock.17
Political Advantages to a Currency Board or Union. The previous
explanation described the economic reasons for establishing currency boards or
currency unions. But it is probable that the primary reason for establishing them in
developing countries is based more on political reasons. As has been shown, these
16 Hard pegs encourage foreign investment for slightly different reasons than they encourage
trade. With trade, there is the danger under a floating exchange rate that a one-time
appreciation will make your exporters uncompetitive until domestic prices adjust. Since the
return on foreign investment is typically denominated in the foreign currency, a one-time
exchange rate depreciation would lower the profitability of the investment held at the time
of the depreciation. But it would have no effect on the profitability of new investment after
the depreciation had ended.
17 Although fiscal policy can still be used as an adjustment mechanism in countries with hard
pegs, there are constraints on its effectiveness in most of these countries. In the euro area,
countries are legally forbidden from running fiscal deficits greater than 3% of GDP
(although that rule has recently been flouted). In developing countries, fiscal policy is
constrained by the willingness of investors to purchase their sovereign debt, and investors
have proven much less willing to finance developing country deficits than deficits in the
developed world.

---

CRS-9
monetary arrangements tie the hands of their country’s policymakers. For some
countries, this is precisely what their policymakers are trying to achieve — a way to
prevent the reinstatement of policies from the “bad old days.” The most stark
example of the “bad old days” is the hyperinflation that many developing countries
experienced. For instance, in 1990, the year before Argentina adopted a currency
board, its inflation rate reached 2,314%. Stable growth is impossible when the price
mechanism has broken down in this way. The currency board quickly brought the
inflation rate in Argentina down to single digits. Whenever a country’s inflation rate
gets extremely high, it is a reflection of its fiscal policy. Large budget deficits cannot
be financed through the sale of debt instruments, so they are instead financed through
the printing of money. Thus, a currency board prevents irresponsible fiscal policy by
preventing monetary policy from supporting it.
Similarly, Ecuador “dollarized” in 2000 — adopting the U.S. dollar and largely
discontinuing the use of its own currency — at a time of economic crisis with the
hope that it would renew investor confidence. While extremely high inflation had
not yet become a problem, events leading up to dollarization appeared to be pointing
in that direction. The country’s banking system had collapsed, its economy had
shrunk by over 7% in 1999, low oil prices and natural disaster had caused budget
financing problems, and it had defaulted on some of its sovereign debt. Investors had
become very concerned that inflationary monetary policy would be used to solve its
fiscal problems, and dollarization quelled these fears by eliminating that policy
option.
Economic analysis sheds little light on the choice between floating exchange
rates and a currency board arrangement when the decision is motivated by the desire
to find a political arrangement that will prevent the pursuit of bad policies.
Economic analysis can identify bad policy; it cannot explain why it is pursued or how
to prevent its recurrence. A currency board is not the only way to tie the hands of
policymakers; various rules and targets have been devised to eliminate policy
discretion that could be used with a floating exchange rate. A currency board may
be a more final commitment, and hence harder to renege on, than rules and targets,
however. Then again, Argentina proved that even currency boards are not
permanent. In any case, the political problem of countries monetizing budget deficits
seems to be waning. In the late 1990s, the median annual inflation rate in developing
countries fell to 5%. If current trends continue, in the future there may be fewer
countries who find it advantageous to accept the harsh medicine of hard pegs to solve
their political shortcomings.18
Hard pegs are also seen by both proponents and opponents as a means to foster
political integration, a topic beyond the scope of this report. This was a primary
consideration behind the adoption of the euro.
18 Michael Mussa et al., “Exchange Rates in an Increasingly Integrated World Economy,”
IMF Occasional Paper 193, 2000, p. 17.

