Journal of Business and Policy Research

Vol. 9. No. 2. December 2014. Pp. 65 – 83

Exchange Rate Policy Conflict

An Obstacle in the Way of East Asian Regional Economic
Cooperation
Maryam Ishaq1 and Muhammad Atiq Ur Rehman2
Association of South East Asian Nations (ASEAN) is acknowledged to be the
second largest regional cooperation association after European Union. However,
unluckily, the region exhibits highly varied and unmatched exchange rate
arrangements, though the growing interdependence of East Asian economies
calls for ensuring intra-regional macroeconomic stability particularly with
reference to monetary policy and exchange rate coordination. The study has
explored whether the real effective exchange rate misalignments in each
member states of the region are more contributed from regional partners or from
non-regional ones. The empirical verification is done through ARDL and Zivot
single equation Error Correction Model and the long run estimates are corrected
rd
for their super consistency through 3 step of Engle-Yoo ECM. Variance
decomposition test is employed in order to see the variability in exports imparted
by the exchange rate of each of regional and non-regional trading partners
individually. Results fairly suggest that East Asian economies are suffering more
from intra-regional exchange rate volatility in comparison to ex regional
disturbances. The fact leads towards a strong support for the hypothesis that
within RIA exchange rate misalignments are more critical in the perspective of
currency crisis in a region in comparison to the same misalignments arising
somewhere else. The probable solutions to the problem is to keep foremost
focus on deviation in exchange rate policies while selecting trading partners,
ensuring more coordinated macroeconomic policy and development of a more
accommodating international financial architecture.

Keywords: Regional cooperation, real effective exchange rate, monetary policy coordination,
exchange rate misalignment, international capital markets.

1. Introduction
The growing interdependence of East Asian economies calls for ensuring intra-regional
macroeconomic stability particularly with reference to monetary policy and exchange rate
coordination, Bayoumi and Mauro (2001). This will indeed be helpful in synchronizing intra-
regional trade and investment and will surely provide the region with a high way to realize its
dream of economic regionalism. It is foreseen that in the next five years East Asia will be the
largest regional trading bloc, probably larger than European Monetary Union. So, under this
anticipation, it is inevitable for the participant economies to have a harmonized and well-
coordinated intra-regional exchange rate mechanism.

1 Doctoral Student (Department of Economics and Finance) University of Canterbury, Christchurch 8140,
Newzealand,
E-mail: maryam.ishaq@pg.canterbury.ac.nz
2 Doctoral Student (Department of Economics),University of the Punjab, Lahore 54890, Pakistan,
E-mail: atiq164@live.com
 Ishaq & Rehman
Nevertheless, the task is highly challenging in today‟s global environment where economies
all over the world are tightly integrated and up to some degree economically interdependent
upon each other. The global financial crisis of 2007 originated from United States has made
the world economy exceedingly volatile and irregular. The tempt for developing Asia‟s strong
fundamentals along with unstable global economic setting has lowered down the
international investors risk appetite, thus lead towards growing volatility in capital
movements.

Keeping all these facts in view, a greater extra-regional exchange rate flexibility against
developed economies may prove to be highly accommodating in a sense that it will indeed
bring macroeconomic and financial stability to the region. Thus Asian economies may gain
much from ensuring a relatively stable exchange rate within the region and opting for
relatively flexible exchange rates against their ex-regional trading partners. But this could
only be achieved by establishing a certain degree of exchange rate policy compatibility.

The depreciation of any other regional member state‟s currency may put all the other
currencies of the region into sheer currency crisis. Specially, it can undermine the integrity of
partners‟ commitment to a fixed parity and may cause to generate speculative attack on its
currency. Such a problem is most likely to occur during the times of financial chaos. The fact
is well evident from Asian financial crisis of 1998 which actually originated from Thailand but
made Indonesia, Philippines and Malaysia to see sizeable currency depreciations and
caused momentous devastations in their financial system. In July 1997, the devaluation of
Thai Baht was prompted by investors‟ observations of worsening financial environment of the
region and of indefensible overvaluation of currency. Resultantly, within one month of this
collapse of Thai Baht, the financial catastrophe entered Philippines, Indonesia, Malaysia and
Korea in an unleashed manner. The momentum and extent of this „regional financial
infection‟ was really frightening as well as alarming.

Keeping in view rising interdependence of East Asian economies, it is highly desirable to

ensure stable and if necessary homogenous kind of intra-regional exchange rate
arrangements through policy synchronization, Kwack and Yeung (2004). US Dollar has
remained no good bench mark after the Asian financial crisis of 1997-98 and global financial
crisis of 2007-08, so efforts should be directed towards obtaining regional exchange rate
stability despite of retaining stability vis-à-vis US Dollar, Taguchi (2010).

However, unluckily, the region exhibits highly varied and unmatched exchange rate
arrangements. Furthermore, region‟s exchange rate regimes and policies are highly
disarrayed. Before the Asian financial crisis 1997-98, most of the East Asian states used to
maintain de jure or de facto regimes (pegged against US Dollar). But the period after this
crisis shows abruptly diverse behavior on part of these economies where they followed a mix
of highly divergent exchange rate regimes. In other words, each state is trying to serve its
own independent and sovereign national monetary policy agendas but this course of action is
highly unwise and lamentable as on the other hand, these economies are profoundly trying to
promote economic cooperation in their region through trade, finance and labor mobility. Thus
the two strategies are highly conflicting as none of the two purposes can be served properly
in these circumstances of exchange rate policy disparity.

