Finance Research Letters 12 (2015) 100–108

Contents lists available at ScienceDirect

Finance Research Letters

journal homepage: www.elsevier.com/locate/frl

Currency competition between the dollar

and euro: Evidence from exchange rate behaviors
Cheol S. Eun a, Soo-Hyun Kim b, Kyuseok Lee c,⇑
a
Scheller College of Business, Georgia Institute of Technology, Atlanta, GA 30308, USA
b
College of Business Administration, Soongsil University, Seoul 156-743, Republic of Korea
c
College of Business, KAIST, Seoul 130-722, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: Using a simple numeraire currency-independent metric to mea-

Received 14 September 2014 sure distances between currencies, it is found that for most of
Accepted 18 November 2014 the 25 sampled floating currencies, excluding the US dollar and
Available online 27 November 2014
euro, there has been a noticeable decrease in the ratio of a cur-
rency’s distance from the euro to its distance from the dollar dur-
JEL classification:
ing the period from 1999 to 2013. Evidence that exchange risk has
F31
F36
increased substantially is also found for dollar-based agents, while
G11 it has decreased for euro-based agents. Overall, the findings indi-
G15 cate that the influence of the euro, relative to the dollar, on other
currencies has increased since its introduction.
Keywords: Ó 2014 Elsevier Inc. All rights reserved.
Currency competition
Dollar
Euro
Exchange rates
Exchange risk

1. Introduction

The two key motives for the adoption of the euro were to enhance the role of Europe in the world
monetary system and to turn the European Union into a truly unified market (Krugman et al., 2012). In
relation to the second motive, a number of prior studies have examined the question of whether the
euro has actually contributed to the integration of the European market (see, for example, Rose, 2000;

⇑ Corresponding author at: KAIST College of Business, 85 Hoegi-ro, Dongdaemoon-gu, Seoul 130-722, Republic of Korea.
Tel.: +82 2 958 3443.
E-mail addresses: cheol.eun@scheller.gatech.edu (C.S. Eun), soo_hyun.kim@ssu.ac.kr (S.-H. Kim), kslee2@business.kaist.ac.kr
(K. Lee).

http://dx.doi.org/10.1016/j.frl.2014.11.003
1544-6123/Ó 2014 Elsevier Inc. All rights reserved.
 C.S. Eun et al. / Finance Research Letters 12 (2015) 100–108 101

Rose and van Wincoop, 2001; Micco et al., 2003; Engel and Rogers, 2004; Dominguez, 2006; Lane,
2006; Abad, 2009; European Central Bank, 2013, 2014). In relation to the first motive, however, there
is little evidence regarding the extent to which the euro (EUR), relative to the US dollar (USD), has
influenced the exchange rate behaviors of other currencies. There have been numerous studies that
have examined such questions as how the world monetary system will evolve over time and whether
the euro will surpass the dollar as the leading vehicle currency (see, for example, Fratianni and
Hauskrecht, 1998; Portes and Rey, 1998; Mundell, 2000; Bergsten, 2002; Kenen, 2002; Chinn and
Frankel, 2005; Eichengreen, 2005; Cohen, 2007; Norrlof, 2009, 2014; Pisani-Ferry and Posen, 2009;
Fields and Vernengo, 2013). However, most of these studies have focused on power analyses of the
dollar against the euro with regard to the reserve position of the two currencies in the central bank
holdings and to the use of the two currencies in international transactions.
The objective of this paper is to examine the currency competition between the dollar and euro by
directly focusing on the influence of the dollar and euro on the exchange rate behaviors of other cur-
rencies and by examining how their relative influence has evolved during the period since the intro-
duction of the euro (1999–2013). In doing so, we introduce a simple distance metric that is
independent of the choice of the numeraire currency in expressing bilateral exchange rates. Using this
distance metric, we first examine the temporal variation of the natural logarithm value of the ratio of a
currency’s distance from the euro to its distance from the dollar, hereafter called the currency distance
ratio. For a given currency, if the influence of the euro relative to that of the dollar on this currency
increases over time, the movement of the currency will more closely track the movement of the euro
than that of the dollar; thus, the currency distance ratio will decrease over time.1 We next examine the
temporal variation of foreign exchange risk for dollar-based and euro-based agents. If the influence of the
euro relative to the dollar on other currencies increases over time, the exchange risk is likely to increase
for dollar-based agents and decrease for euro-based agents.
The results from the sample period of 1999–2013 indicate that, for most of the 25 floating curren-
cies sampled, excluding USD and EUR, there has been a noticeable decrease in the currency distance
ratio. Furthermore, the finding remained robust when subjected to several sensitivity tests. In partic-
ular, the influences of other major currencies, such as JPY, GBP, and CHF, relative to the USD on lesser
currencies are demonstrated to be significantly less than that of EUR. Next, the results for the temporal
variation of exchange risk indicate that the exchange risk increased substantially for dollar-based
agents, while it decreased somewhat for euro-based agents, over two equal sub-periods. Taken
together, our findings indicate that the influence of the euro relative to the dollar on the exchange rate
behaviors of other currencies has increased over time.
In Sections 2 and 3 of this paper, we discuss the data and methodology, respectively. In Section 4,
we present our key empirical results; Section 5 presents our conclusions.