---

CRS-10
Fixed Exchange Rates
In a traditional fixed exchange rate regime, the government has agreed to buy
or sell any amount of currency at a predetermined rate. That rate may be linked to
one foreign currency or (unlike a currency board) it may be linked to a basket of
foreign currencies. In theoretical models, where capital is perfectly mobile and
investors consider all countries to be alike, fixed exchange rates would necessarily
be functionally equivalent to a currency board. Any attempt to unilaterally influence
one’s interest rates, through monetary or fiscal policy, would be unsustainable
because capital would flow in or out of the country until interest rates had returned
to the worldwide level.
In reality, results are not quite so stark. There are transaction costs to
investment. Investors demand different risk premiums of different countries, and
these risk premiums change over time. There is a strong bias among investors
worldwide, particularly in developed countries, to keep more of one’s wealth
invested domestically than economic theory would suggest.19 Due to these factors,
interest rate differentials, which should be theoretically impossible, are abundant.
For instance, interest rates in France and Germany should entail similar risks. Thus,
anytime French interest rates exceeded German rates, capital should flow from
Germany to France until the rates equalized again. Yet the commercial interest
reference rate, as measured by the OECD, between these two countries has varied by
as much as 1.61 percentage points between 1993 and the adoption of the euro in
1999.
As a result, countries with fixed exchange rates have limited freedom to use
monetary and fiscal policy to pursue domestic goals without causing their exchange
rate to become unsustainable. This is not true for countries that operate currency
boards or participate in currency unions. For this reason, these regimes can be
thought of as “soft pegs,” in contrast to the “hard peg” offered by a currency board
or union. But compared to a country with a floating exchange rate, the ability of a
country with a fixed exchange rate to pursue domestic goals is highly limited. If a
currency became overvalued relative to the country to which it was pegged, then
capital would flow out of the country, and the central bank would lose reserves.
When reserves are exhausted and the central bank can no longer meet the demand for
foreign currency, devaluation ensues, if it has not already occurred before events
reach this point.20 The typical reason for a fixed exchange rate to be abandoned in
crisis is due to an unwillingness by the government to abandon domestic goals in
favor of defending the exchange rate. Interest rates can almost always be increased
19 If individuals saw all countries as being equal, to achieve portfolio diversification the
average American investor would hold only about 1/4 of his wealth in American assets and
about 3/4 in foreign assets because the U.S. economy accounts for that fraction of the world
economy. Likewise, foreigners would hold 1/4 of their wealth in American assets and 3/4
abroad. In reality, Americans hold only about 1/10 of their wealth in foreign assets.
20 The problem is asymmetric. If a fixed exchange rate became undervalued, then capital
would flow into the country and the central bank would accumulate reserves. As long as the
central bank is willing to increase its foreign reserves, an undervalued exchange rate can be
sustained. This is believed to be the case with China from 2002-2003.

---

CRS-11
to a point where capital no longer flows out of the country, but great domestic
contraction may accompany those rate increases. It is not uncommon to see interest
rates reach triple digits at the height of an exchange rate crisis. Crises ensue because
investors do not believe that the government will have the political will to accept the
economic hardship required to maintain those interest rates in defense of the
currency.
Economic Advantages of a Fixed Exchange Rate. As with a hard peg,
a fixed exchange rate has the advantage of promoting international trade and
investment by eliminating exchange rate risk. Because the arrangement may be
viewed by market participants as less permanent than a currency board, however, it
may generate less trade and investment.
As with a hard peg, the drawback of a fixed exchange rate is that it gives the
government less scope to use monetary and fiscal policy to promote domestic
economic stability. Thus, it leaves countries exposed to idiosyncratic shocks not
shared by the country to which it has fixed its currency. As explained above, this is
less of a problem than with a hard peg because imperfect capital mobility does allow
for some deviation from the policy of the country or countries to which you are
linked. But the shock would need to be temporary in nature because a significant
deviation could not last.
The scope for the pursuit of domestic goals is greater for countries that fix their
exchange rate to a basket of currencies — unlike a hard peg, the country is no longer
placed at the mercy of the unique and idiosyncratic policies and shocks of any one
foreign country. One method for creating a currency basket is to compose it of the
currencies of the country’s primary trading partners, particularly if the partner has a
hard currency, with shares set in proportion to each country’s proportion of trade. If
the correlation of the business cycle with each trading partner is proportional to the
share of trade with that country, then the potential for idiosyncratic shocks to harm
the economy should be considerably reduced when pegged to a basket of currencies.
On the down side, baskets do not encourage as much bilateral trade and investment
as a peg to a single currency because they reintroduce bilateral exchange rate risk
with each trading partner.
Political Advantages of a Fixed Exchange Rate. In previous decades,
it was believed that developing countries with a profligate past could bolster a new
commitment to macroeconomic credibility through the use of a fixed exchange rate
for two reasons. First, for countries with inflation rates that were previously very
high, the maintenance of fixed exchange rates would act as a signal to market
participants that inflation was now under control. For example, inflation causes the
number of dollars that can be bought with a peso to decline just as it causes the
number of apples that can be bought with a peso to decline. Thus, a fixed exchange
rate can only be maintained if large inflation differentials are eliminated. Second, a
fixed exchange rate was thought to anchor inflationary expectations by providing
stable import prices. For a given change in monetary policy, it is thought that
inflation will decline faster if people expect lower inflation.
After the many crises involving fixed exchange rate regimes in the 1980s and
1990s, this argument has become less persuasive. Unlike a currency board, a fixed

