Description
There has been significant interest in the classification of exchange rate regimes in order
to investigate a wide range of hypotheses. Studies of the effects of exchange rate regimes on crises and
other aspects of economic performance can have important implications for policy choices. The paper
provides a guide to the major new large data sets that classify exchange rate regimes and to critically
analyze important methodological issues.
Journal of Financial Economic Policy
Classifying international aspects of currency regimes
Thomas Willett Eric M.P. Chiu Sirathorn (B.J .) Dechsakulthorn Ramya Ghosh Bernard Kibesse Kenneth
Kim J eff (Yongbok) Kim Alice Ouyang
Article information:
To cite this document:
Thomas Willett Eric M.P. Chiu Sirathorn (B.J .) Dechsakulthorn Ramya Ghosh Bernard Kibesse Kenneth
Kim J eff (Yongbok) Kim Alice Ouyang, (2011),"Classifying international aspects of currency regimes",
J ournal of Financial Economic Policy, Vol. 3 Iss 4 pp. 288 - 303
Permanent link to this document:http://dx.doi.org/10.1108/17576381111182882
Downloaded on: 24 January 2016, At: 21:43 (PT)
References: this document contains references to 40 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 423 times since 2011*
Users who downloaded this article also downloaded:
Puspa Amri, Apanard P. Angkinand, Clas Wihlborg, (2011),"International comparisons of bank regulation,
liberalization, and banking crises", J ournal of Financial Economic Policy, Vol. 3 Iss 4 pp. 322-339 http://
dx.doi.org/10.1108/17576381111182909
Donatella Porrini, Giovanni B. Ramello, (2011),"Class action and financial markets: insights
from law and economics", J ournal of Financial Economic Policy, Vol. 3 Iss 2 pp. 140-160 http://
dx.doi.org/10.1108/17576381111133615
Komain J iranyakul, (2012),"Linkages between Thai stock and foreign exchange markets under
the floating regime", J ournal of Financial Economic Policy, Vol. 4 Iss 4 pp. 305-319 http://
dx.doi.org/10.1108/17576381211279280
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Classifying international aspects
of currency regimes
Thomas Willett
Claremont Graduate University and Claremont McKenna College,
Claremont, California, USA
Eric M.P. Chiu
National Chung-Hsing University, Taichung, Taiwan
Sirathorn (B.J.) Dechsakulthorn
DLA Piper, Los Angeles, California, USA
Ramya Ghosh
LeBow College of Business, Department of Economics, Drexel University,
Sacramento, California, USA
Bernard Kibesse and Kenneth Kim
Claremont Graduate University, Claremont, California, USA
Jeff (Yongbok) Kim
The Bank of Korea, Seoul, Republic of Korea, and
Alice Ouyang
Central University of Finance and Economics, Beijing, China
Abstract
Purpose – There has been signi?cant interest in the classi?cation of exchange rate regimes in order
to investigate a wide range of hypotheses. Studies of the effects of exchange rate regimes on crises and
other aspects of economic performance can have important implications for policy choices. The paper
provides a guide to the major new large data sets that classify exchange rate regimes and to critically
analyze important methodological issues.
Design/methodology/approach – The study surveys and critiques the literature and provides
theoretical analysis of major issues involved in classifying exchange rate regimes.
Findings – The study ?nds that all of the new data sets have problems but some have more
problems than others and several of them are substantial improvements on what was previously
available. It is also shown that the best ways to classify depend on the issue being addressed and that
for detailed studies variants of measures using the concept of exchange market pressure are the most
promising. Directions for future research are also discussed.
Originality/value – The paper makes researchers aware of the new data sets that are available and
discusses their strengths and weaknesses. It also presents original analysis of several of the major
conceptual issues involved in classifying exchange rate regimes.
Keywords Foreign exchange, International ?nance, International economics,
Open economy macroeconomics, International trade
Paper type Research paper
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JEL classi?cation – F31, F3, F41, F4, F
JFEP
3,4
288
Journal of Financial Economic Policy
Vol. 3 No. 4, 2011
pp. 288-303
qEmerald Group Publishing Limited
1757-6385
DOI 10.1108/17576381111182882
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
1. Introduction
There has beenconsiderable recent interest in the classi?cation of exchange rate regimes
in order to investigate a wide range of hypotheses including the relationships of
alternative exchange rate regimes with in?ation, growth, and currency crisis, etc.
It might seem that exchange rate regimes should be an easy type of policy to classify.
After all exchange rates are highly visible and data on them is easily available for most
countries. On closer investigation, however, it turns out that while classifying some
types of regimes such as of?cial crawling pegs is easy in other cases such as attempting
to distinguish between managed ?oaters and unof?cial crawling bands can be quite
hard. As another example there has been controversy about whether after the crisis of
1997-1998 most exchange rate regimes in Asia are better described as managed ?oating
or a continuation of a soft dollar standard (McKinnon and Schnabl, 2004;
Willett et al., 2010.)
A major dif?culty is that for many purposes we would like to classify regimes
by how heavily they are managed by governments. Typically we cannot tell this by
looking at the behavior of the exchange rate alone. Low volatility could be due to heavy
management or to an absence of large shocks. In Section III we discuss how the concept
of exchange market pressure (EMP) can be applied to address this issue.
Early empirical studies used the old IMF classi?cations based on countries’ of?cial
reports. It has become widely recognized, however, that such of?cial reports are often
misleading. Consequently, an important and active research agenda has been to produce
behavioral measures characterizing exchange rate policies based on what governments
actually do rather than what they say they do.
Unfortunately such efforts have often reached con?icting classi?cations for particular
countries’ exchange rate regimes. For example, the newclassi?cations by the IMFand by
Reinhart andRogoff (RR, 2004) wronglyclassifythe Koreanexchange rate regime in2000
as being in their most ?exible categories. While there was considerable movement of the
exchange rate there was also substantial intervention (Willett and Kim, 2006.) Thus,
Korea actually had a managed ?oat. At the other extreme the study by Levy Yeyati and
Sturzenegger (LYS, 2005) classi?ed the regime as a ?xed rate.
Despite such dif?culties these recent efforts at large Nclassi?cations of exchange rate
regimes make substantial improvements over the old of?cial classi?cations. As we will
discuss, however, there is no one ideal method of classifying exchange rate regimes.
While we will indicate some pitfalls to avoid in developing classi?cations, it is important
to recognize that there is no one optimal measure. The best classi?cations may vary
depending on the type of question for which they are being used.
It is also important to recognize that the international aspects of currency policies
are not limited to just the degree of ?exibility of the exchange rate. One important
dimension is the extent to which developments in the foreign exchange market,
i.e. changes in exchange rates and the balance of payments, affect domestic monetary
policy. In this context it can make a huge difference whether or not intervention is
sterilized. Whether currencies are defended by capital controls can also be important.
(The measurement of capital controls is covered in the paper on ?nancial openness.)
In the following section we discuss strengths and weaknesses of the major recent
efforts at classifying exchange rate regimes that cover a large number of countries.
In Section III we discuss a number of the conceptual issues involved in studying
particular countries’ policies in more detail. In the longer version of this paper available
Classifying
currency regimes
289
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
on our web site we provide illustrations of such analysis for India, Korea, and Tanzania
and discussion of the many of the types of questions which classi?cations of currency
policies are used to study.
2. Reviews of major efforts at exchange rate classi?cations
2.1 The new IMF de facto measures
The IMF traditionally provided data on exchange rate regimes based on their of?cial
reports. Studies ?nd that in practice exchange rate regimes are often substantially
different from what governments said they were doing. Therefore, the IMF has
published classi?cations since 1999 based on the staff’s judgment of what kinds of
regimes are actually being followed. The original de facto classi?cations were back
dated to 1990 and are discussed in the IMF working paper by Bubula and O
¨
tker-Robe
(2002). The revised classi?cation is available since 2009 and was published in the 2009
Annual Report on Exchange Arrangements and Exchange Restrictions.
The IMF approach focused on whether a country targeted a path for the exchange
rate? This provided a substantial improvement on the old IMF classi?cation and in our
judgment became the best available large N set of classi?cations. Its main dif?culty
was in its classi?cations of the degree of management of ?exible rate regimes. These
were divided into three types. The most ?exible category was labeled “independent”
?oating and described as follows: The exchange rate is market determined; any foreign
exchange intervention aims at moderating the rate of change and preventing undue
?uctuations that are not justi?ed by the fundamentals, rather than establishing a level
for the exchange rate” (Bubula and O
¨
tker-Robe, 2002, p. 15.)This was contrasted with
“Tightly or Other Managed Floating” where:
The authorities in?uence exchange rate movements through interventions to counter the
long-term trend of the exchange rate, without specifying a predetermined exchange rate path,
or without having a speci?c exchange rate target (“dirty ?oating”) (Bubula and O
¨
tker-Robe,
2002, p. 15.)
Their analytical basis for distinguishing between “tightly” and “other” managed ?oating
was not clear. These three categories need not correspond to the degree of heaviness of
management. Even the distinction between independent and managed ?oating does not
seemclear, however, since “moderatingthe rate of change” and“countering the long-term
trend” canbothbe forms of “leaningagainst the wind” intervention. Inother words, under
both the IMF’s categories of managed and independent ?oating there could be heavy
or light exchange rate management. Since both Japan and Korea were classi?ed as
independently?oaters duringyears inwhich theyintervenedheavily, it is not clear that it
was useful to distinguish between these categories.
In part because of such concerns in 2009 the IMF revised its de facto classi?cations,
using only two categories of ?oating based on the degree of intervention. They also
distinguish more clearly between formal and informal pegs and crawls
(Habermeier et al., 2009.)
2.2 Studies that focus only on the behavior of the exchange rate
2.2.1 RR classi?cations. RR’s heroic effort of classi?cation presents the best currently
available large source for long-time period. Astrong advantage of these classi?cations is
that they use market rather than of?cial rates. Apart fromnot taking the degree of EMP
JFEP
3,4
290
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
into account, RR’s approach has dif?culty with short-lived de facto pegs. They classify
episodes as short-lived de facto pegs if the absolute monthly percentage change in the
exchange rate is equal to zero for four consecutive months or more. This decision rule is
too strict to detect actual short-lived de facto pegs. It is practically impossible for the
monthly percentage changes to be zero consecutively except for hard pegs such as
currency unions. In many studies, pegs are therefore assumed to have a certain level of
narrowbands. In order to discover long-lived de facto pegs, they use the probability that
monthly percentage changes in exchange rates is less than 1 percent over a rolling
?ve-year period. If it is larger than 80 percent, they classify the episodes as pegs.
This scheme cannot detect pegs with frequent breaks in exchange rates. Such two
methodologies in RR therefore have weakness in detecting short-lived de facto pegs.
RR develop a new category they call freely falling regimes. These are ?oating
regimes for countries with in?ation rates above 40 percent. This category is open to the
criticism that it biases results toward ?nding favorable macroeconomic performance
for ?oating regimes by throwing out cases of bad performance. On the other hand it
seems unlikely that the exchange rate regime would be a major cause of very high
in?ation rates (Chiu et al., 2011.)
RR use a special probability measure to distinguish the freely ?oating from the
managed ?oat. They construct an exchange rate ?exibility index:
1
Pð1 , 1%Þ
;
based on the probability that percentage changes of exchange rates over rolling
?ve-year period remain within 1 percent. RR compare the frequency distribution of the
index of each country with what they deem to be free ?oaters such as the US dollar
against the euro over the post-1973 period.
Their decision rule is very conservative for classifying managed ?oats. This
combined with the use only of the behavior of exchange rates results in misclassifying
some cases of managed ?oats as free ?oats. The cases of Korea and Japan who allowed
substantial exchange rate movements while also engaging in heavy intervention in a
number of years are important examples. Since free ?oats are commonly de?ned as
?oats with little or no of?cial intervention this problem could have been avoided by
labeling such cases as ?oats and deleting the “free”.
Another problem concerns their coarse classi?cation. They group
their ?ne classi?cations into ?ve coarse regimes to investigate their associations
with macroeconomic variables such as in?ation and growth. As suggested by
Angkinand et al. (2009), a major problemwith their grouping is that it fails to distinguish
between hard and soft pegs. They group (2) pre-announced peg, (3) pre-announced
horizontal band that is narrower than or equal to 2 percent, and (4) de facto peg with (1)
no separate legal tender all in the same category[1]. While this may be reasonable for
some purposes we have found that distinguishing between hard and soft pegs is
essential for analyzing the effects of exchange rate regimes on currency and ?nancial
crises and on macroeconomic discipline (Angkinand et al., 2009; Angkinand and Willett,
2011; Chiu et al., 2011).