66
 Ishaq & Rehman
The two largest economies of East Asia i.e. China and Japan are said to be mainly
responsible for this concern, Oskooee and Hegerty (2012). Japan has its own policy of freely
floating exchange rate and China is standing on the other pole with tightly managed
exchange rate regime. The other key economies of the region have their own distinct policy
of intermediate exchange rate regimes like managed float where US dollar is serving as the
main anchor currency. This mismatch of exchange rate strategies has made the region to
face numerous events of exchange rate volatility led by hostile capital movements. The most
recent example is of Korean Won which has undergone an intense depreciation immediately
after the global financial crisis of 2008. Nevertheless, since last decade, a clear cut trend
emerges where majority of the South East Asian states‟ monetary authorities have heavily
shadowed Renminbi as China‟s exchange rate has floated sharply against U.S.D. This is
much evident from the cumulative exchange rate movements since mid 2005 and the
weights associated with the implicit currency basket of these state, Henning (2012).

This study aims at exploring whether the exchange rate misalignment in any country is
contributed more from the side of its regional trading partners or ex-regional trading partners.
Thus the findings will lead towards a conclusion that is if the exchange rate disagreements
are more harmful amongst regional trading partners or not.

What will be the gains from this type of exploratory study? Indeed, Regional Trade
Agreements are not damaging themselves, but the problem arises when due to occurrence
of currency crisis in its regional trading partners, a state losses its export market and try to
search for some other good alternative, Horvath and Grabowski (1997). Due to preferential
access under regional trade agreements, it is customary that a country may start exporting
those goods and services to its regional trading partners in which she actually does not hold
any competitiveness in international market, Hamanaka (2012). So, currency depreciation in
any of its regional trading country will make it very difficult for her to search alternative
markets outside the region. In short, our basic hypothesis is to see whether exchange rate
disagreements are more costly for a country if they are with regional trade members besides
ex-regional members.

The rest of the paper is organized as follows: Section 2 discusses the theoretical and
empirical literature important in the discussions of monetary policy disarray and regional
economic coherence. Section 3 highlights the rationale of this study with the primary focus on
its improvement over previous works in terms of conducting an advanced empirical analysis.
Section 4 lays out foundations for a methodological framework for empirically testing the
relative importance of regional and non-regional exchange rates and their respective
volatilities for regional exports. Data and its sources are given in Section 5. The analytical
framework designed in Section 4 is empirically tested in Section 5 through long run
cointegration approaches and short run innovation accounting technique. And finally, Section
6 and 7 concludes the study and highlights its main limitations respectively.

2. Review of Literature
This study is distinctive in a sense that the role of exchange rate policy mismatch in the
context of regional economic cooperation is least talked about earlier. Due to the same
reason, it is expected to give some worthwhile contribution to the existing research on

67
 Ishaq & Rehman
preferential trade agreements in an environment of divergent intra-regional monetary and
exchange rate policies.

Why the absence of compatibility in exchange rate strategies is highly detrimental for all
regional participants, one possible argument in this context can be about those protectionist
pressures which are brought into effect in the periods of currency crisis and they supersede
all the plausible gains of regional integration agreements, Gregoriao and Wolf (1994).

Tracing back the historical background of the idea of monetary unification, Mundell (1961)
was the first person who favored the notion of regional monetary and exchange rate union in
the form of Optimum Currency Area. Stressing the idea of sharing external macroeconomic
shocks, Mundell said that a monetary union is a high way to establish an insurance system
where asymmetric economic shocks can be better combated as compared to a system of
national currency with high degree of uncertainty associated with individual exchange rate
regime. In presence of such exchange rate settings, cross country capital mobility will be
indeed facilitated thereby lessen the blow of adjustment to external macroeconomic shocks.
Secondly, contradicting Balassa-Samuelson theory of productivity-exchange rate pass-
through, Mundell claimed that exchange rate stability is less to do with other macroeconomic
fundamentals like cross sector and cross country productivity and inflation differentials.
Exchange rates are more likely to driven by psychological and „good and bad news‟ factors
thus giving rise to greater volatility. So, in a way, exchange rate movements can be taken as
asymmetric shocks themselves. Hence, discarding the instrument of independent national
exchange rate policy will be more advantageous which are likely to overweigh the associated
costs of such a practice.

In the context of establishment of European Monetary Union, Bayoumi and Eichengreen

(1994) said that the biggest probable rationale for forming this biggest monetary union of the
world is to combat these undesirably adopted resistive protectionist measures. For European
region, there are number of incidence of protectionist pressures that led to momentous
exchange rate swings. The most memorable one is that became cause of September 1992
EMS crisis and strained the economic linkages of EU member states. The basic stimulus
behind this tension was the relocation of those several production plants from Italy, France
and Portugal to United Kingdom which caused the Sterling to depreciate sharply at later
stages. The relationships got more stressed in June 1995 when at the EU summit organized
in Cannes, Jacques Chirac, President of France, publically complained that the devaluation
of Italian Lira has given a historic set back to the France‟s conventional exports to Italy.

Porto and Canuto (2002) analyzed the effects of custom union of Mercosul across various
industries of Brazilian region. Being strongly characterized by exchange rate heterogenity,
the authors revealed quite astonishing results as Mercosul has brought a trade bias from the
period 1990-98. They found that the only beneficiary are the Southern region whereas
Northern region are found to be benefited less. In 2000, the entire region saw a trade
uniformity, still the trade biases are not reverted yet to levels prevailing before the creation of
that custom union.