---

CRS-12
exchange rate regime does nothing concrete to tie policymakers’ hands and prevent
a return to bad macroeconomic policy. Resisting the temptation to finance budget
deficits through inflation ultimately depends on political will; if the political will is
lacking, then the exchange rate regime will be abandoned, as was the case in many
1980s exchange rate crises. Thus burnt in the past, investors may no longer see a
fixed exchange rate as a credible commitment by the government to macroeconomic
stability, reducing the benefits of the fixed exchange rate.21 Furthermore, some
currency board proponents claim that this lack of credibility means that investors will
“test” the government’s commitment to maintaining a soft peg in ways that are costly
to the economy. By contrast, they claim that investors will not test a currency board
because they have no doubt of the government’s commitment.
For this reason, many economists who previously recommended fixed exchange
rates on the basis of their political merits have shifted in recent years towards support
of a hard peg. This has been dubbed the “bipolar view” of exchange rate regimes:
growing international capital mobility has made the world economy behave more
similarly to what models have suggested. As capital flows become more responsive
to interest rate differentials, the ability of “soft peg” fixed exchange rate regimes to
simultaneously pursue domestic policy goals and maintain the exchange rate has
become untenable. As a result, countries are being pushed toward floating exchange
rates (the pursuit of domestic goals) or “hard pegs” (policy directed solely toward
maintaining the exchange rate). In this view, while “soft pegs” may have been
successful in the past, any attempt by a country open to international capital to
maintain a soft peg today is likely to end in an exchange rate crisis, as happened to
Mexico, the countries of Southeast Asia, Brazil, and Turkey. Empirically, the trend
does appear to be moving in this direction. In 1991, 65% of the world’s 55 largest
economies used “soft peg” exchange rate arrangements; in 1999, the number had
fallen to 27%.22
Although the international trend has been towards greater capital mobility and
openness, it should be pointed out that there are still developing countries that are not
open to capital flows. The “bipolar view” argument may not hold for these countries:
without capital flows reacting to changes in interest rates, these countries may be
capable of maintaining a soft peg and an independent monetary policy. This has been
the case for China.
What Have Recent Crises Taught Us About
Exchange Rates?
The previous discussion summarizes the textbook advantages and disadvantages
of different exchange rate regimes. As such, it abstracts and simplifies from many
21 In this light, soft pegs based on a basket of currencies are typically viewed as a less
transparent arrangement involving less political commitment to maintaining discipline than
a soft peg based on a single currency.
22 Stanley Fischer, “Exchange Rate Regimes: Is the Bipolar View Correct?” Journal of
Economic Perspectives, v. 15, n. 2, Spring 2001, p. 9.

---

CRS-13
economic issues that may bear directly on real policymaking. In particular, it
neglects the possibility that crisis could be caused or transmitted through
international goods or capital markets, and the transmission role exchange rates can
play in crisis. The remainder of the report will be devoted to trying to glean some
general lessons from the international crises of the 1990s to enrich our understanding
of how different exchange rate regimes function. The primary lesson seems to be
that fixed exchange rate regimes are prone to crisis, while crisis is extremely
improbable under floating regimes. Unlike the crises of the 1980s, most of the
countries involved in 1990s crises — particularly Southeast Asia — had relatively
good macroeconomic policies in place (e.g., low inflation, balanced budgets,
relatively free capital mobility). Thus, crises cannot be blamed simply on policy
errors.
Fixed exchange rate
The primary lesson of the 1990s seems to be
regimes are prone to crisis
that fixed exchange rate regimes are prone to
because investors are compelled
crisis, while crisis is extremely improbable
to remove their money from a
under floating regimes.
country before it devalues. It is
similar to a fire in a crowded
theater: although everyone entered the theater in an orderly fashion, if everyone tries
to rush out at once, the doors jam and the fire becomes a catastrophe. Proponents of
fixed exchange rate regimes often argue that they can be adjusted if they “get out of
line.” But the weakness of fixed exchange rate regimes is that when economic
fundamentals change in such a way that devaluation becomes necessary, there is no
mechanism to devalue except crisis. Even if a government wanted to announce a
planned devaluation to avoid crisis, the announcement would likely spur anticipatory
capital flight as investors tried to withdraw their investments before the new
exchange rate was implemented.
Corruption, “crony capitalism,” and “greedy speculation” are not needed to
explain why fixed exchange rates collapse. The countries forced to devalue during
the Asian Crisis (Thailand, Malaysia, Philippines, Indonesia, and South Korea) had
very different economic structures and political systems, and were at different stages
of economic development, ranging from a per capita GDP of $15,355 in South Korea
to $4,111 in Indonesia.23 What they all had in common was their exchange rate peg
to the U.S. dollar. The Asian crisis was instigated by the fact that the appreciating
U.S. dollar, to which the crisis countries were fixed, had made their exports less
competitive and encouraged imports, particularly compared to China (which had
devalued its exchange rate in 1994) and Japan.
Investment bubbles, notably in property markets, seemed to be present in all of
the crisis countries, although there is no accepted method to identify them even after
the fact. Some argue that the bursting of these bubbles played a key role in
instigating the crises. Theories for why the bubbles formed include widespread state
allocation of capital, poor local financial regulation, and simple misguided
exuberance on the part of investors. Whether the bursting of such a bubble could
have instigated the crisis under a floating exchange rate is debatable. Some sharp
23 At purchasing power parity in 1997, before the crisis. Source: DRI-WEFA.