2. Data

Our sample spans the 15-year period from January 1, 1999 (Friday) to December 27, 2013 (Friday),
and it includes 27 currencies: those of Australia, Brazil, Canada, Chile, Colombia, Czech Republic, Euro
area, India, Indonesia, Israel, Japan, Korea (South), Malaysia, Mexico, New Zealand, Norway, Philip-
pines, Poland, Russian Federation, Singapore, South Africa, Sweden, Switzerland, Thailand, Turkey,
United Kingdom, and United States.
We selected the sample currencies based on the following procedure. First, we started with the 33
currencies that are consistently ranked in the top 35 in terms of their foreign exchange (FX) turnover
share according to Triennial Central Bank Survey reports from April 2001, 2004, 2007, 2010, and 2013
of the Bank for International Settlements (BIS).2 On average, the 33 currencies selected comprise 97.3%
of the total FX turnover. Next, we examined the history of the exchange rate arrangements and monetary
policy frameworks conducted annually by the International Monetary Fund (IMF) in order to only focus

1
In Section 3, we provide a more formal reasoning using a bipolar model as an approximation for the world monetary system.
2
We consulted various issues in the ‘‘Triennial Central Bank Survey: Foreign Exchange and Derivatives Market Activity’’
published by the BIS. The survey reports the ranking of currencies up to the 35th rank, based on the FX turnover share.
102 C.S. Eun et al. / Finance Research Letters 12 (2015) 100–108

on currencies that were identified as independently floating or managed floating.3 For the purpose of this
paper, we excluded five currencies from the sample (Hong Kong dollar, Danish krone, Taiwanese dollar,
Chinese renminbi, and Saudi riyal) because they were not classified as floating throughout the sample
period. Finally, we excluded one additional currency from the sample (Hungarian forint) because it
was not classified as floating until 2008. Therefore, 27 currencies remained in the final sample.
We used the bilateral exchange rates against the USD sampled weekly on Fridays, which were
obtained from Datastream. In particular, we used the World Market Reuters (WMR) closing spot rates,
which report the mid-quote of the daily closing bid and ask, and these are recorded at 4:00 p.m. Lon-
don time. We used the weekly exchange rates in order to reduce the noise in the data and to mitigate
the problems associated with time zone differences in simultaneously examining the exchange rate
behaviors of many currencies while maintaining a reasonably high frequency. In addition, for each
sample currency, we used its exchange rate data starting from the date when the currency was first
classified as floating during the sample period according to the IMF classification.4 The sampling began
on January 1, 1999, for 21 currencies, but began later for six currencies. The sample start date of each
currency is reported in Table 1.

3. Methodology

We use sX/$ to denote the log of the spot exchange rate in units of currency X per USD, and
DsX=$
t ¼ sX=$
t  sX=$
t1 to denote the weekly change in the log exchange rate during week t (from the
end of week t  1 to the end of week t).
As a measure of dissimilarity between any two currencies, e.g. currencies X and Y, during the period
from week t = 1 to t = T, we use the Euclidean distance between the two T-dimensional vectors, i.e.
DsX=$ ¼ ðDsX=$ X=$
1 ; . . . ; DsT Þ and Ds
Y=$
¼ ðDsY=$ Y=$
1 ; . . . ; DsT Þ; this is denoted by d(X, Y). That is,

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
XT  X=$ 2
dðX; YÞ ¼ t¼1
Dst  DsY=$ t :