2.2.2 Two-way classi?cations: Klein and Shambaugh (2010)
This major book focuses primarily on comparing the effects of ?xed versus ?exible
exchange rates. As we noted above such two-way distinctions may be useful
Classifying
currency regimes
291
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
for some purposes, but for issues such as discipline and currency and banking crises
such a limited breakdown is seriously ?awed. The same type of criticism applies to
the early studies of capital controls that used only zero-one dummies. (See the
accompanying paper on ?nancial openness and interdependence). Like the RR study
their method only employs the behavior of the exchange rate, and thus does not control
for shocks. The criteria for distinguishing between ?xed and ?exible rates are based on
whether the exchange rate stays inside a band. Speci?cally, ^2.0 percent bands
against base currencies are used for annual classi?cations. If the exchange rate change
does not exceed the threshold in 11 out of 12 months, it is classi?ed as the ?xed.
Emphasis on the choice of base currencies is a valuable contribution since the
automatic use of the dollar can be misleading for countries more oriented toward other
currencies such as the euro. Another question involves the extent to which governments
pay attention to different foreign currencies. While it is common to refer to a country’s
exchange rate against the dollar (or the euro) there are many exchange rates, one for each
other independent currency. Focusing on the exchange rate against just one currency
can give a misleading picture when there are substantial changes in cross rates among
other currencies. Anumber of countries peg more or less closely to a basket of currencies.
While the list of these currencies is sometimes made public, the relative weights given to
the different currencies is seldom made public. This has led to efforts to estimate the
implicit weights in such baskets (Frankel and Wei, 1994.) Where countries follow a
managed ?oat such estimation becomes more dif?cult but the issue retains its
importance (Frankel and Wei, 2008; Frankel and Xie, 2010)[2].
2.2.3 Statistical cluster analysis: LYS
LYS classify exchange rate regimes across 183 countries for 1974-2004. They
classify episodes into ?ve categories – ?exible, dirty ?oat, crawling peg, ?xed, and
inconclusive – using three classifying variables, exchange rate volatility, volatility of
exchange rate changes, and volatility of reserves. They use the volatility of the rate of
change of the exchange rate to identify crawling pegs.
Exchange rate volatility is measured by the average of the absolute monthly
percentage changes in the nominal bilateral exchange rate during a calendar year. The
volatility of exchange rate changes is measured by the standard deviation of the
monthly percentage changes in the exchange rate. To compute the volatility of reserves
they ?rst subtract central government deposits from net foreign assets and divide this
by the exchange rate. Then they compute its ?rst-order differenced value and divide
this by the lagged money base. The volatility of reserves is measured by the average
absolute monthly changes in this measure.
They use K-means cluster (KMC) analysis with three classifying variables. The
KMC algorithm produces groups such that they have the smallest total distance
between episodes and the center of the group. They divide each variable into just two
categories, high and low volatility.
While their analysis does have the advantage of being objective, they provide no
clear economic rationale for their particular formulations and the thresholds used, nor
do they explain why they do not consider the case of a highly volatile exchange
rate combined with low volatility of exchange rate changes and reserves. Additional
limitations of the LYS classi?cation are its inability to distinguish hard from soft
?xes and wide-band crawling pegs from dirty ?oat regimes. LYS classify regimes
JFEP
3,4
292
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
by calendar years. For years in which the regimes or the parity changes or there are
transitional periods of instability, it is unclear how the LYS calculations should be
interpreted.
3. Some major conceptual issues
3.1 The EMP approach
The concept of EMP introduced by Girton and Roper (1977) is a measure of the gap
between quantities demanded and supplied in the foreign exchange market at a
particular exchange rate. It provides a way of comparing pressures under alternative
exchange rate regimes by adding changes in reserves (as a measure of of?cial
intervention) and exchange rate changes[3]. Conceptually, these should be weighted by
the slope of the excess demand function in the foreign exchange market but in the
absence of good measures of these parameters, equal weights are often used. The
measure of the degree of ?exibility of an exchange rate regime then becomes the ratio
of EMP taken by changes on reserves versus changes in the exchange rate. Depending
on which way one formulates the ratio this will give a measure of the propensity of the
authorities to intervene in the foreign exchange market or its reciprocal, the degree of
exchange rate ?exibility. This approach is superior to just looking at exchange rate
changes alone since these movements are in?uenced by the size of shocks as well as the
propensity to intervene.
The basic idea of the EMP underlies many of the recent efforts at classi?cation of
exchange rate regimes. However, these studies often do not fully conform to the
concept of the proportion of EMP taken on the exchange rate versus intervention.
The ?rst explicit use of the concept of EMP to classify exchange rate regimes was by
Weymark (1995, 1997, 1998). She de?nes the degree of intervention as the proportion of
EMP absorbed by exchange market intervention. Weymark’s index has a range
from21to þ1. When the sign of changes in exchange rate and reserves is correct for
leaning against wind intervention, then Weymark’s index has a range from 0 to 1, with
values closer to 1 indicating a higher degree of ?xity. When the exchange rate changes
are of the same sign, but have a greater absolute magnitude than the changes that would
have occurred in the case of no intervention, Weymark’s index is negative. When the
exchange rate appreciates (depreciates) with excess supply (demand) of domestic
currency, Weymark’s index is greater than one. As was discussed in Section II the
interpretation of the degree of intervention in cases of wrong sign observations in
Weymark’s index is not straightforward. (Further discussion of technical aspects of
Weymark’s papers is discussed in the longer web version of this paper).
In more recent papers Calvo and Reinhart (CR, 2002), and Herna´ndez and Montiel
(HM, 2003) also drawon the concept of EMP to offer procedures for classifying exchange
rate regimes. Their approaches, however, suffer from several technical problems.
CR and HM both look at the volatilities of the exchange rate, international reserves and
interest rates to characterize exchange rate policies. CR look at the ratio of variances
while HM compare the volatilities of each variable across countries separately rather
than comparing their ratios. However, relative variances can be misleading when trends
are present. This is less of a problem when exchange rate variability is measured in
terms of the variability of changes rather than levels.
Furthermore, ratios of variances do not distinguish cases of wrong signs and as
CR note, variances may be distorted by outliers. Thus, they advocate an approach
Classifying
currency regimes
293
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
based on the frequency with which classifying variables remain within a threshold
limit, arguing that: the probabilistic nature of the statistic conveys information about
the underlying frequency distribution that is not apparent from the variance (p. 384)[4].
The choice of thresholds is of course somewhat arbitrary. CR use 1 and 2.5 percent for
monthly percentage changes. With the use of variances, however, this approach does
not distinguish between correct- and wrong-signed pairs of observations.
3.2 The problem of trends and wrong signs
As noted previously the conceptual basis of the EMP approach assumes governments
intervene to slow or halt exchange rate movements. Cases where intervention is used to
accelerate exchange rate ?exibility are thus not well de?ned. It is not easy to know how
these wrong sign observations should be treated. They can be caused by imperfections
in the reserve change proxy as well as by episodes of leaning with the wind, i.e. pushing
the rate even further than the market has been taking it or of extreme forms of leaning
against the wind where the rate is forced in the opposite direction from market forces.
Actual leaning with the wind in a downward direction is classic beggar thy neighbor
policy and is discouraged by the IMF’s guidelines except for cases where a currency is
judged to be seriously over-valued. Leaning with the wind in the upward direction may
be justi?ed during periods in which country’s currencies are considered to have
over-depreciated. The aftermath of the Asian crisis is a prominent example of such
over-depreciation.
Where imperfect proxies are the cause of the wrong sign, the best solution would
likely be to drop these observations. With true leaning with the wind, one could argue
either that this is a case of government management or that it should be considered as
super ?exibility. Which interpretation is better may depend on the use to be made of
the classi?cation.
Wrong signs could also result from an extreme form of leaning against the wind,
i.e. instead of allowing domestic currency to depreciate in the face of excess supply in
the foreign exchange market, the government could actively appreciate the exchange
rate. Likewise, despite an excess demand in foreign exchange market, a mercantilist
government could force down the exchange rate. It would seem that these should be
treated as cases of strong government management or of super ?xity. Thus, depending
on the cause of the wrong sign, it can be argued that a value of zero or one should be
assigned or that the observation should be deleted.
Trends in exchange rates and reserve changes also present substantial problems for
applying the EMP approach. A major dif?culty is that there is no clear-cut theoretical
rationale for how to compare the comparative degrees of ?exibility of a faster crawl
with a smaller band of ?uctuations around it with a slower rate of crawl but larger
?uctuations around it. For example, the former which cause less problems for an
export-import ?rm whose transactions have a short-time horizon but cause greater
problems for a manufacturing exporter with a much longer time horizon. Given this
problem Willett et al. (2007) argue that a two parameter approach is necessary to fully
characterize exchange rate regimes, one for trend relationships and one for ?uctuations
around trend.
This approach runs into the substantial operational dif?culty involved with
detecting changes in trends. There is no clear-cut rule for deciding when changes are
large enough to justify treating them as shifts in regime, and the most appropriate
JFEP
3,4
294
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
degree of sensitivity can vary with the purpose of the investigation. For the purpose of
deciding whether a country is following a managed ?oat or not taking a single trend line
over a long-time period is likely to be satisfactory, while if one is investigating whether a
country is changing its intervention strategy under a managed ?oat a much more micro
approach needs to be taken and one may need a number of short-termchanges in trends.
For this purpose we are inclined to favor a judgmental approach while admitting that
this is open to possible subjective biases. There are of course a number of formal tests for
structural breaks (Bai and Perron, 1998, 2003.) It is often not clear, however, which
aspects of behavior are responsible for the statistical results. (On this issue see the
discussion in Ghosh (2011) of the con?icting ?ndings for structural breaks in India’s
exchange rate policy).
3.3 Issues of constructing intervention proxies
One of the most dif?cult problems in applying the EMP approach is that few
countries make data on their exchange market intervention publicly available.
Most empirical studies use changes in reserves as a proxy but acknowledge that this is
far from perfect. Reserves can change due to interest earnings, valuation changes,
and of?cial borrowings as well as intervention. And interventions can include actions
in forward, not just spot markets[5]. These problems have led some researchers to give
up using reserve measures all together (Ghosh et al., 2003.) This seems too strong a
reaction, however, since where reserve accumulations are very large, such as those at
times by Japan and Korea, there can be little doubt that intervention was a primary
cause.
One adjustment that can be made fairly easily is to subtract an estimate of interest
earnings from the reserve ?gures[6]. Since dollar values of reserves are used in most
cross-national empirical studies, valuation changes due to exchange rate movements
can also be important. Since exchange rate changes among reserve currencies are often
much larger over short periods than interest rate differences, estimates of the currency
composition of reserves are particularly important for such calculations. For many
countries published ?gures or good estimates about the currency composition of
reserves are not available, but efforts to take valuation changes into account have been
made in some country studies (Ouyang et al., 2008.) Where countries mark their reserve
?gures to market, interest rate changes in reserve currency countries can also affect
valuation for longer maturity assets.
A ?nal problem regarding the intervention proxy is that a constant amount of
intervention per period in the same direction would give rise to varying percentage
changes as reserve levels rose or fell[7]. Likewise, initial reserve levels can make a
substantial difference. Studies such as CR (2002), and HM (2003) characterize Japan as
having a low volatility in foreign reserves. Although Japan intervened heavily in
absolute terms in the foreign exchange market in the early 2000s, the percentage
changes in reserves were fairly small due to the high-initial levels of reserves.
There are several ways to deal with this problem. The most popular method has
been to use scaling variables for intervention proxies. Weymark (1997) and LYS use
the lagged money base, lagged narrow money and the sum of exports and imports for
12 months as scaling variables. This is also a problem for measures of currency crises
that are also based on the EMP concept but typically use unscaled percentage changes
in reserves and exchange rates.
Classifying
currency regimes
295
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.4 Rationales for intervention
There has been considerable debate about the degree of post-crisis exchange rate
?exibility in Asia[8]. While it is widely assumed that there has been a substantial
increase in exchange rate ?exibility, there has also been considerable accumulation of
reserves. This has ledsome economists to argue that there has really beenlittle change in
exchange rate policies (McKinnon and Schnabl, 2004.) We can help clarify this debate by
distinguishing (conceptually at least) between intervention designed to accumulate
reserves such as may be highly desirable after a period of reserve losses, intervention to
hold down the average level of the exchange rate for competitive advantage, and
intervention to smooth out short-run ?uctuations in the exchange rate[9].