Goto and Hamada (1992) analyzed the establishment of a common currency area amongst
emerging Asian economies. Upon analyzing several economic indicators of the region, the

68
 Ishaq & Rehman
author suggests that in terms of synchronized real business cycles, their openness and
degree of factor mobility (labor and capital) the region is highly suitable for developing a
monetary and exchange rate union. Such an arrangement is likely to bring long term
economic benefits to the region.

Inter-American Development Bank conducted a very comprehensive exploratory study

(2002) in order to explore the role of disagreements in exchange rate policies amongst
member states of a region in the perspective of their participation in regional trade
agreements. Using annual data from 1989-2000 for a cross section of 37 countries, a mix of
member states from six various regional integration agreements, the study shows that there
are prominent and significantly large differences in the impacts of regional real exchange rate
misalignments and non-regional ones with respect to trade volumes of a country. Using panel
regression analysis with two way effects, an attempt was made to see the individual effects
of misalignments originating from regional and non-regional trading members on total trade
volumes of a country. Afterwards, same data is split into two distinct categories, developed
and developing countries and the same hypothesis is tested. The results once again favored
the same notion that exchange rate misalignments arising from regional trading members are
more crucial in determining the volumes of trade of a country in comparison to non-regional
trading partners.

Claiming exchange rate coordination vital in this era of growing globalization and economic
interdependence, Gupta (2012) attempted to estimate the degree of exchange rate harmony
among Asian economies. Degree of convergence amongst various currencies is tested by
unit root test with the help of Asian Currency Unit formulate as a basket of currencies of 15
East and South East Asian economies. Significant divergence is found in exchange rate
behaviors of all these participant countries actually driven by adoption of variant exchange
rate mechanisms. However, future exchange rate and monetary coordination, which is
inevitable for ensuring sound working of deepening Asian intra-regional linkages can be
facilitated through multilaterlized swap structures, a well-organized system for regional
economic surveillance and development of stable and reliable Asian bond market.

In the context of regional economic development in ASEAN +3 territory, Achsani and

Partisiwi (2010) regard regional spillover effects and externalities due to financial
interdependence as the most obvious rationale for regional economic macroeconomic
coordination. For regional member states, being highly integrated with each other,
coordination process may take efforts towards formulating a prescribed mechanism for
ensuring intra-regional exchange rate policy consistency, unconventional provisions for
smooth and stable exchange rates or the member states may form a common currency
union, Bayoumi and Eichengreen (1994). Need is to prevent competitive depreciation at the
regional levels which can only be done through developing a more coherent framework of
intra-regional exchange rate policies.

3. Rationale of the Study

One of the primary objectives of this study is to empirically analyze the importance of
monetary unification amongst ASEAN economies. In earlier studies on this issue, the basic

69
 Ishaq & Rehman
limitation has been the lack of empirical evidence, though sufficient theoretical and practical
aspects are discussed.

Branson and Healy (2005) in their paper constructed a currency basket and Real Effective
Exchange Rate (REER) consisting of ASEAN+1 member states‟ currencies in order to show
their co-movements across the region. This practice must be acknowledged as a good
display of exchange rate policy coherency. However, the genuineness of such a belief could
be better supported by some empirical analysis. Similarly, in the context of growing economic
interference amongst Asian economies, Gupta (2012) tried to estimate the degree of
uniformity amongst South and North East Asian economies‟ exchange rate policies. The
author has produced a commendable piece of research by showing a widening deviation in
exchange rate movements of subject economies. This can be attributed towards their highly
divergent exchange rate regimes serving their own independent national macroeconomic
objectives. But the analysis is insufficiently serving the aim of empirically exploring long run
trend movements of subject economies‟ real exchange rate, once their exchange rate regime
are established in accordance with various macroeconomic institutional arrangements like
enhanced territorial trade integration, adoption of more flexibility against external currencies
and enhanced stability within regional currencies, etc.

This study is an improvement over previous work in a way as the author have used latest
empirical techniques to estimates both the short run and the long run trend deviations in
Asian economies, as imparted by both regional and ex-regional trading countries. Such an
analysis gives an insight into the statistical significance and the magnitude of volatility
brought by regional trading partners in comparison to non-regional trading partners. Besides,
this also reveals the existence and the strength of long run association between REER and
its determinants.

4. Methodological Framework
Nominal exchange rate of a country is the value of some country‟s currency in terms of some
other currency. This value may be determined by a number of factors like interest rates, rate
of inflation, growth in money supply and the future anticipated behavior of all these variables.
On contrary, the real exchange rate as the counter part of nominal exchange rate is the ratio
of price levels between two economies. Calculating the real exchange rate and its movement
towards medium and long term equilibrium takes one to decide about the appreciating or
depreciating behavior of a country‟s real exchange rate. Besides, it also highlights the
expected developments in real as well as nominal exchange rate of a country in line with
domestic price trends in comparison to rest of the world.

With the increasing participation in global economic networks and growing economic
linkages, it is preferable to measure and represent real exchange rate in form of a multilateral
index. Such an index is called Real Effective Exchange Rate (REER) Index incorporating the
bilateral exchange rate of some country with its most prominent trading partners, relative
weightage of each trading partner into the trade of subject country and specific price indices
(CPIs or WPIs).

70
 Ishaq & Rehman
Real Effective Exchange Rate can be measured as the geometrically weighted index of price
level of global economy in comparison to price level at home.

∑ . ⁄

represents the weight attached to each trading partner according to its participation in
total trade volume, is the nominal exchange rate against each trading partner, is the
price index of each foreign competitor and is the price index at home.