---

CRS-14
declines in asset prices have sparked serious crises and downturns, as was the case
in Japan in the early 1990s. Other times, sharp price declines have not caused crisis
and have had little lasting effect on the economy, as was the case with the United
States in 1987. But what is clear is that an asset bubble and a fixed exchange rate can
interact in ways more virulent than their individual parts. To the extent that asset
prices would have fallen in Asia to return to their fundamental levels anyway, the
presence of a fixed exchange rate ensured that it would happen suddenly because of
the “fire in a theater” principle. To the extent that a devaluation would have been
necessary anyway, the presence of an asset bubble assured that the outflows would
be larger, placing more of a strain on the countries’ financial systems.
Figure 1. Exchange Rates of Asian Crisis Countries
120
100
80
1=100)
60
40
Index (1997: 20
0
1997:1
1997:3
1998:1
1998:3
1999:1
1999:3
Year
Indonesia
Malaysia
Thailand
South Korea Philippines
Source: International Monetary Fund, International Financial Statistics, (Washington:
March 2000)
Note: Exchange rates are the market bilateral dollar exchange rate measured at the end of
the quarter.
When investors recognize a situation where devaluation becomes likely, even
though they may have had no intention of leaving a country otherwise, they have
every incentive to remove their money before the devaluation occurs because
devaluation makes the local investment worth less in foreign currency. Since the
central bank’s reserves will always be smaller than liquid investment when capital

---

CRS-15
is mobile, devaluation becomes inevitable when investors lose faith in the
government’s willingness to correct the exchange rate’s misalignment. To an extent,
the phenomenon then takes on the aspect of a self-fulfilling prophecy. The reason
the depreciation of a currency in crisis is typically so dramatic is because at that point
investors are no longer leaving because of economic fundamentals, but simply to
avoid being the one “standing when the music stops.”
Notice that in the textbook explanation, a currency depreciation is expected to
boost growth through an improved trade balance. In a currency crisis, this does not
happen at first, although it does happen eventually, because resources cannot be
reallocated towards increased exports quickly enough to compensate for the blow to
the economy that comes through the sudden withdrawal of capital. In the Asian
crisis, businessmen told of export orders they were unable to fill following
devaluation because their credit line had been withdrawn.
The shock of the capital outflow is exacerbated by the tendency for banking
systems to become unbalanced in fixed exchange rate regimes. When foreigners
lending to the banking system start to doubt the sustainability of an exchange rate
regime, they tend to shift exchange rate risk from themselves to the banking system
in two ways. First, foreign investors denominate their lending in their own currency,
so that the financial loss caused by devaluation is borne by the banking system.
Before devaluation, a bank’s assets might exceed its liabilities. With devaluation, the
foreign currency liabilities suddenly multiply in value with the stroke of a pen
without any physical change in the economy, and the banks become insolvent.24
Second, foreign lending to the banking system is done on a short-term basis so that
investments can be repatriated before devaluation takes place. This is problematic
because most of a bank’s investments are longer term. The banks then enter a cycle
where the short term debt is rolled over until crisis strikes, at which point credit lines
are cut. Both of these factors lead to a situation where a currency crisis causes a
banking crisis, which is a much more significant barrier to economic recovery than
the devaluation itself. These two characteristics both tend to be present when lending
to developing countries even in good times; the tendencies are accelerated when
booms look unsustainable. An exception may have been Brazil, which some
economists have suggested recovered so quickly from its devaluation because its
banking system had few short-term, foreign currency denominated assets.25
It is not necessarily illogical for the banking system to take on loans on a short
term basis or denominated in foreign currency when credit conditions tighten. If it
did not accept all forms of financing available to it, it could face insolvency at worst
and a significant contraction in business at best. If the banks believe that the
downturn is temporary and the episode will pass without a currency devaluation, then
the banks will be able to repay the loans once conditions improve. If devaluation
24 The insolvency problem occurs because, in practice, the bank does not denominate its
assets in terms of the foreign currency. This is presumably because its assets are domestic
loans.
25 Paul Krugman, “Crises: The Price of Globalization?” paper presented at Federal Reserve
Bank of Kansas City symposium, August 2000.