Note that the more similarly the two currencies X and Y behave, the smaller the value d(X, Y) will
have.
The currency distance metric (d) has an important property: the distance between any two curren-
cies is independent of the numeraire currency chosen (i.e. USD in our study). In order to explain this
briefly, suppose that in the beginning we use the Japanese yen (JPY) as the numeraire currency and use
sX=U to denote the log of the spot exchange rate in units of currency X per JPY. Then, the claimed inde-
pendence property can be easily obtained from the triangular parity among currencies X, JPY, and USD
(i.e. DsX=U
t ¼ DsX=$ t  DsU=$
t ) because the currency distance between any two currencies, e.g. currencies X
and Y, is based on the deviation of DsX=U t  DsY=U
t ¼ DsX=$
t  DsY=$
t . Note that changing the numeraire cur-
rency from USD to JPY is equivalent to changing the origin of the T-dimensional Euclidean space from
Ds$/$ = (0, . . ., 0) to DsU=$ ¼ ðDsU=$ U=$
1 ; . . . ; DsT Þ.
Before proceeding, it is worthwhile to mention that any dissimilarity metric defined using the
correlation between DsX/$ and DsY/$ does not have a numeraire-independence property. This
results from the correlation between DsX/$ and DsY/$ not being the same as the correlation between
DsX/Z = DsX/$  DsZ/$ and DsY/Z = DsY/$  DsZ/$ for any currency Z.
As a way of investigating the temporal variations of the influence of the dollar versus the euro, we
examined the time trends of the log of the ratio of a currency’s distance from the euro to its distance

3
We consulted various issues in the ‘‘Annual Report on Exchange Arrangements and Exchange Restrictions’’ published by the
IMF. Regarding the changes in the IMF regime classification system in 2009, refer to Habermeier et al. (2009).
4
The Russian ruble has been classified as a managed floating currency throughout the sample period except for the period from
March 1, 2006 to November 1, 2008, when it was classified as a conventional pegged currency. However, removing the Russian
data from this period does little to affect the overall results reported in this study. The Singapore dollar and Indonesian rupiah have
been classified as floating throughout the sample period, except for the period after November 9, 2011 and June 1, 2012,
respectively, when they were classified as crawl-like arrangements. Again, deleting the data during these periods does little to
affect the overall results reported in this study.
 C.S. Eun et al. / Finance Research Letters 12 (2015) 100–108 103

Table 1
Time trends of the currency distance ratio. The currency distance ratio is defined by the natural logarithm value of a currency’s
distance from the euro divided by its distance from the dollar.

Currency Sample start date (1) Currency distance ratio (2) Difference (3) Time trend of
the currency
distance ratio
computed annually
1999/01–2006/06 2006/07–2013/12 Coef. t-Stat.
Australian dollar 1999/01/01 0.059 0.177 0.118 0.010 0.87
Brazilian real 1999/02/01 0.130 0.108 0.238 0.028 2.50
Canadian dollar 1999/01/01 0.350 0.003 0.353 0.042 2.19
Chilean peso 1999/01/01 0.339 0.016 0.355 0.039 3.19
Colombian peso 1999/09/25 0.558 0.038 0.520 0.054 3.45
Czech koruna 1999/01/01 0.745 0.643 0.103 0.005 0.44
Euro 1999/01/01 n.a. n.a. n.a. n.a. n.a.
Indian rupee 1999/01/01 1.254 0.214 1.040 0.148 8.41
Indonesian rupiah 1999/01/01 0.113 0.316 0.204 0.046 2.40
Israeli shekel 2004/01/01 0.597 0.016 0.581 0.073 2.70
Japanese yen 1999/01/01 0.113 0.241 0.128 0.011 0.70
Korean won 1999/01/01 0.439 0.009 0.430 0.023 1.56
Malaysian ringgit 2005/07/21 1.003 0.355 0.648 0.100 2.15
Mexican peso 1999/01/01 0.498 0.008 0.490 0.050 5.99
New Zealand dollar 1999/01/01 0.050 0.160 0.110 0.010 0.99
Norwegian krone 1999/01/01 0.578 0.432 0.147 0.010 1.10
Philippine peso 1999/01/01 0.660 0.453 0.207 0.020 1.02
Polish zloty 2000/04/12 0.282 0.450 0.168 0.040 4.31
Russian ruble 1999/01/01 0.487 0.142 0.629 0.116 3.40
Singapore dollar 1999/01/01 0.682 0.320 0.362 0.037 3.70
South African rand 1999/01/01 0.018 0.110 0.092 0.016 1.60
Swedish krona 1999/01/01 0.666 0.507 0.159 0.006 0.39
Swiss franc 1999/01/01 1.101 0.405 0.695 0.032 1.14
Thai baht 1999/01/01 0.507 0.478 0.030 0.010 0.55
Turkish lira 2001/02/22 0.011 0.092 0.081 0.004 0.37
UK pound 1999/01/01 0.170 0.125 0.045 0.000 0.02
US dollar 1999/01/01 n.a. n.a. n.a. n.a. n.a.
Average 0.162 0.037 0.199