The idea of exchange rate unification can be traced back from Purchasing Power Parity
(PPP) theory. PPP theory is one of the basic building blocks in modeling the modern theories
of exchange rate determination. This theory was originally developed by Cassel (1918,
1920), which explains that under the conditions of free trade, the nominal exchange rate
between two countries is equal to the ratio of the two countries‟ price levels. This approach
assumes that equilibrium real exchange rate remains constant over time while the nominal
exchange rate movements tend to offset the relative price movements. The PPP hypothesis
postulates that exchange rates adjust to price differentials in the open economies to restore
international commodity market equilibrium Enders and Hurn (1994).

According to Mundell, the exchange rates are more likely to driven by psychological and
„good and bad news‟ factors, thus giving rise to greater volatility. So, in a way, exchange rate
movements can be taken as asymmetric shocks themselves. Hence, discarding the
instrument of independent national exchange rate policy will be more advantageous which
are likely to overweigh the associated costs of such a practice.

As recommended by Bergara et al (1995), the goods and services for which it is hard to
relocate them to some other ex-regional market, they will be considered in this study as
„Regional Goods‟. If due to regional integration agreements (RIAs) the importance of
regional goods is increased, then it is anticipated that the elasticity of total exports with
reference to exchange rate misalignments will be considerably higher, provided the sources
of misalignment lie within RIA region. To empirically test this hypothesis, we have taken the
model developed by Baer.et al (2002), where each participant economy‟s total exports
depend on its real effective exchange rate misalignment from equilibrium levels plus a set of
other variables.

i ------------------- (1)

Where

= Total real exports of the country i at time t

= Real effective exchange rate of each country at time t
= Real effective exchange rate misalignments of each country from its
Long run equilibrium at time t
= Total real GDP of each country at time t

71
 Ishaq & Rehman
Starting with the definition of long run equilibrium real exchange rate, it is defined by, Nurkse
(1945) as that value of long run real exchange rate which is consistent with the dual
objectives of internal and external balance of the economy, besides taking into consideration
the specified values of those other macroeconomic variables that may influence these
objectives. By internal balance we mean a situation where the current account deficit of a
country can be well-financed by a sustainable level of capital inflows, Clarck and MacDonald
(1998). On the other hand, external balance stands for a situation where the market for non-
tradables is in a sustainable equilibrium, Hinkle & Montiel (1999).

As the real exchange rate misalignment for each economy is defined as divergence of REER
from its long run equilibrium levels, in this study, these equilibrium levels of REER are
calculated by employing Hodrick-Prescott filter. It is one of the best known and most widely
used de-trending methods by macroeconomists. The method is preferred because we can
transform the filter into the frequency domain and understand its effects on various cycles
that make up the time series.

The most admirable characteristic of the H-P filter is that it makes REER stationary over a
wide range of smoothing values, Hodrick & Prescott (1997), and permit the trend to change
eventually, thus making it preferable over other potential REER estimation techniques of
linear method, band-pass filter and the unobserved components model. Besides, it is flexible
and very effortless to execute, thus providing a very strong rationale for its ample application
in literature.

∑ ∑

This is the weighted average of the difference between actual and potential REER and the
rate of change of trend REER. Here μλ is the weighting factor being an exogenous de-
trending parameter and is determined randomly. Hodrick and Prescott proposed to establish
λ at 1600 for quarterly data and 100 for annual data.

Here it is assumed that exports elasticity with respect to exchange rate misalignments has
nothing to do with the sources of misalignment, no matter they lie within or outside RIA
region. It is anticipated that if the equation (1) holds true then real exchange rate
misalignments must be statistically significant with respect to exports i.e. appreciation in real
effective exchange rate must lead towards decline in export volume of the country and vice
versa, Yusoff (2010).

In order to see whether theses elasticities vary or not, real effective exchange rate and its
misalignment (REERVOL) is decomposed into two components: regional component that is
about the RIA member countries and the non-regional component that is about non-RIA
member countries.

-------------- (2)
-------------- (3)

72
 Ishaq & Rehman
----------------------- (4)
--------- (5)

Where

= weight associated to RIA trading partner on behalf of its share in total trade in country
i..

= weight associated to non RIA trading partner on behalf of its share in total
trade in country i.

= Contribution in total real effective exchange rate misalignment made

by RIA member country i at time t.

= Contribution in total real effective exchange rate

misalignment made non RIA member country i at time t.

Substituting equation (2) and (3) into equation (1) the basic model will be transformed as:

( )
---------- (6)

The main hypothesis to be tested is whether γ > that is the major portions of real effective
exchange rate misalignments are arising more from the side of regional or non-regional
trading partners. In other words, anticipate about equation (6) is made that the contribution
made by regional partners in real exchange rate misalignment must bring about a
quantitatively larger fall in exports in comparison to non-regional trading partners.

5. Data and its Sources

We have employed yearly time series data for the estimation of Equation (1) ranging from
1980-2011. However, Equation (2) - (6) are estimated through quarterly time series for the
same period in order to capture the REER volatilities more realistically. The data is collected
from multiple sourcesv like from World Development Indicators (WDI), International Financial
Statistics (IFS) and Asian Development Bank (ADB), United Nations Conference of Trade
and Development (UNCTAD) and a number of national sources including Department of
Statistics Malaysia, Singapore Economic Survey, Bureau of Trade and Economic Indices
Thailand, Bureau of Statistics Japan, Philippine Statistics Authority.

6. Empirical Results
In order to look for long run dynamics of the models proposed in equation (1) and (6) single
equation cointegration techniques are employed. There are several tests of cointegration
found in the literature. The two widely used cointegration techniques are the Engle Granger
(1987) two step residual-based method and Autoregressive Distributive Lag Model (ARDL).