---

CRS-16
causes them to fail, they may expect the government to bail them out, perhaps
explaining their willingness to accept these currency risks.
These factors make it clear that once a country enters a currency crisis, there is
no policy response that can avoid significant economic dislocation. A policy to lower
interest rates to boost aggregate
demand and add liquidity to the
Once a country enters a currency crisis, there
financial system causes the
is no policy response that can avoid significant
currency to devalue further,
economic dislocation.
increasing the capital outflow
and exacerbating the banking
system’s insolvency. A policy
to raise interest rates in support of the currency exacerbates the economic downturn
brought on by crisis by reducing investment demand further. This too can feed
through to the banking system and capital markets by bankrupting significant
portions of the private sector. And it may not quell the currency crisis. In a textbook
analysis, interest rates can always be increased to attract back the capital leaving. In
reality, after a certain point higher interest rates increase default risk, perhaps causing
more capital flight than lower interest rates would bring.26
Both the Mexican crisis and the East Asian crisis were exacerbated by contagion
effects where crisis spread from country to country in the region. This cannot be
explained by an irrational (and degrading) assumption by investors that “all
Asians/South Americans are crooks.” Rather, it reflects the regional interdependence
of these economies. Although there is no a priori evidence that South Korea’s
currency was overvalued, it became overvalued once its neighbors were forced to
devalue. That is because its exports competed with its neighbors, and exports
accounted for a large fraction of its GDP. After its neighbors devalued, South
Korean exporters — already struggling because the Japanese yen had been
depreciating — could no longer offer competitive prices. Simultaneously, it appears
that investors’ perception of the riskiness of emerging markets in general greatly
increased, curtailing lending to South Korea, which placed pressure on interest rates
and investment.27 At this point, the deterioration in economic fundamentals caused
the Korean won to become overvalued, and currency crisis spread.
One may ask why the Bretton Woods fixed exchange rate system that fixed the
currencies of the major western economies from 1945-1971 was not prone to crisis
(at least before it collapsed). The reason is that capital mobility was largely curtailed
under the Bretton Woods system.28 Without capital mobility, central banks could use
26 Jason Furman and Joseph Stiglitz, “Economic Crises: Evidence and Insights from East
Asia,” Brookings Papers on Economic Activity 2, Brookings Institution, (Washington:
1998), p. 1.
27 Foreign portfolio investment in Korea fell from an inflow of $12,287 million in 1996 to
an outflow of $2,086 million in the fourth quarter of 1997, while foreign lending to banks
fell from an inflow of $9,952 million in 1996 to an outflow of $6,125 million in the fourth
quarter of 1997.
28 “Most countries in Europe did not restore (currency) convertibility until the end of 1958,
(continued...)

---

CRS-17
their reserves to accommodate small changes in fundamentals and could respond to
large changes in fundamentals with a (relatively) orderly devaluation. As long as
capital remains mobile — and almost nobody has supported a return to permanent
capital controls — the Bretton Woods arrangement cannot be replicated. It was not
long after capital controls were removed that the Bretton Woods system experienced
a growing number of currency crises in the 1960s and 1970s, leading to its eventual
demise.
Some economists argue that if short-term, foreign-currency denominated debt
is the real culprit in recent crises, then it makes more sense to address the problem
directly, rather than through the indirect approach of making it more costly through
a floating exchange rate. The problem could be addressed directly through various
forms of capital controls, financial regulations, or taxes on capital flows. They argue
that capital controls are necessary until financial markets become well enough
developed to cope with sudden capital inflows and outflows. Capital controls would
also allow countries to operate an independent monetary policy while maintaining the
trade-related benefits of a fixed exchange rate, similar to how the Bretton Woods
system operated. Yet capital controls deter capital inflows as well as capital
outflows, and rapid development is difficult without capital inflows. Capital controls
may make crises less likely, but they are also likely to reduce a country’s long run
sustainable growth rate.
That is not to argue that floating exchange rates are stable and predictable, as
some economists claimed they would be before their adoption in the 1970s. Rather,
it is to argue that their volatility has very little effect on the macroeconomy. For
example, the South African rand lost half of its value against the U.S. dollar between
1999 and 2001. Yet GDP growth averaged 2.8% and inflation averaged 5.4% in
those years. To be sure, when exchange rates change their value by a significant
amount in a few years, exporting and import-competing sectors of the economy
suffer. Manufacturing and farming are among those sectors in the United States. But
there is very little evidence to suggest that in a well-balanced economy such as the
United States, other sectors of the economy cannot pick up the slack when the
currency appreciates, especially when monetary policy is applied prudently. The one-
third appreciation of the dollar and record trade deficits between 1995-2000 did not
prevent the U.S. economy from achieving stellar growth and unemployment that at
one point dipped below 4%. While floating exchange rates sometimes move by
substantial amounts in a couple of years, they do not move by substantial amounts
overnight, as happens in fixed exchange rate crises. And that is the key reason why
floating exchange rates are not prone to financial and economic crises.
Floating and fixed exchange rates both impose costs on economies. Floating
exchange rates impose a cost by discouraging trade and investment. Fixed exchange
rates impose a cost by limiting policymakers’ ability to pursue domestic stabilization.
But there is a fundamental difference in the types of costs they impose. In most
28 (...continued)
with Japan following in 1964.... The restoration of convertibility did not result in immediate
and complete international financial integration...” Paul Krugman and Maurice Obstfeld,
International Economics, Addison-Wesley, (Reading, MA: 1997), p. 551.