from the dollar (i.e. its currency distance ratio). If the world monetary system can be approximated
using a bipolar model of5

V X ¼ cx  V 1w
$
x
 Vw
€  expðeÞ;
x

where VX represents the value of currency X and e represents all omitted influences with mean zero,
then we have

sX=$ ¼ wx s€=$ þ lnðcx Þ þ e and sX=€ ¼ ð1  wx Þs$=€ þ lnðcx Þ þ e;

thus, dðX; $Þ  wx dð€; $Þ and dðX; €Þ  ð1  wx Þdð€; $Þ. Then, the currency distance ratio ln [d(X, €)/
d(X, $)] will be approximately equal to ln [(1  wx)/wx] and will decrease as wx increases over time.
Because wx can be viewed as a proxy for the influence of EUR relative to USD on currency X, a negative
time trend in the currency distance ratio can be interpreted as the influence of EUR relative to USD on
the currency increasing over time.

5
Fratianni and Hauskrecht (1998) describe the bipolar structure of the international monetary system as a structure that
consists of two currency areas, each of which attracts other countries to their gravitational centers. However, we could not find the
bipolar structure in the international monetary system formulated mathematically in the literature. Although it is ad hoc, our
model is simple and appears to well approximate the hypothetical bipolar structure of the international monetary system that
Fratianni and Hauskrecht (1998) described.
104 C.S. Eun et al. / Finance Research Letters 12 (2015) 100–108

Table 2
Number of portfolios of size N constructed.

Number of currencies, N
N=1 2 3–23 24 25 26
Number of portfolios of size N constructed 26 325 1000 325 26 1

Next, we investigated the effect of diversification on the foreign exchange risk from the perspective
of the dollar-based and euro-based agents. In particular, following Evans and Archer (1968), Wagner
and Lau (1971), and Solnik (1974), we examined the relationship between the riskiness of a currency
portfolio and the number of currencies included in the portfolio. From the perspective of USD-based
agents, for example, for given N, we randomly selected N currencies from the 26 other sample curren-
cies, i.e. excluding USD, without replacement. Then, for the equally weighted portfolio of the N sam-
pled currencies, we computed the variance of the changes in the weekly bilateral USD exchange rate.6
We repeated this procedure 1000 times, computed their average variances, and denoted it using Var(PN).
As a measure of the effect of the diversification on the reduction of the foreign exchange risk, we then
computed the ratio of Var(PN) to Var(P1). The ratio Var(PN)/Var(P1) illustrates the relationship between
the number of currencies in a portfolio and the risk of the portfolio relative to that of a typical one-cur-
rency portfolio. As a method of investigating the temporal variation of the effect of diversification on for-
eign exchange risk, we divided the sample period into two equal sub-periods and examined how the
ratio Var(PN)/Var(P1) changed over time.
Because the total number of possible combinations that select N out of 26 currencies is not too
large for N values equal to 1, 2, 24, 25, or 26, we computed Var(PN) for those Ns using all possible com-
binations. Table 2 provides the number of portfolios generated for the portfolio of size N.