73
 Ishaq & Rehman
In order to see whether each participant series is having unit root or not, DF-GLS unit root
test. DF-GLS unit root test was introduced by Eliot, Rothenberg and Stock (ERS) in 1996 as
an improvement over DF type unit root testing procedures. Using a generalized least squares
rationale, this test has the best overall capability in terms of small sample size and power
when an unknown mean or trend is present in the series.

The results obtained against DF-GLS test are more in favor of non-stationarity; however
some of the series come up with week evidence of unit root. Because the consequences of
treating a non-stationary series as if it was stationary are more serious than the loss of
efficiency that would result from treating a stationary series as if it was non-stationary, we
treat all variables as being non-stationary, Thomas and King (2008).

Following the footsteps of Chinn (2000) and Thomas and King (2008), Zivot single-equation
ECM approach is adopted, in comparison to a system approach (i.e. the Johansen
procedure) which strictly requires the development of a full system of coherent and properly
specified equations.

Following Zivot (1994), the existence of a valid cointegrating vector for equation (1) and (6) is
assessed by testing the significance of λ in the following regression:

ii( ) ∑ ∑

Where X = [REER, REERVOL and REALGDP] for equation (1) and [REGREER, REGVOL,
NONREGREER, NONREGVOL and REALGDP] for equation (6). and are the estimated
least-squares coefficients obtained by first regressing exports on X.

Results for equation (1) are reported in Table 2. A statistically significant and negative
coefficient of error correction term is found in case of Japan, Malaysia, Philippines and
Singapore. However, China is an exception because it has a positive and statistically
insignificant error correction coefficient. China may be an outlier in the sample due to its
economic structure and exchange rate policy. iii The positive coefficient of REER
commensurate well with the theory in case of Japan, Malaysia and Philippines. It can be
interpreted as one unit appreciation (depreciation) of real effective exchange rate leads to
decrease (increase) in exports, ranging from 0.09 to 1.58 units. Similarly, real GDP is holding
a theoretically correct (positive) sign for all the countries except Singapore.

In general, we found that a significant positive long run relationship is found between the real
effective exchange rate and exports. These results are in line with the theory and the
previously done studies, Asseery & Peel (1997) ; Baera et al. (2002). Furthermore, the other
explanatory variables and error correction coefficients are significant besides bearing
desirable signs.

Inclusion of volatility series in our earlier estimations enable us to reach more precise results.
This importance of exchange rate volatility while calculating the impact of real exchange on
exports has been well acknowledged in quite a few studies, Chit et al. (2010). From Table 3,

74
 Ishaq & Rehman
our results suggest that in the error correction framework, real exchange rate volatility has a
significant impact on exports which is significant for all the countries expect Singapore.
Theoretically, the exchange-rate volatility can have an impact on trade in either positive or
negative directions, Bailey (1987) and McKenzie (1999).

In order to see the robustness of results obtained from Zivot ECM, we have employed ARDL
bound testing approach. This model will serve two purposes. As discussed earlier, some of
the series included in model are with weak levels of unit roots. A time series if I (0) or I(1) are
considerably different from each other. Being I(0), the series will be having zero mean with
finite variance. So any shock will temporarily affect the value of this series. On contrary, if it is
I(1), it will be having infinite variance as t goes to infinity. Thus any shock will have
permanents impacts on this series as it is the sum of all previous changes. Theoratically
spaeking, the finite variance of an I(0) vector comes due to the presence of high frequencies
or short run partof the participant series whereas if it is I(1) the infinite variance can be well
explained due to the contribution of low frequencies or long run part of the time series
involved, Engle & Granger (1987) and Boswijk (1994) in this context states that “cointegration
entails the existence of stable relationships that are individually integrated of order one i.e.
time series whose univariate models have a unit root”

In this paper, we are confined to unrestricted ECM and F- test onlyiv. The equation of a
general unrestricted model can be given as:
∑ ∑ ---------- (7)

∑ ∑ --------------- (8)v

The coefficients and both from equation (7) and (8) will be tested via Wald Coefficients
against the null hypothesis of = =0 and the calculated F-statistics are compared to the
critical bounds to explore the existence of a cointegration relationship. The exact critical
bound values for F-test are not existing for an arbitrary mixture of I(0) and I(1) variables.
However, Pesaran et al. (2001) provide bounds on the critical values for an
asymptotic distribution of the F-statistic. For a variety of situations (e.g. different numbers of
variables), they supply lower and upper bound critical values. The lower bound assumes that
all the variables are I(0), while the upper bound considers that all the variables are I(1).
Actually, the precision may be lying somewhere in between these two upper and lower
bounds.

If the computed F-value falls below the lower bound we arrive at the conclusion that there is
no cointegration. If the F-statistic exceeds the upper bound, we conclude that there is a
cointegration relationship present. If the F-statistic falls between the upper and lower bounds,
the test is said to be inconclusive.

For Equation 1, the results are given in Table 1. Results show that the calculated F-statistics
is 5.087967 which is higher than the upper bounds critical values at all levels of significance
in case of China unlike Zivot ECM findings. Thus, there is a Co-integration relationship
amongst the variables in case of China. Similarly, the cointegration exists in case of
Philippines at all levels as the calculated F-stat of 5.124644 is sufficiently greater than its
upper critical bounds. The null hypothesis of no Co-integration is rejected for Japan and

75
 Ishaq & Rehman
Singapore also. Similarly, in equation 2, all the calculated F-statistics (reported in Table 2)
strongly reject the null of no cointegration. So there is strong evidence about the existence of
long run equilibrium relationship among the variables.