The ?rst two motives will be observationally equivalent in terms of the statistical
data during the early stages of recouping reserve losses. In the later stages of reserve
accumulation distinctions would have to be based on judgments about whether reserve
accumulations were becoming “excessive”. The appropriate level of reserves for a
country can of course be a matter of considerable dispute.
The third type of intervention – to limit short-term ?uctuations in the exchange
rate – is more easily identi?ed. Indeed that is what our suggested methodology is
designed to capture, once we detrend changes in reserves as well as changes in the
exchange rate.
Even when we distinguish trends from deviations from trends, we have no clear
theoretical basis for comparing their relative importance in terms of the overall degree
of ?exibility of a regime. How do we compare the degree of ?exibility of two crawling
band regimes, one of which has a faster rate of crawl while the other has a wider band?
Since there is no clear theoretical basis for the weight in adding these two dimensions
together we suggest in the next section that a two-parameter approach is needed.
The same type of problem applies to the use of measures of interest rate variability
(Hausmann et al., 2001; Baig, 2001; CR, 2002; HM, 2003; Cavoli and Rajan, 2005; Siregar
and Pontines, 2005, etc.) It is clearly appropriate to use changes in interest rates or other
monetary variables such as money supplies or degrees of sterilization in classifying
countries’ overall monetary cum exchange rate regimes, but it is not clear how the
behaviors of these monetary policy variables should be related to the classi?cation of the
degree of ?exibility of the exchange rate regime. CR are able to make a strong link only
by making the assumption that interest rate changes are only used to limit exchange rate
movements. They argue correctly that “such interest rate volatility is not the result of
adhering to strict monetary targets in the face of large and frequent money
demand shocks [. . .] ” (p. 392). However, they then jump directly from this statement to
the conclusion that “Interest rate volatility would appear to be the byproduct of a
combination of trying to stabilize the exchange rate through domestic open market
operation and lack of credibility” (p. 392). This leaves out the possibility of the effects of
other types of shocks and of domestically motivated monetary policy actions dictated by
discretion rather than a monetary rule. Interest rate changes can also be used to protect
reserve levels. Thus, we see no clear basis for a presumption that higher interest rate
variability should be considered as an indicator of less ?exibility in the exchange rate
regime. This is certainly an issue worthy of further investigation.
Post-crisis exchange rate regimes in Asian countries have been important issues with
regard to this subject recently but there has still been considerable disagreement. While
the view that most of the crisis countries shifted frompre-crisis ?xed rates to post-crisis
JFEP
3,4
296
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
?oating rates has become popular, many scholars point out its oversimpli?cation.
Siregar and Pontines (2005) ?nd that Indonesia, Korea, and Thailand have maintained
a de-facto ?exible exchange rate regime during the post-1997 period but Singapore
has increased intervention. They use an intervention index which is composed of
probabilities that exchange rate, international reserves, and interest rate are in regimes
with high volatilities based on Markov-Regime Switching ARCH model. However,
CR (2002) argue that countries announcing of?cially free ?oating tend to still intervene
in the foreign exchange market, so-called fear of ?oating. McKinnon and Schnabl (2004)
even argue that post-crisis Asian foreign exchange policies shows strong intervention
and they indeed have returned to soft dollar peg. Dooley et al. (2003) describe recent
phenomenon including the Asian regimes as a new de facto Bretton Woods regime.
Finally, HM (2003) argue that post-crisis foreign exchange policies in East Asia have
become more ?exible than before but less than real free ?oating.
3.5 Sterilization and the effects of exchange rates on domestic monetary policy
Exchange rate policy is only one aspect of monetary regimes (Rose, 2007). An exchange
rate regime requires a corresponding domestic monetary policy to avoid generating
crises. In the short-run countries may use sterilized foreign exchange interventions to
break the link between exchange rate and monetary policy, for example to deal with
capital movements that are believed to be temporary. Unsterilized intervention on the
other hand will generate corresponding changes in domestic monetary aggregates.
Thus, it is important to distinguish between sterilized and unsterilized intervention.
As is emphasized in Mundell-Fleming models and trilemma analysis under ?xed
exchange rates and perfect capital mobility sterilization is not possible. The trilemma
analysis argues that one cannot have all three of ?xed exchange rates, independent
monetary policy, and absence of controls as these are the three main ways to correct
balance of payments disequilibria. With imperfect capital mobility, however, this set of
constraints can be violated in the short-run through sterilized intervention. Many
commentators have wrongly assumed that because China has been following a fairly
tight peg against the dollar it must be importing the USA’ monetary policy. This
misses that China has substantial capital controls and has been able to sterilize most of
its balance of payments surplus over a number of years (Ouyang et al., 2010).
There has been considerable controversy over how much attention countries
should pay to exchange rate and balance of payments developments in setting domestic
monetary policy. This is an issue even for open economies that have adopted in?ation
targeting. In this context Willett (2003a, b) has suggested that instead of just being used
to decide if a country should join a currency area or adopt some other form of hard ?x
such as a currency board, the optimal currency area criteria can be used to investigate
how much weight a country should give to foreign exchange market developments in
setting domestic monetary policy. This can run from 100 percent for a hard ?x to
0 percent with a free ?oat for a large country.
Many countries that have ?oating rates have practiced heavy intervention in the
foreign exchange market. Korea is a good example. Of?cials frequently describe
Korea’s exchange rate regime as free ?oating, but it has at times intervened heavily in
the foreign exchange market (Kim and Willett, 2010). Korea has adopted in?ation
targeting but the exchange rate seems to still play an important role in its monetary
policy. Both Eichengreen (2004) and Parsley and Popper (2009) have found
Classifying
currency regimes
297
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
that Korea’s overnight call rate has responded to movement in the dollar-won exchange
rate as well as to expected in?ation and the output gap. Ouyang and Willett (2010) ?nd
that a high proportion of Korea’s intervention in the foreign exchange market has been
sterilized.
Estimating the degree of sterilization can be tricky. Simple correlations do
not distinguish between whether changes in monetary policy are causing changes in
reserves or changes in reserves are causing changes in monetary policy. Thus, it is
necessary to estimate policy reaction functions to attempt to make this distinction. Many
of the issues involved in such efforts are discussed in the longer version of this paper
because it is necessary to attempt to distinguish.
There is also an issue of whether interest rate variability should be included in
measures of exchange rate ?exibility. It is clearly appropriate to use changes in interest
rates or other monetary variables such as money supplies or degrees of sterilization in
classifying countries’ overall monetary cum exchange rate regimes, but it is not clear
how the behaviors of these monetary policy variables should be related to the
classi?cation of the degree of ?exibility of the exchange rate regime. CRare able to make
a strong link only by making the assumption that interest rate changes are only used to
limit exchange rate movements. They argue correctly that “suchinterest rate volatility is
not the result of adhering to strict monetary targets in the face of large and frequent
money demand shocks [. . .] ” (p. 392). However, they then jump directly from
this statement to the conclusion that “Interest rate volatility would appear to be the
byproduct of a combination of trying to stabilize the exchange rate through domestic
open market operation and lack of credibility” (p. 392). This leaves out the possibility of
the effects of other types of shocks and of domestically motivated monetary policy
actions dictated by discretion rather than a monetary rule. Interest rate changes can also
be used to protect reserve levels. Thus, we see no clear basis for a presumption that
higher interest rate variability should be considered as an indicator of less ?exibility in
the exchange rate regime. This is certainly an issue worthy of further investigation.
4. Concluding remarks
In the limited space of this article there are many issues with which we have not been
able to deal. We hope, however, that we have given the reader a useful introduction to
the major large datasets available that classify exchange rate regimes and an overview
of some of the major conceptual issues involved in classifying exchange rate regimes
and their relation to domestic monetary policies.
In the longer version of this paper available on the web we provide examples of how
our favored EMP approach can be applied and how it compares with the other major
approaches using the cases of India, Korea, and Tanzania. We also discuss a number of
the research areas for which classi?cations of currency regimes are used. In particular
we review studies of the effects of currency regimes on international trade, economic
growth, in?ation and other aspects of macroeconomic discipline, and on currency and
?nancial crises and the debate about best to characterize post-crisis exchange rate
regimes in Asia.
In closing we would like to reiterate what we view as some of the most important
methodological conclusions that we have reached:
.
The degree of exchange rate ?exibility is not the only important international
dimension of a country’s currency policies. Also important are the degree
JFEP
3,4
298
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
to which international exchanges are controlled and the role given to
developments in the foreign exchange market in setting domestic monetary
policy. For this latter purpose it is important to distinguish between sterilized
and unsterilized interventions in the foreign exchange market. Since it is dif?cult
to know to what extent a country changes its interest to protect its exchange rate
or its reserves it is probably not a good idea to use changes in interest rates in
measures of exchange rate ?exibility.
.
Of?cial statements about exchange rate policies are often misleading guides to
the practices that countries actually follow.
.
In general the degree of ?exibility of an exchange rate regime cannot be measured
correctly by the behavior of the exchange rate alone. Lowlevels of ?uctuation can
be caused either by heavy intervention or by there being only small shocks.
To distinguish between these cases it is useful to use the concept of EMP and look
at the ratio of the pressures that lead to changes in the exchange rate versus
changes in reserves. This gives us an index of ?exibility or its inverse, the
propensity to intervene. This framework assumes that countries intervene to lean
against the wind, i.e. to slowdown exchange rate movements. Where governments
intervene to accelerate exchange rate movements the degree of exchange rate
?exibility is not well de?ned. Measures that use the ratios of variances do not
make such distinctions
.
In using this EMP approach it will often be important to distinguish between
relationships between trends in reserves and exchange rates and ?uctuations
around these trends.
.
Each country’s currency has exchange rates against many other currencies.
It will often be misleading to consider its exchange rate against only one other
currency such as the dollar or the euro.
.
One should be very suspicious of classi?cations that are binary, i.e. that only
distinguish between ?xed and ?exible exchange rates since there are many
versions of both ?xed and ?exible rates. For example, when investigating
relationships with macroeconomic discipline and currency and ?nancial crises it
is essential to distinguish between hard and soft ?xes. Likewise under a two-way
classi?cation it is not at all clear whether crawling band regimes should be
placed under ?xed or ?exible rates. By varying such placements a researcher
could often get almost any results that they wanted a priori.
.
There is no one best number of categories to use. That will dependonthe purpose of
the study. The same goes for the best time periods over which to measure regimes.
.
The most complex statistical measures are not always the most accurate. It is
sometimes not clear what they are actually measuring.
.
Since there is no perfect way to implement the concept of exchange rate
?exibility or its inverse, how heavily a government is managing its exchange
rate it is important for researchers to test the robustness of their results to
alternative methods.
This may seem a daunting list of considerations to take into account. However, the fact
that it is unlikely that we can create perfect measures should not de?ect us from trying
Classifying
currency regimes
299
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
to construct less imperfect measures, nor from using measures that we know are
imperfect but we think are the best available. When we do this, however, we should be
humble about our results and test them against alternative measures. Science often
progresses in little jumps.
Notes
1. Numbers in parenthesis refer to their ?ne classi?cation.
2. These issues are discussed in the longer version of this paper.
3. In the monetary model used by Girton and Roper this slope is one.
4. RR also use the similar approach. For example, RR classify a case as a de facto peg if the
probability that the monthly exchange rate change remains within 1 percent band over
rolling ?ve-year period is 80 percent or higher.
5. CR give a nice discussion of several of these issues.
6. Willett et al. (2007) adjust reserve changes for Japan for estimates of interest earnings, but
just ?nd that this makes little difference for their estimates. Both series fairly closely track
the data on Japan’s actual interventions over the period 1991-2005. Of course, this proxy may
not work as well for other countries, but our results for Japan suggest that this approach is
worth using in the absence of better information.
7. This problem can of course also apply to continued appreciation or depreciation.
8. This debate is discussed in the longer version of this paper.
9. Of course, there could also be the intervention to prop up the rate to avoid in?ation. This is
especially likely before elections.
References
Angkinand, A. and Willett, T.D. (2011), “Exchange rate regimes and banking crises: the channels
of in?uence investigated”, International Journal of Finance & Economics (in press).
Angkinand, A., Chiu, E.M.P. and Willett, T.D. (2009), “Testing the unstable middle and two
corners hypotheses about exchange rate regimes”, Open Economies Review, Vol. 20 No. 1,
pp. 61-83.
Bai, J. and Perron, P. (1998), “Estimating and testing linear models with multiple structural
changes”, Econometrica, Vol. 66 No. 1, pp. 47-78.