4. Results

4.1. Currency distance ratio

Table 1 illustrates that, for the majority of 25 floating sample currencies (excluding USD and EUR),
the currency distance ratio decreased over the sample period. In particular, the columns labeled (1)
and (2) demonstrate that over the two equal sub-periods, the ratio decreased for 18 out of 25 curren-
cies. The ratio increased for seven currencies, but only the Swiss franc exhibited a notable increase.7
On average, the ratio decreased from 0.162 to 0.037 (a difference of 0.199) over the two sub-periods,
and the difference of 0.199 is significant at the 5% level based on both the paired sample t-test
(p = 0.011) and Wilcoxon signed rank sum test (p = 0.010). Next, the column labeled (3) reports the coef-
ficient estimates on yearly basis and the associated t-statistics from the ordinary least squares (OLS)
regression of the currency distance ratio, computed annually, on the intercept and year. This demon-
strates that the time trend coefficients were negative for 18 out of 25 currencies. Despite the sample size
being very small (ranging from 9 to 15), the t-statistics for the negative time trend coefficients were
greater than the two in absolute value terms for 11 currencies.
Fig. 1 presents the time trend of the average currency distance ratio. It confirms the overall nega-
tive time trend of the currency distance ratio. Although the average currency distance ratio increased
from 2010, which was when the Eurozone crisis intensified, the average ratio was far smaller during
the second sub-period compared with the first sub-period.

6
Although the sampling of the exchange rates of six currencies began later than that for the other 20 currencies, we used
exchange rate data starting from January 1, 1999 for all 27 sampled currencies for the analysis of the effect of diversification on
foreign exchange risk.
7
Gençay and Gradojevic (2013) show that the probability of informed trading for the USD/CHF exchange rate is higher than that
for the EUR/USD exchange rate. A possible interpretation of the increased distance ratio for the CHF may be that it was driven away
by the informed traders. In addition, since late 2009 when Europe’s sovereign debt crisis began, investors began to sell euro-
denominated assets and purchase Swiss francs and other safe haven assets on a large scale. As a result of this portfolio shift, the
Swiss franc may have stopped tracking the euro closely in more recent years, which may have led to an increased distance ratio.
 C.S. Eun et al. / Finance Research Letters 12 (2015) 100–108 105

Fig. 1. Time trend of the average currency distance ratio. The currency distance ratio is defined by the natural logarithm value
of a currency’s distance from the euro divided by its distance from the dollar.

Table 3
Sub-period analysis of the average pairwise correlation of the currency exchange rate changes measured against USD and EUR.

Average pairwise correlation of the currency Average pairwise correlation of the currency
exchange rate changes measured against USD exchange rate changes measured against EUR
First sub-period 0.2731 0.3790
(1999/01–2006/06)
Second sub-period 0.4340 0.3187
(2006/07–2013/12)

As a robustness test, we repeated our analysis by restricting the sample period to weeks where the
average of the absolute exchange rate changes was smaller than its 99-percentile or 95-percentile val-
ues during the sample period.8 We performed this robustness test because the currency distance metric,
which is based on the Euclidean distance, could be influenced heavily by a few extreme exchange rate
change observations. The results from the robustness test remained qualitatively the same. We also
repeated our analysis using the currency distance based on the mean absolute deviation (MAD). It is clear
that the MAD-based distance is more robust to outliers than the Euclidean distance, which is based on
the mean quadratic deviation. However, we found that the overall results remained robust.
As another robustness test for our results, we examined the time trends of the average pairwise
correlation of the currency exchange rate changes measured against USD and EUR.9 Table 3 reports
the average pairwise correlations computed for the two sub-periods. Fig. 2 depicts the time trend of
the currency correlation ratio, which was defined using the natural logarithm value of the average pair-
wise correlation of the currency exchange rate changes measured against EUR divided by that measured
against USD. Generally, if currencies co-move more closely with USD, the average pairwise correlation of
the currency exchange rate changes measured against USD is likely to decrease. Likewise, if currencies
co-move more closely with EUR, the average pairwise correlation of currency exchange rate changes
measured against EUR is likely to decrease. Thus, the results of Table 3 and Fig. 2 have the same inter-
pretation as the previous results. That is, the influence of the euro, relative to the dollar, on other curren-
cies increased during the sample period.

4.2. Effects of other major currencies

Although USD and EUR are two dominant currencies, other major currencies, such as the Japanese
yen (JPY), British pound (GBP), and Swiss franc (CHF), might have also affected the currency exchange
rate behaviors of many lesser currencies. Thus, as a further robustness test of our results, we also eval-

8
In particular, we follow Menkhoff et al. (2012) to compute the average of the absolute exchange rate changes over a weekly
interval.
9
We thank the anonymous referee for proposing the use of this rolling correlation analysis.
106 C.S. Eun et al. / Finance Research Letters 12 (2015) 100–108

Fig. 2. Time trend of the average currency correlation ratio. The currency correlation ratio is defined by the natural logarithm
value of the average pairwise correlation of the currency exchange rate changes measured against EUR divided by that
measured against USD.