76
 Ishaq & Rehman

Table 1
Long-Run Elasticity Estimates For Equation 1 Obtained From Zivot ECM
Country λ REER REERVOL REALGDP
0.4531
China
(0.6088) - - -
-0.3238 2.1443 1.0441 3.3877
Japan
(0.0493) [8.2384] [6.3996] [15.0161]
-0.7241 1.4720 0.8766 1.1391
Malaysia
(0.0587) [7.4322] [3.6320] [14.8806]
-0.1471 1.1136 0.0307 0.8430
Philippines
(0.0388) [2.5875] [1.9403] [4.0655]
-0.2722 -3.9836 -1.6545 -0.0647
Singaporevi (0.0379) [-1.3075] [-1.4722] [0.3222]

F-Test For Equation 1 Obtained From ARDL Unrestricted ECM

Country Order of F-Statistics 50% C.V.Bs 75% C.V.Bs 90% C.V.Bs

Variables I(0) I(1) I(0) I(1) I(0) I(1)
China FREER 5.087967 2.86 4.01 3.25 4.49 3.74 5.06
Japan FREER 4.408592
Malaysia FREALGDP 3.437355
Philippines FREER 5.124644
Singapore FEXP 4.279261

77
 Ishaq & Rehman

Table 2
Long-Run Elasticity Estimates For Equation 2 Obtained From Zivot ECM
Country λ REGEXRATE NONREGEXRATE REGVOL NONREGVOL REALGDP
China -0.5563 0.4822 2.7544 0.0199 0.0397 1.9038
(0.0974) [0.7430] [1.5497] [0.0467] [0.0865] [9.8775]
Japan -0.8737 1.0987 0.0543 0.8970 1.4321
(0.0313) [2.2875] [5.8744] [2.8663] [7.0971]
Malaysia -0.7371 4.6631 -0.0133 5.9854 -0.4383 2.1432
(0.0053) [1.8753] [-0.7657] [3.4328] [-0.8771] [12.4432]
Philippines -0.7955 3.6504 1.9648 -6.9073 4.1906 -0.2341
(0.1053) [7.0519] [7.9866] [-6.0711] [6.7754] [-2.0977]
Singapore -0.6739 -1.4876 0.5439 2.9647 -1.4760 1.0876
(0.7877) [-11.9432] [1.7654] [14.0688] [-0.9877] [3.9995]
Indonesia -0.4006 0.8439 -0.4532 -2.3549 1.6549 1.2764
(0.0711) [1.9649] [-1.8465] [-2.0754] [1.1443] [5.0639]
Thailand -0.8022 1.1174 0.8544 -2.4388 1.9876 1.8654
(-0.6449) [2.8549] [3.0541] [-1.9225] [1.9965] [3.7439]

NOTE: λ is the ECM Coefficient. The p-values are given in parenthesis and t-ratios are written in square brackets

F-Test For Equation 2 Obtained From ARDL Unrestricted ECM

Country Order of F-Statistics 50% C.V.Bs 75% C.V.Bs 90% C.V.Bs

Variables I(0) I(1) I(0) I(1) I(0) I(1)
China FEXP 5.718607 2.45 3.61 2.75 3.99 3.15 4.43
Japan FRVOL 3.644036
Malaysia FEXP 15.75747
Philippines FEXP 5.607604
Singapore FEXP 5.607604
Indonesia FNRXRATE 9.661050
Thailand FREALGDP 10.00103

78
 Ishaq & Rehman

6.1 Short Run Innovation Accounting: Variance Decomposition Test

In order to account for the short run innovation accounting and for supplementary
conclusions, variance decomposition test (VDC) and impulse response function (IRF) are the
most suggested tools. VDC and IRF serve the purpose of assessing the dynamic
connections and magnitude of causal relations among system variables. The forecast errors
of VDC will reveal the percentage of REER series forecast error variance credited to their
own shocks versus the shocks from the REER of other competitor states. Thus, VDC is a
short run tool to calculate the comparative significance of two types of REER volatilities in
explaining fluctuations in home REER.

Table 3 states the variance decomposition of forecast errors of exports in percentage form at
four, ten and twenty quarters forecast horizon for participant economies. It is clearly evident
from the statistics that except for China, regional REER misalignments are noticeably
dominant in comparison to their non-regional counter parts. As for regional REER, there‟s a
mix of evidence as half of the countries are representing non-regional REER more influential
with respected to export volume. However, the figure for twentieth horizon clearly depicts that
for all the East Asian participant states, the importance of regional members‟ exchange rate
and its misalignment is getting more crucial than the non-regional members. This calls for an
immediate policy action to ensure region wide monetary and exchange rate policy
coherence, so that East Asian exports can better prevented from undesirable volatilities
caused by exchange rate shocks arising from neighbor states.