---

CRS-18
countries, the cost of floating exchange rates is internalized and can be managed
through the market in the form of hedging.29 (Developing countries with
undeveloped financial systems may not be able to adequately hedge exchange rate
risk, however.) Part of the cost of fixed exchange rates is an externality and cannot
be hedged away. In other words, society as a whole bears some of the costs of fixed
exchange rate regimes, so that market participants do not take that cost into account
in their transactions. The costs that society bears are threefold. First, to the extent
that a country faces unique shocks to its economy, it gives up the ability to protect
its economy against these shocks. Those involved in international trade and
investment do not compensate society at large for the fact that the volatility of
aggregate unemployment and inflation has been increased. Second, the fixed
exchange rate regime is more prone to crisis, which further increases the probability
of high unemployment episodes. Even if floating exchange rates were to lead to
lower growth because they dampen the growth of trade and foreign investment, risk
averse individuals may prefer that outcome if it leads to fewer crises. Third, in some
historical instances, fixed exchange rates have weakened the banking system through
their incentives to take on debt that cannot be repaid in the event of devaluation. Of
the three factors, the last is the only one that could theoretically be rectified through
regulation, although implementing such regulation in practice could be difficult.
This is not to argue that fixed exchange rate regimes are never superior to
floating regimes. The United States would not be better off with 50 separate
currencies for each state even though it would ameliorate regional recessions. When
countries economies are interdependent enough, the benefits of fixed exchange rates
outweigh the costs: regions experience fewer unique shocks, labor mobility
improves, product markets may benefit from greater competition and economies of
scale, and capital market integration increases. But few countries meet this criterion.
Whether the countries of the euro zone become interdependent enough to make the
euro sustainable remains to be seen. At the time the euro was introduced, growth
between the “core” (countries like Germany and Italy) and the “periphery” (countries
like Ireland and Finland) were widely divergent, although they seem to have
narrowed since the euro was introduced.
But many developing countries that have adopted (or have considered adopting)
fixed exchange rates are not well integrated with the economy to which they are
linked (see appendix). That is because these countries are looking to link to the
world’s major “hard” currencies, the U.S. dollar, the euro, the Japanese yen, the
British pound, or the Swiss franc. Since they are often choosing to fix their exchange
rate to gain credibility (e.g., after an episode of high inflation), only a hard currency
would provide that credibility. But since the economies of most developing countries
are not closely tied to these hard currency economies, they are likely to face very
different economic shocks that require adjustment that would not be provided
through the policies of the hard currency countries to which they are tied. This
makes these countries more prone to boom and bust than they would be with a
(responsibly run) floating exchange rate. Certainly, Russia and the countries of East
29 If floating exchange rates do fluctuate irrationally as some economists have posited, this
imposes another cost on an economy, a cost that can be eliminated with no sacrifice by a
fixed exchange rate.

---

CRS-19
Asia and Latin America that were struck by currency crises in the 1990s were not
closely enough integrated with the U.S. economy to make a dollar peg sustainable.
Of these countries, only Mexico and the Philippines experienced growth that was
positively correlated with U.S. growth in the 1990s.
Proponents of currency boards argue that they do not suffer the vulnerabilities
of traditional fixed exchange rates because devaluation becomes too costly an option
for the government to consider. For that reason, they argue, investors have no
qualms about the safety of their money, and speculators know they cannot undermine
the currency so they do not try. The example of Argentina’s currency board
demonstrates why this argument is unpersuasive.30 In making this argument,
currency board proponents are only focusing on the political advantage to a currency
board — it makes profligate fiscal and monetary policy impossible. But this is not
the only factor that makes economies grow and investors choose them as an
investment location. A currency board eliminates currency risk, but it does nothing
to eliminate a country’s macroeconomic risk, to which investors are just as sensitive.
For example, there are good reasons why the overall U.S. economy would not be
significantly affected by the dollar’s one-third appreciation since 1995, but there is
no reason why the Argentine economy would be unaffected. It had not received the
large capital inflows or experienced the rapid economic growth that made the dollar’s
appreciation sustainable — some would argue, desirable — for the United States
despite its implication for exporters. Thus, Argentine’s exporters and import-
competing industries became uncompetitive in the last 5 years with no countervailing
factors to make other sectors of the economy competitive. In fact, developments to
the Argentine economy suggest a floating exchange rate would have naturally
depreciated in recent years to offset negative factors. The prices of commodities
(which are important exports for Argentina) had been falling, foreign investment to
developing nations had fallen since the Asian crisis, and Argentina’s largest trading
partner, Brazil, underwent a significant devaluation in 1998. Although the currency
board may have lowered political risk in Argentina, for these reasons, it greatly
increased macroeconomic risk, and that is why the currency board collapsed in 2002.
In the face of macroeconomic risk and political upheaval, Argentina proved that no
currency arrangement is permanent.
It is beyond the scope of this report to explore the question of whether
developing countries with a profligate economic past can make a credible new start
without fixing their exchange rates. Some economists go farther and suggest that in
today’s globalized economy, fixed exchange rates are no longer viable, and adopting
a foreign currency becomes necessary for a country trying to make a new start. In
those few cases where a natural currency union partner already exists, a fixed
exchange rate offers considerable economic advantages, particularly for a country
trying to overcome a profligate past. For all other countries, after considering the
experience of recent years, the economic advantages to floating exchange rates seem
considerable.
30 For more information on the Argentine economy, see CRS Report RL31169, Argentina:
Economic Problems and Solutions, by Gail Makinen.