Table 4
Changes in the currency distance ratio among the major currencies between two equal sub-periods. The number in each cell
represents the difference of the currency distance ratio between the two equal sub-periods. The numbers in parentheses and
square brackets are the p-values based on the paired-sample t-test and the Wilcoxon signed rank sum test, respectively.

X = EUR X = USD X = JPY X = GBP X = CHF

Change in the cross-currency average of the log ratio of a 0.199 0.000 0.036 0.105 0.070
currency’s distance from X to its distance from USD between (0.011) (0.453) (0.083) (0.340)
two equal sub-periods [0.010] [0.175] [0.146] [0.323]
Change in the cross-currency average of the log ratio of a 0.000 0.199 0.235 0.094 0.128
currency’s distance from X to its distance from EUR between (0.011) (<0.001) (0.002) (<0.001)
two equal sub-periods [0.010] [<0.001] [<0.001] [<0.001]

Fig. 3. Foreign exchange risk. The figure plots the relationships between the number of currencies in a currency portfolio (N)
and the risk of the portfolio relative to that of a typical one-currency portfolio (Var(PN)/Var(P1)) for two equal sub-periods.

uated the influences of JPY, GBP, and CHF, relative to USD and EUR, on the exchange rate behaviors of
other lesser currencies.10 Table 4 reports the results.
The results of the currency distance ratio measured against USD indicate that (i) the average cur-
rency distance from EUR relative to USD decreased significantly (at the 5% level) over the two sub-

10
We thank the anonymous referee for this suggestion.
 C.S. Eun et al. / Finance Research Letters 12 (2015) 100–108 107

periods and that (ii) the average currency distance from JPY, GBP, and CHF relative to USD exhibited no
significant difference (at the 5% level) over the two sub-periods. In addition, the results of the currency
distance ratio measured against EUR indicated that the average currency distance from USD, JPY, GBP,
and CHF relative to EUR increased significantly (at the 5% level) over the two sub-periods. This indi-
cates that the behaviors of other currencies were more tightly linked to EUR than they were to USD,
JPY, GBP, and CHF during the second sub-period compared with the first sub-period. Taken together,
the results of Table 4 indicate that the influence of EUR, relative to USD, on other lesser currencies
increased significantly over the two sub-periods and that this finding cannot be explained by the influ-
ences of other hard currencies such as JPY, GBP, and CHF.

4.3. Foreign exchange risk

Fig. 3 illustrates the relationship between the number of currencies in a currency portfolio and the
risk of the portfolio relative to that of a typical one-currency portfolio for two equal sub-periods for dol-
lar-based and euro-based agents. The graph on the left presents the relationship from the perspective of
a dollar-based agent. It indicates clearly that the exchange risk increased substantially over the two
sub-periods: specifically, the ratio Var(P26)/Var(P1) increased from 21.1% to 44.4%. In contrast, the graph
on the right, which illustrates the relationship from the perspective of a euro-based agent, indicates
that the exchange risk decreased, albeit not substantially, over the two sub-periods; specifically, the
ratio Var(P26)/Var(P1) decreased from 39.3% to 34.8%. Taken together, the two graphs in Fig. 3 illustrate
that during the first sub-period, the exchange risk was much higher for euro-based agents than for dol-
lar-based agents, but also that during the second sub-period, the situation reversed.

5. Conclusion

Using a simple distance metric between currencies, we demonstrated that the majority of floating
currencies behave more similarly to the euro relative to the dollar than they did earlier, for the period
from 1999 to 2013. In addition, we also presented evidence that the exchange risk increased substan-
tially over this period for dollar-based agents, whereas it decreased somewhat for euro-based agents.
In summary, our findings indicate that the influence of the euro relative to the dollar on the exchange
rate behaviors of other currencies has increased since the introduction of the euro. The euro now
appears to have established its status as another leading global currency, which rivals the dollar, as
the European Union intended with the launch of the euro.

Acknowledgments

We thank Ramazan Gençay (the Editor-in-Chief) and the anonymous referee for their valuable
comments and suggestions that improved the paper. We also thank Trisha Poole for English help.
All remaining errors are our own. Kyuseok Lee acknowledges the financial support from KAIST (Grant
No. 04120011).