79
 Ishaq & Rehman

Table 3
Variance Decomposition (% Of Forecast Error Variance Of Exports)
Countries Quarters EXP REGREER REGVOL NONREGREER NONREGVOL GDP

Quarters Relative Variance of Chinese Exports

China 4 95.19 0.85 0.81 2.87 0.24 0.03
10 90.55 1.67 1.60 5.64 0.48 0.02
20 88.63 2.01 1.93 6.79 0.58 0.03
Quarters Relative Variance of Indonesian Exports
Indonesia 4 99.99 0.00 0.00 0.00 0.00 0.00
10 99.97 0.00 0.00 0.00 0.00 0.00
20 99.91 0.02 0.01 0.03 0.00 0.00
Quarters Relative Variance of Japanese Exports
Japan 4 99.57 0.01 0.14 0.16 0.05 0.04
10 97.88 0.05 0.72 0.82 0.28 0.21
20 95.41 0.11 1.57 1.79 0.62 0.47
Quarters Relative Variance of Malaysian Exports
Malaysia 4 94.51 0.44 4.21 0.76 0.00 0.05
10 73.50 2.14 20.34 3.71 0.00 2.83
20 49.89 4.06 38.48 7.02 0.00 0.53
Quarters Relative Variance of Philippines Exports
Philippines 4 99.97 0.00 0.00 0.00 0.01 0.01
10 99.89 0.00 0.02 0.00 0.03 0.03
20 99.78 0.00 0.04 0.01 0.07 0.07
Quarters Relative Variance of Singapore Exports
Singapore 4 99.08 0.50 0.03 0.06 0.01 0.28
10 87.34 7.06 0.51 0.91 0.24 3.91
20 47.74 29.16 2.13 3.78 1.01 16.15
Quarters Relative Variance of Thailand Exports
Thailand 4 99.55 0.17 0.05 0.10 0.03 0.07
10 97.03 1.17 0.38 0.70 0.22 0.47
20 90.97 3.58 1.16 2.14 0.69 1.44

80
 Ishaq & Rehman

7. Conclusion
The paper analyzed the probable impacts of exchange rate dissension among the ASEAN +2
member states on total export volumes of the region. By taking regional and ex-regional real
exchange rate misalignments in account, the authors have made an attempt to draw
attention to the growing role of intra-regional factors which may once again make the region
to see another episode of currency crisis, like the one of 1997-99 financial meltdown of East
Asia. The estimated impact of real exchange rate misalignment originating from the member
economies of ASEAN+2 region is much more pronounced than the impacts arising from ex-
regional trading partners of the region. The fact leads towards a strong support for the
hypothesis that within-RIA exchange rate misalignments are more critical in the perspective
of currency crisis in a region in comparison to the same misalignments arising somewhere
else.

Regardless of the need of intra-regionally harmonized exchange rate mechanism, at present

there are no such prominent arrangements for monetary and exchange rate policy
coordination as each East Asian state wants to pursue its own independent macroeconomic
goals. In the last thirteen years the region has experienced a significant economic integration
through the channels of trade and finance and is anticipated to appear as the largest
economic block in the years to come. This calls for a dire need of stable and well-coordinated
exchange rate framework for the region. Such a highly integrated market driven
regionalization needs much more assurance than before that national macroeconomic
objectives of each East Asian state would not be damaged by exchange rate misalignments
of other regional member states. Keeping in mind the Asian financial crisis of 1997, to save
the region from experiencing any such economic catastrophe once again, the entire region
will have to surrender their individual economic interests and their sovereign economic
agendas and will have to work towards developing a more organized and coherent monetary
and exchange rate policy structure. This does not only serve their personal national interests
but also the role of East Asia as a regional economic entity will be more distinct and reliable
in the entire continent.

8. Limitations of the Study

This study could be considered as significant contribution in already done empirical works on
regional exchange rate policy coordination. The authors have made an extensive use of
modern econometric techniques in order to better analyze the situation in East Asian
economies. However, for the purpose of brevity, many of the institutional and policy
measures which may better support the agenda of monetary unification in the region are not
discussed. Improved trade integration, some kind of operational mechanism to combat the
speculative currency attacks like Chiang Mail Initiative (CMI) or ASEAN Swap Arrangement
(ASA), effective and efficient working of Asian Bond market, etc. can be some of the
examples whose practical significance needs to tested empirically in order to investigate their
effectiveness to ensure exchange rate policy coherency in East Asian territory. Hence, there
is much room to conduct empirical research on this dimension in order to come up with some
practical solutions to ensure enhanced monetary and exchange rate policy uniformity
amongst the major players of East Asian region.

81
 Ishaq & Rehman

Endnotes

ii
is calculated through recursive approach to attain more efficient coefficients. Initially two lags are considered
on part of each variable. By employing general to specific approach, those lagged variables have been omitted
which are clearly insignificant (not significant up to 20%).
iii
China has been devaluing its currency
iv
Only F-test results are reported in case of ARDL. Due to some chance of weakly non-stationary series in the
model, we rely on the F-statistics of ARDL Unrestricted ECM to make the decision about the existence of a
cointegrating vector.
v
Equation 8 has empirically been tested by altering the order of variables and the obtained results are
consistent with a few exceptions. Detailed results can be requested from authors.