Bai, J. and Perron, P. (2003), “Computation and analysis of multiple structural change models”,
Journal of Applied Econometrics, Vol. 18 No. 1, pp. 1-22.
Baig, T. (2001), “Characterizing exchange rate regimes in post-crisis east asia”,
IMF Working Paper No. 125, International Monetary Fund, Washington, DC.
Bubula, A. and O
¨
tker-Robe, I. (2002), “The evolution of exchange rate regimes since 1990:
evidence from de facto policies”, IMF Working Paper No. 155, International Monetary
Fund, Washington, DC.
Calvo, G. and Reinhart, C. (2002), “Fear of ?oating”, Quarterly Journal of Economics, Vol. 117
No. 2, pp. 379-408.
Cavoli, T. and Rajan, R. (2005), “Have exchange rate regimes in Asia become more ?exible post
crisis? Re-visiting the evidence”, Centre for International Economic Studies Discussion
Paper No. 0503, University of Adelaide, Adelaide.
JFEP
3,4
300
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Chiu, E.M.P., Dechsakulthorn, S., Walter, S., Walton, J. and Willett, T.D. (2011), “The discipline
effects of ?xed exchange rates: the distinction between hard and soft pegs”, working paper,
Claremont Institute for Economic Policy Studies, Claremont, CA.
Dooley, M.P., Folkerts-Landau, D. and Garber, P.M. (2003), “An essay on the revived Bretton
Woods system”, NBER Working Paper No. 9971, National Bureau of Economic Research,
Cambridge, MA.
Eichengreen, B. (2004), “Monetary and exchange rate policy in Korea: assessments and policy
issues”, Discussion Paper No. 4676, Center for Economic and Policy Research,
Washington, DC.
Frankel, J. and Wei, S.J. (1994), “Yen bloc or dollar bloc? Exchange rates in the East Asian
economies”, in Ito, T. and Krueger, A. (Eds), Macroeconomic Linkage: Savings, Exchange
Rates, and Capital Flows, Univerity of Chicago Press, Chicago, IL.
Frankel, J. and Wei, S.J. (2008), “Estimation of de facto exchange rate regimes: synthesis of the
techniques for inferring ?exibility and basket weights”, NBER Working Paper No. 14016,
National Bureau of Economic Research, Cambridge, MA.
Ghosh, R. (2011), “Capital controls and exchange rate regime in India”, PhD dissertation,
UMI No. 3449833, Claremont Graduate University, Claremont, CA.
Ghosh, A.R., Gulde, A. and Wolf, H. (2003), Exchange Rate Regimes: Choices and Consequences,
MIT Press, Cambridge, MA.
Girton, L. and Roper, D. (1977), “A monetary model of exchange market pressure applied to the
postwar Canadian experience”, American Economic Review, Vol. 67, pp. 537-48.
Habermeier, K., Kokenyne, A., Veyrune, R. and Anderson, H. (2009), “Revised system for the
classi?cation of exchange rate arrangements”, IMF Working Paper No. 211, International
Monetary Fund, Washington, DC.
Hausmann, R., Panizza, U. and Stein, E. (2001), “Why do countries ?oat the way they ?oat?”,
Journal of Development Economics, Vol. 66, pp. 387-414.
Herna´ndez, L. and Montiel, P.J. (2003), “Post-crisis exchange rate policy in ?ve Asian countries:
?lling in the hollow middle?”, Journal of the Japanese and International Economics, Vol. 17
No. 3, pp. 336-69.
Kim, K.S., Willett, T.D. and Willett, T.D. (2010), “Korea’s exchange rate policies between the
crises”, The Global Crisis and Korea’s International Financial Policies, Korea Economic
Institute, Washington, DC, pp. 11-38.
Klein, M.W. and Shambaugh, J. (2010), Exchange Rate Regimes in the Modern Era, MIT Press,
Cambridge, MA.
Levy Yeyati, E. and Sturzenegger, F. (2005), “Classifying exchange rate regimes: deeds vs words”,
European Economic Review, Vol. 49, pp. 1603-35.
McKinnon, R. and Schnabl, G. (2004), “The East Asian dollar standard, fear of ?oating,
and original sin”, Review of Development Economics, Vol. 8 No. 3, pp. 331-60.
Ouyang, A.Y. and Willett, T.D. (2010), “International aspects of Korea’s monetary policies”,
in Willett, T.D., Auerbach, N.N., Kim, K.S., Kim, Y., Ouyang, A., Permpoon, O.,
Sompornserm, T., Srisorn, L., Sula, O. et al. (Eds), The Global Crisis and Korea’s
International Financial Policies, Korea Economic Institute, Washington, DC, pp. 76-88.
Ouyang, A.Y., Rajan, R.S. and Willett, T.D. (2008), “Managing the monetary consequences of
reserve accumulation in emerging Asia”, Global Economic Review, Vol. 37 No. 2, pp. 171-99.
Ouyang, A.Y., Rajan, R.S. and Willett, T.D. (2010), “China as a reserve sink”, Journal of
International Money and Finance, Vol. 29 No. 5, pp. 951-72.
Classifying
currency regimes
301
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Parsley, D. and Popper, H. (2009), “Evaluating exchange rate management: an application to
Korea”, Working Paper No. 282009, Hong Kong Institute for Monetary Research, Central.
Reinhart, C. and Rogoff, K. (2004), “The modern history of exchange rate arrangements:
a reinterpretation”, Quarterly Journal of Economics, Vol. 119, pp. 1-48.
Rose, A.K. (2007), “A stable international monetary system emerges: in?ation targeting is
Bretton Woods, reversed”, Journal of International Money and Finance, Vol. 26, pp. 663-81.
Siregar, R.Y. and Pontines, V. (2005), “External debt and exchange rate overshooting: the case of
selected East Asian countries”, School of Economics Working Paper No. 14, University of
Adelaide, Adelaide.
Weymark, D. (1995), “Estimating exchange market pressure and the degree of exchange market
pressure for Canada”, Journal of International Economics, Vol. 39, pp. 273-95.
Weymark, D. (1997), “Measuring the degree of exchange rate market intervention in a small open
economy”, Journal of International Money and Finance, Vol. 16, pp. 55-79.
Weymark, D. (1998), “A general approach to measuring exchange market pressure”,
Oxford Economic Papers, Vol. 50, pp. 106-21.
Willett, T.D. (2003a), “Fear of ?oating needn’t imply ?xed rates: feasible options for intermediate
exchange rate regimes”, Open Economies Review, Vol. 14 No. 1, pp. 71-91.
Willett, T.D. (2003b), “The OCA approach to exchange rate regimes”, in Dean, J.W., Salvatore, D.
and Willett, T.D. (Eds), The Dollarization Debate, Oxford University Press, New York, NY,
pp. 154-71.
Willett, T.D. and Kim, Y. (2006), “Korea’s post-crisis foreign exchange rate policy”, Korea’s
Economy, Korea Economic Institute, Washington, DC.
Willett, T.D., Kim, Y. and Nitithanprapas, I. (2007), “Taking the concept of exchange market
pressure seriously for classifying exchange rate regimes”, working paper, Claremont
Institute for Economic Policy Studies, Claremont, CA.
Willett, T.D., Nancy, N.A., Kenneth, S.K., Yoonmin, K., Alice, O., Orawan, P., Thana, S., Lalana, S.
and Ozan, S. (2010), The Global Crisis and Korea’s International Financial Policy,
Korea Economic Institute, Washington, DC.
Further reading
Chiu, E.M.P. and Willett, T.D. (2009), “The interactions of government strength and alternative
exchange rate regimes in avoiding currency crises”, International Studies Quarterly, Vol. 53
No. 4, pp. 1001-25.
Frankel, J. and Xie, D. (2010), “Estimation of de facto ?exibility parameter and basket weights in
evolving exchange rate regimes”, NBER Working Paper No. 15620, National Bureau of
Economic Research, Cambridge, MA.
About the authors
Thomas Willett (PhD, University of Virginia) is the Horton Professor of Economics at Claremont
Graduate University and Claremont McKenna College and Director of the Claremont Institute of
Economic Policy Studies and of its China-Asia Paci?c Political Economy Program. Before
coming to Claremont he was on the faculties of Cornell and Harvard Universities and served in
Washington, DC as a senior economist on the President’s Council of Economic Advisers and as
Director of international research at the US Treasury. He has also been a visiting scholar at the
International Monetary Fund and served as Editor or member of the editorial board of
14 professional journals in the areas of economics, public policy, and international relations.
He has written widely in the areas of monetary and ?nancial economics and the political
JFEP
3,4
302
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
economy of public policy with a particular focus on issues of international money and ?nance.
He has been a leader in the movement to better integrate economic and political analysis,
especially in the area of international political economy. His recent research has focused
particularly on international capital ?ows, currency policies, and currency and ?nancial crises
and issues of domestic and international ?nancial reform where he combines analysis from
political economy, behavioural ?nance, and complexity economics. Thomas Willett is the
corresponding author and can be contacted at: [email protected]
Eric M.P. Chiu is an Associate Professor at the Graduate Institute of National Policy and
Public Affairs at the National Chung Hsing University in Taiwan, a position he has held since
September 2006. He received his PhD from the Claremont Graduate University in International
Political Economy. His research concentrates on how politics interact with exchange rate regimes
and capital controls in shaping macroeconomic stability such as currency and ?nancial crises as
well as international capital movement, with speci?c focus on both emerging and developing
countries. His work has been published in International Studies Quarterly, Open Economies
Review, and Journal of Economic Policy Reform.
Sirathorn (B.J.) Dechsakulthorn is an Economist based in Los Angeles, California, USA.
Ramya Ghosh is a Faculty member at Drexel University’s LeBow College of Business. His
research interests are in the area of international ?nance focusing on international capital ?ows,
capital controls and exchange rate policies. Prior to joining Drexel University, Ghosh taught
Economics and Finance at University of California – Irvine and California State Polytechnic
University – Pomona. He was also a consultant for the World Bank and the evaluation of?ce of
the International Monetary Fund (IMF) in Washington, DC.
Bernard Kibesse is a PhD candidate in Economics at Claremont Graduate University
specializing in international money and ?nance. His current research interest is centred on
exchange rate regime, monetary policy and currency union of East African countries. Kibesse
also holds a Masters degree in ?nance from the University of Strathclyde (UK). He is also a
Certi?ed Public Accountant (CPA) and has been with the Central Bank of Tanzania for the
last 15 years.
Kenneth Kim is currently a PhD candidate at the Claremont Graduate University specializing
in the ?eld of International Money and Finance focusing on the Asian Crisis, Shared Mental
Models, and Korean exchange rate regime and policy. He has been teaching undergraduate
Intermediate Macroeconomics and International Economics courses at the California State
University, Fullerton and has also taught Public Finance (MS in Taxation Program) at the
California State University at Fullerton, Irvine Campus.
Jeff ( Yongbok) Kim has worked for the Bank of Korea since 1997 and joined in the
International Policy Analysis Division of the European Central Bank as a seconded economist in
January 2011. He specializes in International Money and Finance. Speci?c topics of interest
include exchange rate regimes, international capital ?ows, and Asian economic regionalism.
Alice Ouyang (PhDin Economics fromClaremont Graduate University) is Associate Professor
of China Academy of Public Finance and Public Policy at Central University of Finance and
Economics. Her research interests mainly focus on International Money and Finance, Chinese and
Asian Economies, Macroeconomics, and Quantitative Method. Her papers have been published in
Journal of International Money and Finance, Applied Economics, and Global Economic Review.
Classifying
currency regimes
303
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This article has been cited by:
1. Stephen G. Hall, Amangeldi Kenjegaliev, P.A.V.B. Swamy, George S. Tavlas. 2013. Measuring currency
pressures: The cases of the Japanese yen, the Chinese yuan, and the UK pound. Journal of the Japanese
and International Economies 29, 1-20. [CrossRef]
2. THOMAS D. WILLETT, JEFF (YONGBOK) KIM, ISRIYA NITITHANPRAPAS BUNYASIRI.
2012. MEASURING EXCHANGE RATE FLEXIBLITY: A TWO-PARAMETER EXCHANGE
MARKET PRESSURE APPROACH. Global Journal of Economics 01, 1250007. [CrossRef]
3. William R. Clark, Mark Hallerberg, Manfred Keil, Thomas D. Willett. 2012. Measures of financial
openness and interdependence. Journal of Financial Economic Policy 4:1, 58-75. [Abstract] [Full Text]
[PDF]
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_797384061.pdf
There has been significant interest in the classification of exchange rate regimes in order
to investigate a wide range of hypotheses. Studies of the effects of exchange rate regimes on crises and
other aspects of economic performance can have important implications for policy choices. The paper
provides a guide to the major new large data sets that classify exchange rate regimes and to critically
analyze important methodological issues.