References

Abad, P., 2009. EMU and European Government Bond Market Integration. ECB Working Paper, No. 1079. <http://
www.ecb.europa.eu/pub/pdf/scpwps/ecbwp1079.pdf>.
Bergsten, C.F., 2002. The euro versus the dollar: will there be a struggle for dominance? J. Policy Model. 24, 307–314.
Chinn, M., Frankel, J., 2005. Will the Euro Eventually Surpass the Dollar as Leading International Reserve Currency? NBER
Working Paper, No. 11510. <http://www.nber.org/papers/w11510>.
Cohen, B.J., 2007. Enlargement and international role of the euro. Rev. Int. Polit. Econ. 14, 746–773.
Dominguez, K.M.E., 2006. The European Central Bank, the euro and global financial markets. J. Econ. Perspect. 20, 67–88.
Eichengreen, B., 2005. Sterling’s Past, Dollar’s Future: Historical Perspectives on Reserve Currency Competition. NBER Working
Paper, No. 11336. <http://www.nber.org/papers/w11336>.
Engel, C., Rogers, J.H., 2004. European product market integration after the euro. Econ. Policy 19, 347–384.
European Central Bank, 2013. Intra-euro Area Trade Linkages and External Adjustment. ECB Monthly Bulletin (January 2013),
pp. 59–74. <http://www.ecb.europa.eu/pub/pdf/other/art2_mb201301en_pp59-74en.pdf>.
European Central Bank, 2014. Financial Integration in Europe. ECB Annual Report (April 2014), pp. 59–74. <http://
www.ecb.europa.eu/pub/pdf/other/financialintegrationineurope201404en.pdf>.
108 C.S. Eun et al. / Finance Research Letters 12 (2015) 100–108

Evans, J.L., Archer, S.H., 1968. Diversification and the reduction of dispersion: an empirical analysis. J. Financ. 23, 761–767.
Fields, D., Vernengo, M., 2013. Hegemonic currencies during the crisis: the dollar versus the euro in a Cartalist perspective. Rev.
Int. Polit. Econ. 20, 740–759.
Fratianni, M., Hauskrecht, A., 1998. From the gold standard to a bipolar monetary system. Open Econ. Rev. 9, 609–635.
Gençay, R., Gradojevic, N., 2013. Private information and its origins in an electronic foreign exchange market. Econ. Model. 33,
86–93.
Habermeier, K.F., Kokenyne, A., Veyrune, R., Anderson, H., 2009. Revised System for the Classification of Exchange Rate
Arrangements. IMF Working Paper, No. 09-211. <http://www.imf.org/external/pubs/ft/wp/2009/wp09211.pdf>.
Kenen, P.B., 2002. The euro versus the dollar: will there be a struggle for dominance? J. Policy Model. 24, 347–354.
Krugman, P.R., Obstfeld, M., Melitz, M.J., 2012. International Economics: Theory and Policy, ninth ed. Pearson Education
Limited, New York, USA.
Lane, P.R., 2006. The real effects of European monetary union. J. Econ. Perspect. 20, 47–66.
Menkhoff, L., Sarno, L., Schmeling, M., Schrimpf, A., 2012. Carry trades and global foreign exchange volatility. J. Financ. 67, 681–
718.
Micco, A., Stein, E., Ordonez, G., 2003. The currency union effect on trade: early evidence from EMU. Econ. Policy 18, 316–356.
Mundell, R., 2000. Currency areas, volatility and intervention. J. Policy Model. 22, 281–299.
Norrlof, C., 2009. Key currency competition: the euro versus the dollar. Coop. Confl. 44, 420–442.
Norrlof, C., 2014. Dollar hegemony: a power analysis. Rev. Int. Polit. Econ. 21, 1042–1070.
Pisani-Ferry, J., Posen, A.S., 2009. The Euro at Ten: The Next Global Currency? Peterson Institute for International Economics,
Washington, DC, USA.
Portes, R., Rey, H., 1998. The emergence of the euro as an international currency. Econ. Policy 26, 307–343.
Rose, A.K., 2000. One money, one market: the effect of common currencies on trade. Econ. Policy 15, 9–45.
Rose, A.K., van Wincoop, E., 2001. National money as a barrier to international trade: the real case for currency union. Am. Econ.
Rev. 91, 386–390.
Solnik, B.H., 1974. Why not diversify internationally rather than domestically? Financ. Anal. J. 30, 48–54.
Wagner, W.H., Lau, S.C., 1971. The effect of diversification on risk. Financ. Anal. J. 27, 48–54.