References
Achsani, NA. & Partisiwi, T 2010, 'Testing the Feasibility of Asean+3 Single Currency
Comparing Optimum Currency Area and Clusturing Approach', International Research
Journal of Finance and Economics, vol. 37, pp. 69-75.
Asseery, A & Peel, DA 1991, 'The effects of exchange rate volatility on exports : Some new
estimates', Economic Letters , vol. 37, no. 2, pp. 173-77.
B, YM 2010, 'Bilateral Trade Balance, Exchange Rates, and Income: Evidence from
Malaysia', Global Economy Journal , vol. 9, no. 4, pp. 213-229.
Baera, W, Cavalcantib, T & Silva, P 2002 'Economic integration without policy coordination:
the case of Mercosur', Emerging Markets Review, vol. 2, pp. 269-291.
Bailey, M J 1987, 'The impact of exchange-rate volatility on export growth: Some theoretical
considerations and empirical results', Journal of Policy Modeling, vol. 9, no. 1, pp. 225-
243.
Bayoumi, T & Mauro,P 2001, 'The Suitability of ASEAN for a Regional Currency Agreement',
World Economy , vol. 24, no. 3, pp. 933-954.
Bayoumi, T & Eichengreen, B 1994, 'One Money or Many? Analyzing the Prospects for
Monetary Unification in Various Parts of the World' Princeton Studies in Internationall
Finance.
Boswijk, HP 1994, 'Testing for Unstable Root in Conditional and Structural Error Correction
Models' Journal of Econometrics , vol. 63, pp. 37-60.
Bergara, M, Dominioni, D & Licandro, JA 1995, 'Un modelo para
comprender la Enfermedad Uruguaya', Banco Central del Uruguay, Revista de Economía,
vol. 2, no. 2, pp. 39-75.
Branson, WH & Healy, CN 2005, 'Monetary and exchange rate policy coordination in
ASEAN+1'. National Bureau of Economic Research, Working Paper 1173.
Cassel, G 1918, 'Abnormal Deviations in International Exchanges' Economic Journal, vol. 28,
pp. 413-415.
Chinn, MD 2000, 'The Usual Suspects? Productivity and Demand Shocks and Asia-Pacific
Real Exchange Rates', Review of International Economics, vol. 8, no.1, pp. 20-43.
Clarck, PB & MacDonald, R 1998, 'Exchange Rate and Economic Fundamentals: A
Methodologiacl Comparison between BEERs and FEERs', International Monetary Fund,
Working Paper Series, 67 .
Elliott, G, Rothenberg, TJ & Stock, JH 1996, 'Efficient Tests for an Autoregressive Unit Root',
Econometricia , vol. 64, no. 4, pp. 813-836.

82
 Ishaq & Rehman

Enders, W & Hurn, S 1994, 'Theory and Tests of Generalized Purchasing-Power Parity:
Common Trends and Real Exchange Rates in the Pacific Rim' Review of International
Economics, vol 2, no. 2, pp. 179-190.
Engle, RF & Granger, CW 1987, 'Co-Integration and Error Correction: Representation,
Estimation, and Testing', Econometricia, vol. 55,no. 2, pp. 251-267.
Gregoriao, JD & Wolf, HC 1994, 'Terms of Trade, Productivity and Real Exchange Rate',
Nantional Bureau of Economic Research, Working Paper Series, 4807 .
Gupta, SA 2012, 'Exchange Rate Coordination in Asia: Evidence using the Asian Currency
Unit', Asian Development Bank Institute, Working Paper Series, 356.
Hamanaka, S 2012, 'Trends of Trade Interdependence and Prolifiration of FTAs in Asia',
Global Economy Journal , vol. 12 no. 2, pp. 164-183.
Henning, CR 2012, 'Choice and Coercion in East Asian Exchange Rate Regime. Ptereson
Institute for International Economics, Working Paper Series, 4854 .
Hinkle, LE & Montiel, PJ 1999, 'Exchange Rate Misalignment: Comcept nad Measurement
for Developing Countries', 1st edn, Oxford University Press, New York.
Hodrick, RJ & Prescott, EC 1997, 'Postwar U.S. Business Cycles: An Empirical Investigation',
Journal of Money, Credit and Banking , vol. 29, no. 1, pp. 1-16.
Horvath, J & Grabowski, R 1997, 'Prospects of African Integration in Light of the Theory of
Optimum Currency Areas', Journal of Economic Integration , vol. 1, no. 12, pp. 1-25.
Ishaq, M & Atiq ur Rehman, M 2013, 'Surmounting the Individual: Establishing a Common
Currency in Asia- A Case Study East Asian Economies', Global Economy Journal, vol.
13. no. 1, pp. 63-88.
Kwack, S & Yeung 2004, 'An Optimum Currency Area in East Asia: Feasibility, Coordination,
and Leadership Role', Journal of Asian Economics , vol. 15, no. 1, pp. 153-169.
Mundell, AR 1961, 'A Tehory of Optimum Currency Areas', The American Economic Review,
vol. 4, no. 51, pp. 657-665.
Nurkse, R 1945, 'Conditions of International Monetary Equlibrium. In R. Nurkse, Essays in
International Finance 4 (Spring) Princeton University Press, Princeton, New Jersey.
Oskooee, MB & Hegerty, SW 2012, 'Exchange Rate Volatility and Industry Trade between
Japan and China', Global Economy Journal , vol. 12, no. 3, pp. 1-21.
Pesaran, HM, Shin, Y & Smith, RJ 2001), 'Bounds Testing Approaches to the Analysis of
Level Relationships', Journal of Applied Econometrics, vol. 16, no. 3, pp. 289-326.
Porto & Canuto 2002, 'Mercosul: Gains from Regional Integration and Exchange Rate
Regimes', Paper Presented at Latin American Meeting of Econometric Society.
Taguchi, H 2010, 'Feasibility of Currency Unions in Asia- An Assessment Using Generalized
Purchasing Power Parity', Policy Resaerch Institute, Ministery of Finance, Japan Public
Policy Review , vol. 5, no. 6, pp. 859-872.
Thomas, A & King, A 2008, 'The Balassa–Samuelson Hypothesis in the Asia-Pacific Region
Revisited', Review of International Economics, vol. 16, no. 1, pp. 127-141.
Zivot, E 1994, 'Single Equation Conditional Error Correction Model Based Tests for
Cointegration', Discussion Paper 94-12, Department of Economics, University of
Washington .

83