Journal of Financial Economic Policy
Classifying international aspects of currency regimes
Thomas Willett Eric M.P. Chiu Sirathorn (B.J .) Dechsakulthorn Ramya Ghosh Bernard Kibesse Kenneth
Kim J eff (Yongbok) Kim Alice Ouyang
Article information:
To cite this document:
Thomas Willett Eric M.P. Chiu Sirathorn (B.J .) Dechsakulthorn Ramya Ghosh Bernard Kibesse Kenneth
Kim J eff (Yongbok) Kim Alice Ouyang, (2011),"Classifying international aspects of currency regimes",
J ournal of Financial Economic Policy, Vol. 3 Iss 4 pp. 288 - 303
Permanent link to this document:http://dx.doi.org/10.1108/17576381111182882
Downloaded on: 24 January 2016, At: 21:43 (PT)
References: this document contains references to 40 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 423 times since 2011*
Users who downloaded this article also downloaded:
Puspa Amri, Apanard P. Angkinand, Clas Wihlborg, (2011),"International comparisons of bank regulation,
liberalization, and banking crises", J ournal of Financial Economic Policy, Vol. 3 Iss 4 pp. 322-339 http://
dx.doi.org/10.1108/17576381111182909
Donatella Porrini, Giovanni B. Ramello, (2011),"Class action and financial markets: insights
from law and economics", J ournal of Financial Economic Policy, Vol. 3 Iss 2 pp. 140-160 http://
dx.doi.org/10.1108/17576381111133615
Komain J iranyakul, (2012),"Linkages between Thai stock and foreign exchange markets under
the floating regime", J ournal of Financial Economic Policy, Vol. 4 Iss 4 pp. 305-319 http://
dx.doi.org/10.1108/17576381211279280
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Classifying international aspects
of currency regimes
Thomas Willett
Claremont Graduate University and Claremont McKenna College,
Claremont, California, USA
Eric M.P. Chiu
National Chung-Hsing University, Taichung, Taiwan
Sirathorn (B.J.) Dechsakulthorn
DLA Piper, Los Angeles, California, USA
Ramya Ghosh
LeBow College of Business, Department of Economics, Drexel University,
Sacramento, California, USA
Bernard Kibesse and Kenneth Kim
Claremont Graduate University, Claremont, California, USA
Jeff (Yongbok) Kim
The Bank of Korea, Seoul, Republic of Korea, and
Alice Ouyang
Central University of Finance and Economics, Beijing, China
Abstract
Purpose – There has been signi?cant interest in the classi?cation of exchange rate regimes in order
to investigate a wide range of hypotheses. Studies of the effects of exchange rate regimes on crises and
other aspects of economic performance can have important implications for policy choices. The paper
provides a guide to the major new large data sets that classify exchange rate regimes and to critically
analyze important methodological issues.
Design/methodology/approach – The study surveys and critiques the literature and provides
theoretical analysis of major issues involved in classifying exchange rate regimes.
Findings – The study ?nds that all of the new data sets have problems but some have more
problems than others and several of them are substantial improvements on what was previously
available. It is also shown that the best ways to classify depend on the issue being addressed and that
for detailed studies variants of measures using the concept of exchange market pressure are the most
promising. Directions for future research are also discussed.
Originality/value – The paper makes researchers aware of the new data sets that are available and
discusses their strengths and weaknesses. It also presents original analysis of several of the major
conceptual issues involved in classifying exchange rate regimes.
Keywords Foreign exchange, International ?nance, International economics,
Open economy macroeconomics, International trade
Paper type Research paper
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JEL classi?cation – F31, F3, F41, F4, F
JFEP
3,4
288
Journal of Financial Economic Policy
Vol. 3 No. 4, 2011
pp. 288-303
qEmerald Group Publishing Limited
1757-6385
DOI 10.1108/17576381111182882
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
1. Introduction
There has beenconsiderable recent interest in the classi?cation of exchange rate regimes
in order to investigate a wide range of hypotheses including the relationships of
alternative exchange rate regimes with in?ation, growth, and currency crisis, etc.
It might seem that exchange rate regimes should be an easy type of policy to classify.
After all exchange rates are highly visible and data on them is easily available for most
countries. On closer investigation, however, it turns out that while classifying some
types of regimes such as of?cial crawling pegs is easy in other cases such as attempting
to distinguish between managed ?oaters and unof?cial crawling bands can be quite
hard. As another example there has been controversy about whether after the crisis of
1997-1998 most exchange rate regimes in Asia are better described as managed ?oating
or a continuation of a soft dollar standard (McKinnon and Schnabl, 2004;
Willett et al., 2010.)
A major dif?culty is that for many purposes we would like to classify regimes
by how heavily they are managed by governments. Typically we cannot tell this by
looking at the behavior of the exchange rate alone. Low volatility could be due to heavy
management or to an absence of large shocks. In Section III we discuss how the concept
of exchange market pressure (EMP) can be applied to address this issue.
Early empirical studies used the old IMF classi?cations based on countries’ of?cial
reports. It has become widely recognized, however, that such of?cial reports are often
misleading. Consequently, an important and active research agenda has been to produce
behavioral measures characterizing exchange rate policies based on what governments
actually do rather than what they say they do.
Unfortunately such efforts have often reached con?icting classi?cations for particular
countries’ exchange rate regimes. For example, the newclassi?cations by the IMFand by
Reinhart andRogoff (RR, 2004) wronglyclassifythe Koreanexchange rate regime in2000
as being in their most ?exible categories. While there was considerable movement of the
exchange rate there was also substantial intervention (Willett and Kim, 2006.) Thus,
Korea actually had a managed ?oat. At the other extreme the study by Levy Yeyati and
Sturzenegger (LYS, 2005) classi?ed the regime as a ?xed rate.
Despite such dif?culties these recent efforts at large Nclassi?cations of exchange rate
regimes make substantial improvements over the old of?cial classi?cations. As we will
discuss, however, there is no one ideal method of classifying exchange rate regimes.
While we will indicate some pitfalls to avoid in developing classi?cations, it is important
to recognize that there is no one optimal measure. The best classi?cations may vary
depending on the type of question for which they are being used.
It is also important to recognize that the international aspects of currency policies
are not limited to just the degree of ?exibility of the exchange rate. One important
dimension is the extent to which developments in the foreign exchange market,
i.e. changes in exchange rates and the balance of payments, affect domestic monetary
policy. In this context it can make a huge difference whether or not intervention is
sterilized. Whether currencies are defended by capital controls can also be important.
(The measurement of capital controls is covered in the paper on ?nancial openness.)
In the following section we discuss strengths and weaknesses of the major recent
efforts at classifying exchange rate regimes that cover a large number of countries.
In Section III we discuss a number of the conceptual issues involved in studying
particular countries’ policies in more detail. In the longer version of this paper available
Classifying
currency regimes
289
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
on our web site we provide illustrations of such analysis for India, Korea, and Tanzania
and discussion of the many of the types of questions which classi?cations of currency
policies are used to study.
2. Reviews of major efforts at exchange rate classi?cations
2.1 The new IMF de facto measures
The IMF traditionally provided data on exchange rate regimes based on their of?cial
reports. Studies ?nd that in practice exchange rate regimes are often substantially
different from what governments said they were doing. Therefore, the IMF has
published classi?cations since 1999 based on the staff’s judgment of what kinds of
regimes are actually being followed. The original de facto classi?cations were back
dated to 1990 and are discussed in the IMF working paper by Bubula and O
¨
tker-Robe
(2002). The revised classi?cation is available since 2009 and was published in the 2009
Annual Report on Exchange Arrangements and Exchange Restrictions.
The IMF approach focused on whether a country targeted a path for the exchange
rate? This provided a substantial improvement on the old IMF classi?cation and in our
judgment became the best available large N set of classi?cations. Its main dif?culty
was in its classi?cations of the degree of management of ?exible rate regimes. These
were divided into three types. The most ?exible category was labeled “independent”
?oating and described as follows: The exchange rate is market determined; any foreign
exchange intervention aims at moderating the rate of change and preventing undue
?uctuations that are not justi?ed by the fundamentals, rather than establishing a level
for the exchange rate” (Bubula and O
¨
tker-Robe, 2002, p. 15.)This was contrasted with
“Tightly or Other Managed Floating” where:
The authorities in?uence exchange rate movements through interventions to counter the
long-term trend of the exchange rate, without specifying a predetermined exchange rate path,
or without having a speci?c exchange rate target (“dirty ?oating”) (Bubula and O
¨
tker-Robe,
2002, p. 15.)
Their analytical basis for distinguishing between “tightly” and “other” managed ?oating
was not clear. These three categories need not correspond to the degree of heaviness of
management. Even the distinction between independent and managed ?oating does not
seemclear, however, since “moderatingthe rate of change” and“countering the long-term
trend” canbothbe forms of “leaningagainst the wind” intervention. Inother words, under
both the IMF’s categories of managed and independent ?oating there could be heavy
or light exchange rate management. Since both Japan and Korea were classi?ed as
independently?oaters duringyears inwhich theyintervenedheavily, it is not clear that it
was useful to distinguish between these categories.
In part because of such concerns in 2009 the IMF revised its de facto classi?cations,
using only two categories of ?oating based on the degree of intervention. They also
distinguish more clearly between formal and informal pegs and crawls
(Habermeier et al., 2009.)
2.2 Studies that focus only on the behavior of the exchange rate
2.2.1 RR classi?cations. RR’s heroic effort of classi?cation presents the best currently
available large source for long-time period. Astrong advantage of these classi?cations is
that they use market rather than of?cial rates. Apart fromnot taking the degree of EMP
JFEP
3,4
290
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
into account, RR’s approach has dif?culty with short-lived de facto pegs. They classify
episodes as short-lived de facto pegs if the absolute monthly percentage change in the
exchange rate is equal to zero for four consecutive months or more. This decision rule is
too strict to detect actual short-lived de facto pegs. It is practically impossible for the
monthly percentage changes to be zero consecutively except for hard pegs such as
currency unions. In many studies, pegs are therefore assumed to have a certain level of
narrowbands. In order to discover long-lived de facto pegs, they use the probability that
monthly percentage changes in exchange rates is less than 1 percent over a rolling
?ve-year period. If it is larger than 80 percent, they classify the episodes as pegs.
This scheme cannot detect pegs with frequent breaks in exchange rates. Such two
methodologies in RR therefore have weakness in detecting short-lived de facto pegs.
RR develop a new category they call freely falling regimes. These are ?oating
regimes for countries with in?ation rates above 40 percent. This category is open to the
criticism that it biases results toward ?nding favorable macroeconomic performance
for ?oating regimes by throwing out cases of bad performance. On the other hand it
seems unlikely that the exchange rate regime would be a major cause of very high
in?ation rates (Chiu et al., 2011.)
RR use a special probability measure to distinguish the freely ?oating from the
managed ?oat. They construct an exchange rate ?exibility index:
1
Pð1 , 1%Þ
;
based on the probability that percentage changes of exchange rates over rolling
?ve-year period remain within 1 percent. RR compare the frequency distribution of the
index of each country with what they deem to be free ?oaters such as the US dollar
against the euro over the post-1973 period.
Their decision rule is very conservative for classifying managed ?oats. This
combined with the use only of the behavior of exchange rates results in misclassifying
some cases of managed ?oats as free ?oats. The cases of Korea and Japan who allowed
substantial exchange rate movements while also engaging in heavy intervention in a
number of years are important examples. Since free ?oats are commonly de?ned as
?oats with little or no of?cial intervention this problem could have been avoided by
labeling such cases as ?oats and deleting the “free”.
Another problem concerns their coarse classi?cation. They group
their ?ne classi?cations into ?ve coarse regimes to investigate their associations
with macroeconomic variables such as in?ation and growth. As suggested by
Angkinand et al. (2009), a major problemwith their grouping is that it fails to distinguish
between hard and soft pegs. They group (2) pre-announced peg, (3) pre-announced
horizontal band that is narrower than or equal to 2 percent, and (4) de facto peg with (1)
no separate legal tender all in the same category[1]. While this may be reasonable for
some purposes we have found that distinguishing between hard and soft pegs is
essential for analyzing the effects of exchange rate regimes on currency and ?nancial
crises and on macroeconomic discipline (Angkinand et al., 2009; Angkinand and Willett,
2011; Chiu et al., 2011).