---

CRS-20
Appendix: How Interdependent Are International
Economies?
The statement that some international economies are naturally suited for floating
exchange rate regimes while some economies are naturally suited for a fixed
exchange rate with a major trading partner is an uncontroversial statement among
economists. It is based on the insights first provided by Robert Mundell’s model of
an optimum currency area, which outlines the criteria that determine under what
circumstances a fixed exchange rate would succeed.31 This model underlines the
discussion of advantages and disadvantages presented in the first part of this report.
Controversy arises among economists on two points. First, it arises on the political
question of how important the political benefits of fixed exchange rates should be,
which cannot be addressed by the model. Second, it arises from the fact that the
empirical parameters of the optimum currency area model are not well established,
with economists disagreeing about how much integration is actually needed for a
fixed exchange rate to succeed.
This appendix attempts to offer some empirical evidence on the latter question.
It approximates a country’s interdependence with its largest trading partner based on
two key criteria from the optimum currency area model:
! How closely linked the two countries are through trade, measured as
exports to the trading partner as a percentage of GDP in 1999.
! The degree of correlation between the two countries’ business
cycles, measured as correlation of economic growth from 1990-
1999.32
The results are presented for selected developed countries and areas in Table 2 and
for selected developing countries in Table 3.
31 Robert Mundell, “A Theory of Optimum Currency Areas,” American Economic Review,
v. 51, September 1961, pp. 657-665.
32 Correlation is a measure of the typical relationship in the movement of two variables. It
is measured such that correlation equals -1 when the variables move in exactly opposite
direction, equals 0 when there is no relationship in the movement of the two variables, and
equals 1 when the variables move in exactly the same direction.

---

CRS-21
Table 2. Economic Interdependence of Selected Developed
Countries and Hong Kong
Exports to
Correlation of
Largest Trading
Country
Largest Partner
Growth with
Partner
(as % GDP)
Largest Partner
Australia
Japan
3.6%
-.54
Canada
U.S.
37.9%
.90
Denmark
euro area
16.6%
.36
Hong Kong
China
60.1%
.47
New Zealand
Australia
6.7%
.60
Norway
euro area
18.4%
-.12
Singapore
U.S.
25.9%
-.19
Sweden
euro area
18.1%
.57
Switzerland
euro area
22.6%
.68
United Kingdom
euro area
12.9%
-.12
Source: International Monetary Fund, International Financial Statistics
Note: Data are for 1999 except ; correlation is between 1990-1999; All countries in the table maintain
a floating exchange rate regime except for Hong Kong, which operates a currency board, and
Denmark, which operates a fixed exchange rate.
Using any specific cutoff point to define two economies as interdependent vs.
independent for either measure would be arbitrary, but one can see that many
countries do not achieve even the bare minimum of interdependence. Negative
growth correlation means that, overall, the business cycle in the largest trading
partner was typically moving in the opposite direction of the country for any given
year in the 1990s. Typically, this would put pressure on their exchange rates to move
in opposite directions as well. The disclaimer that past relationships do not imply
future causation is always needed in macroeconomics. Still, there is not much reason
for hope that two countries with negative correlation in the past would be sufficiently
positively correlated in the future to form a successful fixed exchange rate regime.
Similarly, it would be difficult to argue that the largest trading partner had a
significant effect on the country if the country’s exports were not equivalent to even,
say, 10% of the country’s GDP.
By these measures, of the countries in Table 2, Australia, Norway, and the
United Kingdom seemed poorly suited for a fixed exchange rate in the 1990s. To an
extent, these results reflect common sense observation, as all three countries
experienced shocks that were idiosyncratic from their major trading partner —
Norway’s economy is heavily (positively) influenced by oil prices; the major British
economic crisis of the 1990s was the devaluation of the pound, caused by the
disparity between its economic conditions and conditions in Western Europe; and