2.2.2 Two-way classi?cations: Klein and Shambaugh (2010)
This major book focuses primarily on comparing the effects of ?xed versus ?exible
exchange rates. As we noted above such two-way distinctions may be useful
Classifying
currency regimes
291
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
for some purposes, but for issues such as discipline and currency and banking crises
such a limited breakdown is seriously ?awed. The same type of criticism applies to
the early studies of capital controls that used only zero-one dummies. (See the
accompanying paper on ?nancial openness and interdependence). Like the RR study
their method only employs the behavior of the exchange rate, and thus does not control
for shocks. The criteria for distinguishing between ?xed and ?exible rates are based on
whether the exchange rate stays inside a band. Speci?cally, ^2.0 percent bands
against base currencies are used for annual classi?cations. If the exchange rate change
does not exceed the threshold in 11 out of 12 months, it is classi?ed as the ?xed.
Emphasis on the choice of base currencies is a valuable contribution since the
automatic use of the dollar can be misleading for countries more oriented toward other
currencies such as the euro. Another question involves the extent to which governments
pay attention to different foreign currencies. While it is common to refer to a country’s
exchange rate against the dollar (or the euro) there are many exchange rates, one for each
other independent currency. Focusing on the exchange rate against just one currency
can give a misleading picture when there are substantial changes in cross rates among
other currencies. Anumber of countries peg more or less closely to a basket of currencies.
While the list of these currencies is sometimes made public, the relative weights given to
the different currencies is seldom made public. This has led to efforts to estimate the
implicit weights in such baskets (Frankel and Wei, 1994.) Where countries follow a
managed ?oat such estimation becomes more dif?cult but the issue retains its
importance (Frankel and Wei, 2008; Frankel and Xie, 2010)[2].
2.2.3 Statistical cluster analysis: LYS
LYS classify exchange rate regimes across 183 countries for 1974-2004. They
classify episodes into ?ve categories – ?exible, dirty ?oat, crawling peg, ?xed, and
inconclusive – using three classifying variables, exchange rate volatility, volatility of
exchange rate changes, and volatility of reserves. They use the volatility of the rate of
change of the exchange rate to identify crawling pegs.
Exchange rate volatility is measured by the average of the absolute monthly
percentage changes in the nominal bilateral exchange rate during a calendar year. The
volatility of exchange rate changes is measured by the standard deviation of the
monthly percentage changes in the exchange rate. To compute the volatility of reserves
they ?rst subtract central government deposits from net foreign assets and divide this
by the exchange rate. Then they compute its ?rst-order differenced value and divide
this by the lagged money base. The volatility of reserves is measured by the average
absolute monthly changes in this measure.
They use K-means cluster (KMC) analysis with three classifying variables. The
KMC algorithm produces groups such that they have the smallest total distance
between episodes and the center of the group. They divide each variable into just two
categories, high and low volatility.
While their analysis does have the advantage of being objective, they provide no
clear economic rationale for their particular formulations and the thresholds used, nor
do they explain why they do not consider the case of a highly volatile exchange
rate combined with low volatility of exchange rate changes and reserves. Additional
limitations of the LYS classi?cation are its inability to distinguish hard from soft
?xes and wide-band crawling pegs from dirty ?oat regimes. LYS classify regimes
JFEP
3,4
292
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
by calendar years. For years in which the regimes or the parity changes or there are
transitional periods of instability, it is unclear how the LYS calculations should be
interpreted.
3. Some major conceptual issues
3.1 The EMP approach
The concept of EMP introduced by Girton and Roper (1977) is a measure of the gap
between quantities demanded and supplied in the foreign exchange market at a
particular exchange rate. It provides a way of comparing pressures under alternative
exchange rate regimes by adding changes in reserves (as a measure of of?cial
intervention) and exchange rate changes[3]. Conceptually, these should be weighted by
the slope of the excess demand function in the foreign exchange market but in the
absence of good measures of these parameters, equal weights are often used. The
measure of the degree of ?exibility of an exchange rate regime then becomes the ratio
of EMP taken by changes on reserves versus changes in the exchange rate. Depending
on which way one formulates the ratio this will give a measure of the propensity of the
authorities to intervene in the foreign exchange market or its reciprocal, the degree of
exchange rate ?exibility. This approach is superior to just looking at exchange rate
changes alone since these movements are in?uenced by the size of shocks as well as the
propensity to intervene.
The basic idea of the EMP underlies many of the recent efforts at classi?cation of
exchange rate regimes. However, these studies often do not fully conform to the
concept of the proportion of EMP taken on the exchange rate versus intervention.
The ?rst explicit use of the concept of EMP to classify exchange rate regimes was by
Weymark (1995, 1997, 1998). She de?nes the degree of intervention as the proportion of
EMP absorbed by exchange market intervention. Weymark’s index has a range
from21to þ1. When the sign of changes in exchange rate and reserves is correct for
leaning against wind intervention, then Weymark’s index has a range from 0 to 1, with
values closer to 1 indicating a higher degree of ?xity. When the exchange rate changes
are of the same sign, but have a greater absolute magnitude than the changes that would
have occurred in the case of no intervention, Weymark’s index is negative. When the
exchange rate appreciates (depreciates) with excess supply (demand) of domestic
currency, Weymark’s index is greater than one. As was discussed in Section II the
interpretation of the degree of intervention in cases of wrong sign observations in
Weymark’s index is not straightforward. (Further discussion of technical aspects of
Weymark’s papers is discussed in the longer web version of this paper).
In more recent papers Calvo and Reinhart (CR, 2002), and Herna´ndez and Montiel
(HM, 2003) also drawon the concept of EMP to offer procedures for classifying exchange
rate regimes. Their approaches, however, suffer from several technical problems.
CR and HM both look at the volatilities of the exchange rate, international reserves and
interest rates to characterize exchange rate policies. CR look at the ratio of variances
while HM compare the volatilities of each variable across countries separately rather
than comparing their ratios. However, relative variances can be misleading when trends
are present. This is less of a problem when exchange rate variability is measured in
terms of the variability of changes rather than levels.
Furthermore, ratios of variances do not distinguish cases of wrong signs and as
CR note, variances may be distorted by outliers. Thus, they advocate an approach
Classifying
currency regimes
293
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
based on the frequency with which classifying variables remain within a threshold
limit, arguing that: the probabilistic nature of the statistic conveys information about
the underlying frequency distribution that is not apparent from the variance (p. 384)[4].
The choice of thresholds is of course somewhat arbitrary. CR use 1 and 2.5 percent for
monthly percentage changes. With the use of variances, however, this approach does
not distinguish between correct- and wrong-signed pairs of observations.
3.2 The problem of trends and wrong signs
As noted previously the conceptual basis of the EMP approach assumes governments
intervene to slow or halt exchange rate movements. Cases where intervention is used to
accelerate exchange rate ?exibility are thus not well de?ned. It is not easy to know how
these wrong sign observations should be treated. They can be caused by imperfections
in the reserve change proxy as well as by episodes of leaning with the wind, i.e. pushing
the rate even further than the market has been taking it or of extreme forms of leaning
against the wind where the rate is forced in the opposite direction from market forces.
Actual leaning with the wind in a downward direction is classic beggar thy neighbor
policy and is discouraged by the IMF’s guidelines except for cases where a currency is
judged to be seriously over-valued. Leaning with the wind in the upward direction may
be justi?ed during periods in which country’s currencies are considered to have
over-depreciated. The aftermath of the Asian crisis is a prominent example of such
over-depreciation.
Where imperfect proxies are the cause of the wrong sign, the best solution would
likely be to drop these observations. With true leaning with the wind, one could argue
either that this is a case of government management or that it should be considered as
super ?exibility. Which interpretation is better may depend on the use to be made of
the classi?cation.
Wrong signs could also result from an extreme form of leaning against the wind,
i.e. instead of allowing domestic currency to depreciate in the face of excess supply in
the foreign exchange market, the government could actively appreciate the exchange
rate. Likewise, despite an excess demand in foreign exchange market, a mercantilist
government could force down the exchange rate. It would seem that these should be
treated as cases of strong government management or of super ?xity. Thus, depending
on the cause of the wrong sign, it can be argued that a value of zero or one should be
assigned or that the observation should be deleted.
Trends in exchange rates and reserve changes also present substantial problems for
applying the EMP approach. A major dif?culty is that there is no clear-cut theoretical
rationale for how to compare the comparative degrees of ?exibility of a faster crawl
with a smaller band of ?uctuations around it with a slower rate of crawl but larger
?uctuations around it. For example, the former which cause less problems for an
export-import ?rm whose transactions have a short-time horizon but cause greater
problems for a manufacturing exporter with a much longer time horizon. Given this
problem Willett et al. (2007) argue that a two parameter approach is necessary to fully
characterize exchange rate regimes, one for trend relationships and one for ?uctuations
around trend.
This approach runs into the substantial operational dif?culty involved with
detecting changes in trends. There is no clear-cut rule for deciding when changes are
large enough to justify treating them as shifts in regime, and the most appropriate
JFEP
3,4
294
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
degree of sensitivity can vary with the purpose of the investigation. For the purpose of
deciding whether a country is following a managed ?oat or not taking a single trend line
over a long-time period is likely to be satisfactory, while if one is investigating whether a
country is changing its intervention strategy under a managed ?oat a much more micro
approach needs to be taken and one may need a number of short-termchanges in trends.
For this purpose we are inclined to favor a judgmental approach while admitting that
this is open to possible subjective biases. There are of course a number of formal tests for
structural breaks (Bai and Perron, 1998, 2003.) It is often not clear, however, which
aspects of behavior are responsible for the statistical results. (On this issue see the
discussion in Ghosh (2011) of the con?icting ?ndings for structural breaks in India’s
exchange rate policy).
3.3 Issues of constructing intervention proxies
One of the most dif?cult problems in applying the EMP approach is that few
countries make data on their exchange market intervention publicly available.
Most empirical studies use changes in reserves as a proxy but acknowledge that this is
far from perfect. Reserves can change due to interest earnings, valuation changes,
and of?cial borrowings as well as intervention. And interventions can include actions
in forward, not just spot markets[5]. These problems have led some researchers to give
up using reserve measures all together (Ghosh et al., 2003.) This seems too strong a
reaction, however, since where reserve accumulations are very large, such as those at
times by Japan and Korea, there can be little doubt that intervention was a primary
cause.
One adjustment that can be made fairly easily is to subtract an estimate of interest
earnings from the reserve ?gures[6]. Since dollar values of reserves are used in most
cross-national empirical studies, valuation changes due to exchange rate movements
can also be important. Since exchange rate changes among reserve currencies are often
much larger over short periods than interest rate differences, estimates of the currency
composition of reserves are particularly important for such calculations. For many
countries published ?gures or good estimates about the currency composition of
reserves are not available, but efforts to take valuation changes into account have been
made in some country studies (Ouyang et al., 2008.) Where countries mark their reserve
?gures to market, interest rate changes in reserve currency countries can also affect
valuation for longer maturity assets.
A ?nal problem regarding the intervention proxy is that a constant amount of
intervention per period in the same direction would give rise to varying percentage
changes as reserve levels rose or fell[7]. Likewise, initial reserve levels can make a
substantial difference. Studies such as CR (2002), and HM (2003) characterize Japan as
having a low volatility in foreign reserves. Although Japan intervened heavily in
absolute terms in the foreign exchange market in the early 2000s, the percentage
changes in reserves were fairly small due to the high-initial levels of reserves.
There are several ways to deal with this problem. The most popular method has
been to use scaling variables for intervention proxies. Weymark (1997) and LYS use
the lagged money base, lagged narrow money and the sum of exports and imports for
12 months as scaling variables. This is also a problem for measures of currency crises
that are also based on the EMP concept but typically use unscaled percentage changes
in reserves and exchange rates.
Classifying
currency regimes
295
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.4 Rationales for intervention
There has been considerable debate about the degree of post-crisis exchange rate
?exibility in Asia[8]. While it is widely assumed that there has been a substantial
increase in exchange rate ?exibility, there has also been considerable accumulation of
reserves. This has ledsome economists to argue that there has really beenlittle change in
exchange rate policies (McKinnon and Schnabl, 2004.) We can help clarify this debate by
distinguishing (conceptually at least) between intervention designed to accumulate
reserves such as may be highly desirable after a period of reserve losses, intervention to
hold down the average level of the exchange rate for competitive advantage, and
intervention to smooth out short-run ?uctuations in the exchange rate[9].