---

CRS-22
Australia is relatively physically isolated and not overly reliant on any particular
trading partner. A case could be made for a fixed exchange rate for the other
countries in the table; a strong case could be made for Canada, Sweden, Switzerland,
and Hong Kong (to China).
But a closer look at Canada suggests that a successful floating exchange rate
may not be incompatible even with a country as closely interdependent with its
neighbor as Canada is with the United States. Despite its interdependence, the
Canadian dollar experienced significant depreciation from 1991-1999, yet the
Canadian economy showed no obvious adverse effects. Between 1991 and 1999, the
Canadian dollar depreciated by 22% against the U.S. dollar in nominal terms. There
were 4 years when the currency depreciated by more than 5% in a single year. Yet
Canadian growth was strong from 1994 onwards, with the exception of 1996, and its
inflation rate was lower than the United States in all but 3 years from 1990-1999.33
Table 3 suggests many countries operate the opposite exchange rate regime
from what the optimum currency model would suggest — despite the fact that the
implementation of a particular exchange rate regime would be expected to create
economic conditions more amenable to the regime. Among developing countries,
a case could be made for a fixed exchange rate in the 1990s in Bulgaria, Hungary, the
Philippines, and Poland; of which the latter two actually operated a floating exchange
rate. A floating exchange rate seems more suitable in Ecuador, Saudi Arabia, Brazil,
Chile, Colombia, Pakistan, South Africa, and South Korea; of which the first two
operated fixed exchange rates. The remaining cases are more ambiguous.
One could object to the fact that growth correlation in the 1990s is artificially
low as a predictor of future growth because many developing countries underwent
economic crises in the 1990s that will be unlikely to be repeated in the future. In
particular, many of the countries listed in the table as maintaining a floating exchange
rate maintained a fixed exchange rate until crisis forced them to abandon their peg.
Countries to abandon their fixed exchange rate in crisis since the 1990s include
Mexico, Thailand, South Korea, Indonesia, the Philippines, Russia, the Czech
Republic, and Brazil. (Argentina and Turkey abandoned their currency pegs after
1999, the period covered in the table.34) This is a valid argument, but on the other
hand, one could argue that the countries arguably underwent these crises because
their economies experience very different shocks than the United States, to whom
they had all fixed their exchange rate. There is no obvious reason to think that these
idiosyncracies will disappear in the future.
33 Inflation is measured by the price deflator, as recorded by the IMF. The lower Canadian
inflation rate signifies that the real bilateral exchange rate depreciation was greater than the
nominal depreciation. In 3 of the 4 years when the currency depreciated by more than 5%,
Canadian inflation was lower than U.S. inflation.
34 Before the currency crisis, the Czech Republic fixed its exchange rate to a dollar-
deutschmark basket. Also, Malaysia and Hong Kong both experienced significant economic
dislocation during the Asian crisis, but Hong Kong maintained its currency board and
Malaysia maintained its peg after a devaluation.

---

CRS-23
Perhaps the most surprising result of the analysis was the number of countries
whose largest trading partner was not the country to which they have or previously
had fixed their exchange rate. According to the optimum currency area model,
Argentina should have fixed its exchange rate to Brazil instead of the United States,
Hong Kong should have fixed to China instead of the United States,35 Indonesia
should have fixed to Japan instead of the United States, Russia should have fixed to
the euro instead of the United States, and Brazil should have fixed to the euro instead
of the United States. This observation underlines the fact that exchange rate regimes
are often pursued as much for political reasons as economic reasons, and the
economic risks that political decision entails.
35 The Hong Kong problem is in fact solved since China also maintains a fixed exchange rate
with the United States.

---

CRS-24
Table 3. Economic Interdependence of Selected Developing
Countries
Exports to
Correlation of
Largest Trading
Country
Largest Partner
Growth with
Partner
(as % GDP)
Largest Partner
Argentina
Brazil
2.4%
.07
Peg
Bulgaria
euro area
24.2%
.30
Hard
Ecuador
U.S.
9.5%
-.43
China
U.S.
4.2%
.09
ate
Egypt
euro area
5.2%
.39
Hungary
euro area
37.3%
.64

Malaysia
U.S.
26.7%
-.41
Morocco
euro area
14.3%
.16
ixed Exchange R
Saudi Arabia
U.S.
6.7%
-.70
F
Turkey
euro area
11.3%
-.21
Brazil
euro area
1.8%
-.41
Chile
U.S.
5.6%
-.29
Colombia
U.S.
9.2%
-.20
Czech Republic
euro area
40.7%
0.0
India
U.S.
2.5%
.78
Indonesia
Japan
7.0%
.73
Israel
U.S.
12.8%
-.49
ate
Mexico
U.S.
27.1%
.14
Nigeria
U.S.
13.3%
-.52
Pakistan
U.S.
3.6%
-.38
Peru
U.S.
4.0%
.25
ing Exchange R
Philippines
U.S.
15.2%
.44
loat
F
Poland
euro area
17.2%
.40
Romania
euro area
17.6%
-.16
Russia
euro area
12.4%
-.25
South Africa
euro area
5.7%
-.10
South Korea
U.S.
8.7%
-.39
Thailand
U.S.
11.7%
-.55
Venezuela
U.S.
11.1%
-.68
Source: International Monetary Fund, International Financial Statistics
Note: Data are for 1999 except ; correlation is between 1990-1999 except for ex-Soviet bloc
countries, in which case correlation is between 1993-1999 to exclude transition period. Currency
arrangement is identified as of 1999; some countries have changed since. Israel and Venezuela
officially peg their exchange rates within a band; since this band is large (7.5% and 20%, respectively)
they have been classified in this report as maintaining floating exchange rates.

---