The ?rst two motives will be observationally equivalent in terms of the statistical
data during the early stages of recouping reserve losses. In the later stages of reserve
accumulation distinctions would have to be based on judgments about whether reserve
accumulations were becoming “excessive”. The appropriate level of reserves for a
country can of course be a matter of considerable dispute.
The third type of intervention – to limit short-term ?uctuations in the exchange
rate – is more easily identi?ed. Indeed that is what our suggested methodology is
designed to capture, once we detrend changes in reserves as well as changes in the
exchange rate.
Even when we distinguish trends from deviations from trends, we have no clear
theoretical basis for comparing their relative importance in terms of the overall degree
of ?exibility of a regime. How do we compare the degree of ?exibility of two crawling
band regimes, one of which has a faster rate of crawl while the other has a wider band?
Since there is no clear theoretical basis for the weight in adding these two dimensions
together we suggest in the next section that a two-parameter approach is needed.
The same type of problem applies to the use of measures of interest rate variability
(Hausmann et al., 2001; Baig, 2001; CR, 2002; HM, 2003; Cavoli and Rajan, 2005; Siregar
and Pontines, 2005, etc.) It is clearly appropriate to use changes in interest rates or other
monetary variables such as money supplies or degrees of sterilization in classifying
countries’ overall monetary cum exchange rate regimes, but it is not clear how the
behaviors of these monetary policy variables should be related to the classi?cation of the
degree of ?exibility of the exchange rate regime. CR are able to make a strong link only
by making the assumption that interest rate changes are only used to limit exchange rate
movements. They argue correctly that “such interest rate volatility is not the result of
adhering to strict monetary targets in the face of large and frequent money
demand shocks [. . .] ” (p. 392). However, they then jump directly from this statement to
the conclusion that “Interest rate volatility would appear to be the byproduct of a
combination of trying to stabilize the exchange rate through domestic open market
operation and lack of credibility” (p. 392). This leaves out the possibility of the effects of
other types of shocks and of domestically motivated monetary policy actions dictated by
discretion rather than a monetary rule. Interest rate changes can also be used to protect
reserve levels. Thus, we see no clear basis for a presumption that higher interest rate
variability should be considered as an indicator of less ?exibility in the exchange rate
regime. This is certainly an issue worthy of further investigation.
Post-crisis exchange rate regimes in Asian countries have been important issues with
regard to this subject recently but there has still been considerable disagreement. While
the view that most of the crisis countries shifted frompre-crisis ?xed rates to post-crisis
JFEP
3,4
296
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
?oating rates has become popular, many scholars point out its oversimpli?cation.
Siregar and Pontines (2005) ?nd that Indonesia, Korea, and Thailand have maintained
a de-facto ?exible exchange rate regime during the post-1997 period but Singapore
has increased intervention. They use an intervention index which is composed of
probabilities that exchange rate, international reserves, and interest rate are in regimes
with high volatilities based on Markov-Regime Switching ARCH model. However,
CR (2002) argue that countries announcing of?cially free ?oating tend to still intervene
in the foreign exchange market, so-called fear of ?oating. McKinnon and Schnabl (2004)
even argue that post-crisis Asian foreign exchange policies shows strong intervention
and they indeed have returned to soft dollar peg. Dooley et al. (2003) describe recent
phenomenon including the Asian regimes as a new de facto Bretton Woods regime.
Finally, HM (2003) argue that post-crisis foreign exchange policies in East Asia have
become more ?exible than before but less than real free ?oating.
3.5 Sterilization and the effects of exchange rates on domestic monetary policy
Exchange rate policy is only one aspect of monetary regimes (Rose, 2007). An exchange
rate regime requires a corresponding domestic monetary policy to avoid generating
crises. In the short-run countries may use sterilized foreign exchange interventions to
break the link between exchange rate and monetary policy, for example to deal with
capital movements that are believed to be temporary. Unsterilized intervention on the
other hand will generate corresponding changes in domestic monetary aggregates.
Thus, it is important to distinguish between sterilized and unsterilized intervention.
As is emphasized in Mundell-Fleming models and trilemma analysis under ?xed
exchange rates and perfect capital mobility sterilization is not possible. The trilemma
analysis argues that one cannot have all three of ?xed exchange rates, independent
monetary policy, and absence of controls as these are the three main ways to correct
balance of payments disequilibria. With imperfect capital mobility, however, this set of
constraints can be violated in the short-run through sterilized intervention. Many
commentators have wrongly assumed that because China has been following a fairly
tight peg against the dollar it must be importing the USA’ monetary policy. This
misses that China has substantial capital controls and has been able to sterilize most of
its balance of payments surplus over a number of years (Ouyang et al., 2010).
There has been considerable controversy over how much attention countries
should pay to exchange rate and balance of payments developments in setting domestic
monetary policy. This is an issue even for open economies that have adopted in?ation
targeting. In this context Willett (2003a, b) has suggested that instead of just being used
to decide if a country should join a currency area or adopt some other form of hard ?x
such as a currency board, the optimal currency area criteria can be used to investigate
how much weight a country should give to foreign exchange market developments in
setting domestic monetary policy. This can run from 100 percent for a hard ?x to
0 percent with a free ?oat for a large country.
Many countries that have ?oating rates have practiced heavy intervention in the
foreign exchange market. Korea is a good example. Of?cials frequently describe
Korea’s exchange rate regime as free ?oating, but it has at times intervened heavily in
the foreign exchange market (Kim and Willett, 2010). Korea has adopted in?ation
targeting but the exchange rate seems to still play an important role in its monetary
policy. Both Eichengreen (2004) and Parsley and Popper (2009) have found
Classifying
currency regimes
297
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
that Korea’s overnight call rate has responded to movement in the dollar-won exchange
rate as well as to expected in?ation and the output gap. Ouyang and Willett (2010) ?nd
that a high proportion of Korea’s intervention in the foreign exchange market has been
sterilized.
Estimating the degree of sterilization can be tricky. Simple correlations do
not distinguish between whether changes in monetary policy are causing changes in
reserves or changes in reserves are causing changes in monetary policy. Thus, it is
necessary to estimate policy reaction functions to attempt to make this distinction. Many
of the issues involved in such efforts are discussed in the longer version of this paper
because it is necessary to attempt to distinguish.
There is also an issue of whether interest rate variability should be included in
measures of exchange rate ?exibility. It is clearly appropriate to use changes in interest
rates or other monetary variables such as money supplies or degrees of sterilization in
classifying countries’ overall monetary cum exchange rate regimes, but it is not clear
how the behaviors of these monetary policy variables should be related to the
classi?cation of the degree of ?exibility of the exchange rate regime. CRare able to make
a strong link only by making the assumption that interest rate changes are only used to
limit exchange rate movements. They argue correctly that “suchinterest rate volatility is
not the result of adhering to strict monetary targets in the face of large and frequent
money demand shocks [. . .] ” (p. 392). However, they then jump directly from
this statement to the conclusion that “Interest rate volatility would appear to be the
byproduct of a combination of trying to stabilize the exchange rate through domestic
open market operation and lack of credibility” (p. 392). This leaves out the possibility of
the effects of other types of shocks and of domestically motivated monetary policy
actions dictated by discretion rather than a monetary rule. Interest rate changes can also
be used to protect reserve levels. Thus, we see no clear basis for a presumption that
higher interest rate variability should be considered as an indicator of less ?exibility in
the exchange rate regime. This is certainly an issue worthy of further investigation.
4. Concluding remarks
In the limited space of this article there are many issues with which we have not been
able to deal. We hope, however, that we have given the reader a useful introduction to
the major large datasets available that classify exchange rate regimes and an overview
of some of the major conceptual issues involved in classifying exchange rate regimes
and their relation to domestic monetary policies.
In the longer version of this paper available on the web we provide examples of how
our favored EMP approach can be applied and how it compares with the other major
approaches using the cases of India, Korea, and Tanzania. We also discuss a number of
the research areas for which classi?cations of currency regimes are used. In particular
we review studies of the effects of currency regimes on international trade, economic
growth, in?ation and other aspects of macroeconomic discipline, and on currency and
?nancial crises and the debate about best to characterize post-crisis exchange rate
regimes in Asia.
In closing we would like to reiterate what we view as some of the most important
methodological conclusions that we have reached:
.
The degree of exchange rate ?exibility is not the only important international
dimension of a country’s currency policies. Also important are the degree
JFEP
3,4
298
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
to which international exchanges are controlled and the role given to
developments in the foreign exchange market in setting domestic monetary
policy. For this latter purpose it is important to distinguish between sterilized
and unsterilized interventions in the foreign exchange market. Since it is dif?cult
to know to what extent a country changes its interest to protect its exchange rate
or its reserves it is probably not a good idea to use changes in interest rates in
measures of exchange rate ?exibility.
.
Of?cial statements about exchange rate policies are often misleading guides to
the practices that countries actually follow.
.
In general the degree of ?exibility of an exchange rate regime cannot be measured
correctly by the behavior of the exchange rate alone. Lowlevels of ?uctuation can
be caused either by heavy intervention or by there being only small shocks.
To distinguish between these cases it is useful to use the concept of EMP and look
at the ratio of the pressures that lead to changes in the exchange rate versus
changes in reserves. This gives us an index of ?exibility or its inverse, the
propensity to intervene. This framework assumes that countries intervene to lean
against the wind, i.e. to slowdown exchange rate movements. Where governments
intervene to accelerate exchange rate movements the degree of exchange rate
?exibility is not well de?ned. Measures that use the ratios of variances do not
make such distinctions
.
In using this EMP approach it will often be important to distinguish between
relationships between trends in reserves and exchange rates and ?uctuations
around these trends.
.
Each country’s currency has exchange rates against many other currencies.
It will often be misleading to consider its exchange rate against only one other
currency such as the dollar or the euro.
.
One should be very suspicious of classi?cations that are binary, i.e. that only
distinguish between ?xed and ?exible exchange rates since there are many
versions of both ?xed and ?exible rates. For example, when investigating
relationships with macroeconomic discipline and currency and ?nancial crises it
is essential to distinguish between hard and soft ?xes. Likewise under a two-way
classi?cation it is not at all clear whether crawling band regimes should be
placed under ?xed or ?exible rates. By varying such placements a researcher
could often get almost any results that they wanted a priori.
.
There is no one best number of categories to use. That will dependonthe purpose of
the study. The same goes for the best time periods over which to measure regimes.
.
The most complex statistical measures are not always the most accurate. It is
sometimes not clear what they are actually measuring.
.
Since there is no perfect way to implement the concept of exchange rate
?exibility or its inverse, how heavily a government is managing its exchange
rate it is important for researchers to test the robustness of their results to
alternative methods.
This may seem a daunting list of considerations to take into account. However, the fact
that it is unlikely that we can create perfect measures should not de?ect us from trying
Classifying
currency regimes
299
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
to construct less imperfect measures, nor from using measures that we know are
imperfect but we think are the best available. When we do this, however, we should be
humble about our results and test them against alternative measures. Science often
progresses in little jumps.
Notes
1. Numbers in parenthesis refer to their ?ne classi?cation.
2. These issues are discussed in the longer version of this paper.
3. In the monetary model used by Girton and Roper this slope is one.
4. RR also use the similar approach. For example, RR classify a case as a de facto peg if the
probability that the monthly exchange rate change remains within 1 percent band over
rolling ?ve-year period is 80 percent or higher.
5. CR give a nice discussion of several of these issues.
6. Willett et al. (2007) adjust reserve changes for Japan for estimates of interest earnings, but
just ?nd that this makes little difference for their estimates. Both series fairly closely track
the data on Japan’s actual interventions over the period 1991-2005. Of course, this proxy may
not work as well for other countries, but our results for Japan suggest that this approach is
worth using in the absence of better information.
7. This problem can of course also apply to continued appreciation or depreciation.
8. This debate is discussed in the longer version of this paper.
9. Of course, there could also be the intervention to prop up the rate to avoid in?ation. This is
especially likely before elections.
References
Angkinand, A. and Willett, T.D. (2011), “Exchange rate regimes and banking crises: the channels
of in?uence investigated”, International Journal of Finance & Economics (in press).
Angkinand, A., Chiu, E.M.P. and Willett, T.D. (2009), “Testing the unstable middle and two
corners hypotheses about exchange rate regimes”, Open Economies Review, Vol. 20 No. 1,
pp. 61-83.
Bai, J. and Perron, P. (1998), “Estimating and testing linear models with multiple structural
changes”, Econometrica, Vol. 66 No. 1, pp. 47-78.
Bai, J. and Perron, P. (2003), “Computation and analysis of multiple structural change models”,
Journal of Applied Econometrics, Vol. 18 No. 1, pp. 1-22.
Baig, T. (2001), “Characterizing exchange rate regimes in post-crisis east asia”,
IMF Working Paper No. 125, International Monetary Fund, Washington, DC.
Bubula, A. and O
¨
tker-Robe, I. (2002), “The evolution of exchange rate regimes since 1990:
evidence from de facto policies”, IMF Working Paper No. 155, International Monetary
Fund, Washington, DC.
Calvo, G. and Reinhart, C. (2002), “Fear of ?oating”, Quarterly Journal of Economics, Vol. 117
No. 2, pp. 379-408.
Cavoli, T. and Rajan, R. (2005), “Have exchange rate regimes in Asia become more ?exible post
crisis? Re-visiting the evidence”, Centre for International Economic Studies Discussion
Paper No. 0503, University of Adelaide, Adelaide.
JFEP
3,4
300
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Chiu, E.M.P., Dechsakulthorn, S., Walter, S., Walton, J. and Willett, T.D. (2011), “The discipline
effects of ?xed exchange rates: the distinction between hard and soft pegs”, working paper,
Claremont Institute for Economic Policy Studies, Claremont, CA.
Dooley, M.P., Folkerts-Landau, D. and Garber, P.M. (2003), “An essay on the revived Bretton
Woods system”, NBER Working Paper No. 9971, National Bureau of Economic Research,
Cambridge, MA.
Eichengreen, B. (2004), “Monetary and exchange rate policy in Korea: assessments and policy
issues”, Discussion Paper No. 4676, Center for Economic and Policy Research,
Washington, DC.
Frankel, J. and Wei, S.J. (1994), “Yen bloc or dollar bloc? Exchange rates in the East Asian
economies”, in Ito, T. and Krueger, A. (Eds), Macroeconomic Linkage: Savings, Exchange
Rates, and Capital Flows, Univerity of Chicago Press, Chicago, IL.
Frankel, J. and Wei, S.J. (2008), “Estimation of de facto exchange rate regimes: synthesis of the
techniques for inferring ?exibility and basket weights”, NBER Working Paper No. 14016,
National Bureau of Economic Research, Cambridge, MA.
Ghosh, R. (2011), “Capital controls and exchange rate regime in India”, PhD dissertation,
UMI No. 3449833, Claremont Graduate University, Claremont, CA.
Ghosh, A.R., Gulde, A. and Wolf, H. (2003), Exchange Rate Regimes: Choices and Consequences,
MIT Press, Cambridge, MA.
Girton, L. and Roper, D. (1977), “A monetary model of exchange market pressure applied to the
postwar Canadian experience”, American Economic Review, Vol. 67, pp. 537-48.
Habermeier, K., Kokenyne, A., Veyrune, R. and Anderson, H. (2009), “Revised system for the
classi?cation of exchange rate arrangements”, IMF Working Paper No. 211, International
Monetary Fund, Washington, DC.
Hausmann, R., Panizza, U. and Stein, E. (2001), “Why do countries ?oat the way they ?oat?”,
Journal of Development Economics, Vol. 66, pp. 387-414.
Herna´ndez, L. and Montiel, P.J. (2003), “Post-crisis exchange rate policy in ?ve Asian countries:
?lling in the hollow middle?”, Journal of the Japanese and International Economics, Vol. 17
No. 3, pp. 336-69.
Kim, K.S., Willett, T.D. and Willett, T.D. (2010), “Korea’s exchange rate policies between the
crises”, The Global Crisis and Korea’s International Financial Policies, Korea Economic
Institute, Washington, DC, pp. 11-38.
Klein, M.W. and Shambaugh, J. (2010), Exchange Rate Regimes in the Modern Era, MIT Press,
Cambridge, MA.
Levy Yeyati, E. and Sturzenegger, F. (2005), “Classifying exchange rate regimes: deeds vs words”,
European Economic Review, Vol. 49, pp. 1603-35.
McKinnon, R. and Schnabl, G. (2004), “The East Asian dollar standard, fear of ?oating,
and original sin”, Review of Development Economics, Vol. 8 No. 3, pp. 331-60.
Ouyang, A.Y. and Willett, T.D. (2010), “International aspects of Korea’s monetary policies”,
in Willett, T.D., Auerbach, N.N., Kim, K.S., Kim, Y., Ouyang, A., Permpoon, O.,
Sompornserm, T., Srisorn, L., Sula, O. et al. (Eds), The Global Crisis and Korea’s
International Financial Policies, Korea Economic Institute, Washington, DC, pp. 76-88.
Ouyang, A.Y., Rajan, R.S. and Willett, T.D. (2008), “Managing the monetary consequences of
reserve accumulation in emerging Asia”, Global Economic Review, Vol. 37 No. 2, pp. 171-99.
Ouyang, A.Y., Rajan, R.S. and Willett, T.D. (2010), “China as a reserve sink”, Journal of
International Money and Finance, Vol. 29 No. 5, pp. 951-72.
Classifying
currency regimes
301
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Parsley, D. and Popper, H. (2009), “Evaluating exchange rate management: an application to
Korea”, Working Paper No. 282009, Hong Kong Institute for Monetary Research, Central.
Reinhart, C. and Rogoff, K. (2004), “The modern history of exchange rate arrangements:
a reinterpretation”, Quarterly Journal of Economics, Vol. 119, pp. 1-48.
Rose, A.K. (2007), “A stable international monetary system emerges: in?ation targeting is
Bretton Woods, reversed”, Journal of International Money and Finance, Vol. 26, pp. 663-81.
Siregar, R.Y. and Pontines, V. (2005), “External debt and exchange rate overshooting: the case of
selected East Asian countries”, School of Economics Working Paper No. 14, University of
Adelaide, Adelaide.
Weymark, D. (1995), “Estimating exchange market pressure and the degree of exchange market
pressure for Canada”, Journal of International Economics, Vol. 39, pp. 273-95.
Weymark, D. (1997), “Measuring the degree of exchange rate market intervention in a small open
economy”, Journal of International Money and Finance, Vol. 16, pp. 55-79.
Weymark, D. (1998), “A general approach to measuring exchange market pressure”,
Oxford Economic Papers, Vol. 50, pp. 106-21.
Willett, T.D. (2003a), “Fear of ?oating needn’t imply ?xed rates: feasible options for intermediate
exchange rate regimes”, Open Economies Review, Vol. 14 No. 1, pp. 71-91.
Willett, T.D. (2003b), “The OCA approach to exchange rate regimes”, in Dean, J.W., Salvatore, D.
and Willett, T.D. (Eds), The Dollarization Debate, Oxford University Press, New York, NY,
pp. 154-71.
Willett, T.D. and Kim, Y. (2006), “Korea’s post-crisis foreign exchange rate policy”, Korea’s
Economy, Korea Economic Institute, Washington, DC.
Willett, T.D., Kim, Y. and Nitithanprapas, I. (2007), “Taking the concept of exchange market
pressure seriously for classifying exchange rate regimes”, working paper, Claremont
Institute for Economic Policy Studies, Claremont, CA.
Willett, T.D., Nancy, N.A., Kenneth, S.K., Yoonmin, K., Alice, O., Orawan, P., Thana, S., Lalana, S.
and Ozan, S. (2010), The Global Crisis and Korea’s International Financial Policy,
Korea Economic Institute, Washington, DC.
Further reading
Chiu, E.M.P. and Willett, T.D. (2009), “The interactions of government strength and alternative
exchange rate regimes in avoiding currency crises”, International Studies Quarterly, Vol. 53
No. 4, pp. 1001-25.
Frankel, J. and Xie, D. (2010), “Estimation of de facto ?exibility parameter and basket weights in
evolving exchange rate regimes”, NBER Working Paper No. 15620, National Bureau of
Economic Research, Cambridge, MA.
About the authors
Thomas Willett (PhD, University of Virginia) is the Horton Professor of Economics at Claremont
Graduate University and Claremont McKenna College and Director of the Claremont Institute of
Economic Policy Studies and of its China-Asia Paci?c Political Economy Program. Before
coming to Claremont he was on the faculties of Cornell and Harvard Universities and served in
Washington, DC as a senior economist on the President’s Council of Economic Advisers and as
Director of international research at the US Treasury. He has also been a visiting scholar at the
International Monetary Fund and served as Editor or member of the editorial board of
14 professional journals in the areas of economics, public policy, and international relations.
He has written widely in the areas of monetary and ?nancial economics and the political
JFEP
3,4
302
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
economy of public policy with a particular focus on issues of international money and ?nance.
He has been a leader in the movement to better integrate economic and political analysis,
especially in the area of international political economy. His recent research has focused
particularly on international capital ?ows, currency policies, and currency and ?nancial crises
and issues of domestic and international ?nancial reform where he combines analysis from
political economy, behavioural ?nance, and complexity economics. Thomas Willett is the
corresponding author and can be contacted at: [email protected]
Eric M.P. Chiu is an Associate Professor at the Graduate Institute of National Policy and
Public Affairs at the National Chung Hsing University in Taiwan, a position he has held since
September 2006. He received his PhD from the Claremont Graduate University in International
Political Economy. His research concentrates on how politics interact with exchange rate regimes
and capital controls in shaping macroeconomic stability such as currency and ?nancial crises as
well as international capital movement, with speci?c focus on both emerging and developing
countries. His work has been published in International Studies Quarterly, Open Economies
Review, and Journal of Economic Policy Reform.
Sirathorn (B.J.) Dechsakulthorn is an Economist based in Los Angeles, California, USA.
Ramya Ghosh is a Faculty member at Drexel University’s LeBow College of Business. His
research interests are in the area of international ?nance focusing on international capital ?ows,
capital controls and exchange rate policies. Prior to joining Drexel University, Ghosh taught
Economics and Finance at University of California – Irvine and California State Polytechnic
University – Pomona. He was also a consultant for the World Bank and the evaluation of?ce of
the International Monetary Fund (IMF) in Washington, DC.
Bernard Kibesse is a PhD candidate in Economics at Claremont Graduate University
specializing in international money and ?nance. His current research interest is centred on
exchange rate regime, monetary policy and currency union of East African countries. Kibesse
also holds a Masters degree in ?nance from the University of Strathclyde (UK). He is also a
Certi?ed Public Accountant (CPA) and has been with the Central Bank of Tanzania for the
last 15 years.
Kenneth Kim is currently a PhD candidate at the Claremont Graduate University specializing
in the ?eld of International Money and Finance focusing on the Asian Crisis, Shared Mental
Models, and Korean exchange rate regime and policy. He has been teaching undergraduate
Intermediate Macroeconomics and International Economics courses at the California State
University, Fullerton and has also taught Public Finance (MS in Taxation Program) at the
California State University at Fullerton, Irvine Campus.
Jeff ( Yongbok) Kim has worked for the Bank of Korea since 1997 and joined in the
International Policy Analysis Division of the European Central Bank as a seconded economist in
January 2011. He specializes in International Money and Finance. Speci?c topics of interest
include exchange rate regimes, international capital ?ows, and Asian economic regionalism.
Alice Ouyang (PhDin Economics fromClaremont Graduate University) is Associate Professor
of China Academy of Public Finance and Public Policy at Central University of Finance and
Economics. Her research interests mainly focus on International Money and Finance, Chinese and
Asian Economies, Macroeconomics, and Quantitative Method. Her papers have been published in
Journal of International Money and Finance, Applied Economics, and Global Economic Review.
Classifying
currency regimes
303
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This article has been cited by:
1. Stephen G. Hall, Amangeldi Kenjegaliev, P.A.V.B. Swamy, George S. Tavlas. 2013. Measuring currency
pressures: The cases of the Japanese yen, the Chinese yuan, and the UK pound. Journal of the Japanese
and International Economies 29, 1-20. [CrossRef]
2. THOMAS D. WILLETT, JEFF (YONGBOK) KIM, ISRIYA NITITHANPRAPAS BUNYASIRI.
2012. MEASURING EXCHANGE RATE FLEXIBLITY: A TWO-PARAMETER EXCHANGE
MARKET PRESSURE APPROACH. Global Journal of Economics 01, 1250007. [CrossRef]
3. William R. Clark, Mark Hallerberg, Manfred Keil, Thomas D. Willett. 2012. Measures of financial
openness and interdependence. Journal of Financial Economic Policy 4:1, 58-75. [Abstract] [Full Text]
[PDF]
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
3
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_797384061.pdf