Global growth has been derived significantly from fast increases in trade and
investment among countries in recent decades. In addition, the way the trade
is done has been altered because of emerging markets advents specially China,
expansion in services tradability and change in the type of products which are
traded like shares. Although the recent crises have had undesirable impact on
this growing trend, factors such as emerging markets, income increase, taste
variety increase and world trade integration have stimulated the rapid growth
of the level of trade internationally. This can be seen in quadruple proliferation
of goods traded between 1980 and 2008. However, this increase was not similar
among developing and developed countries. Many developing countries in Asia
regarded as emerging markets have been the fastest growing markets in supply
of goods and services while not much increase has been seen in African, South
American and central European countries. Yet the developed countries have been
maintained to be the pioneers in trade and investment development. In recent
years the emerging markets in Asia have been targeted as one of the main investment
purposes as well as multinationals future investment plans. This will be more
considerable because of the increase in regional liberalization and coordination
among Asian countries such as ASEAN countries (Association of Southeast Asian
Nations) (BIS, 2011).
Although, ASEAN countries have the prominent potentials for investment and
trade purposes, there have been fluctuations in growth percentage and the real
exchange rate movements in recent years.
||Real exchange rate movements from, (a) 2002-2007 and (b) 2000-2007
Above graphs (Fig. 1a-b) show real exchange
rate movements in some of these countries from 2000-2007.
Therefore, it is important to forecast the changes in exchange rate in ASEAN
members for investment and trade purposes because without this estimation there
would be significant losses to the investment and trade in these countries despite
the great potentials and opportunities in them (Grenville,
One of the theories which can be used for exchange rate appreciation or depreciation prediction which is widely used is International Fisher Effect (IFE). This theory illustrates the relationship between exchange, interest and inflation rates of the countries. Based on IFE, the future amount of spot exchange rate can be estimated by the nominal interest rate differential between two countries. The real interest rate is considered to be equal in all countries for more simplification. So, the nominal interest rate differential is affected by the inflation rate differential. Then, the inflation rate differential will affect the future spot exchange rate based on IFE. This study aims the IFE validity via exchange rate movements for ten ASEAN countries currencies relative to Malaysian currency within a period of ten years, from 2002 to 2012.
Overview of ASEAN: ASEAN was established in 8 August 1967 to form an
economic, political organization of Southeast Asian countries including Indonesia,
Malaysia, the Philippines, Singapore and Thailand. From that time onwards, five
more countries (Brunei, Myanmar, Cambodia, Laos and Vietnam) were added to this
membership. The motivations of making this organization are economic growth
enhancement, cultural elaboration, safeguarding the stability in the region
and talking about differences in a calm and peaceful manner. If ASEAN was considered
as a unique commodity, it would have the eighth grade among the largest economies
in the world (EC. Europa.eu, 2010).
However, ASEAN members have faced some challenges for having a proliferated
trade. The main challenge is the competitive competitions which are imposed
by China and India in the market either in the region or in the other parts
of the world. ASEAN countries ought to provide an appropriate political environment
to improve their competition and competitive advantages. This environment must
offer more profitability and productivity for ASEAN members so as to fit themselves
with the changes and needs of the markets. These challenges will be mitigated
by initiating liberalization in trade among ASEAN members via offering facilities
in transportation, logistics, policies and the proper and enhanced usage of
information technology in trade (Intal, 2010).
Empirical evidence: The analysis of International Fisher Effect has
offered different results for validity of this theory. On one hand, some researchers
have shown that IFE can be applicable empirically (Hill,
2004) in declared that IFE exists in the long run while it does not hold
in the short run. Moreover, in the long run inflation rate differential can
be used to predict exchange rate movement, however, there is no exact relationship
between them (Hakkio, 1986). In addition, currency
realignment in European Monetary System is able to have significant influence
on maintaining Purchasing Power Parity theory among European countries (Cheung
et al., 1995). Aliber and Stickney (1975)
stated that IFE can be useful in a long run. This is supported by Kane
and Rosenthal (1982) in about the validity of International Fisher Effect
in long run (Madura, 2012).
On the other hand, Cumby and Obstfeld (1981) illustrated
that IFE is not an appropriate indicator for currency movements in short time
horizon. The profit of more than a half of the transactions resulted from buying
high interest future contracts with discount and selling low interest future
contracts with premium proves the fact that IFE may not come true in reality
(Thomas, 1985). Others including Adler
and Lehmann (1983) as well as Adler and Dumas (1983)
have come to conclusion about lacking significant results for IFE to hold. Also,
Madura and Nosari (1984) designed an investment strategy
which refuted IFE in practice because of the profit this strategy gained by
borrowing from low interest rate currency and investing in high interest rate
However, in one study to test IFE, this theory was valid for Japan while there
was no significant relationship between nominal interest rate differentials
to offset the exchange rate movements in other countries. These mismatches clarify
that other factors such as foreign exchange supply and demand, balance of payments
problems, rising inflation, interest rate, national income, monetary policy,
expectations and speculations have significant influences on currency movements
as well (Khalwaty, 2000).
RESEARCH FRAMEWORK AND HYPOTHESIS
Theoretical framework: The combination of Fisher Effect (FE) and Purchasing Power Parity (PPP) theories will make the International Fisher Effect (IFE). First, FE is used to measure the expected inflation rates for a certain country. Then, exchange rate movement is estimated by using PPP theory.
So, the equation presenting IFE is derived as follows:
where, ef is foreign currency percentage change, rh,t is home country nominal interest rate, rf,t is foreign country nominal interest rate, ih,t is home country inflation rate, if,t is foreign country inflation rate.
Laconically speaking, IFE suggests that future spot exchange rates between two countries can be forecasted by the difference between their nominal interest rates. Based on this theory, investors will gain the same return on their investment in a foreign country with higher nominal interest rate in comparison to their home country. This is because of the foreign exchange rate depreciation occurred due to higher nominal interest rate suggested by IFE.
Hypothesis: The null hypothesis based on IFE hypothesizes the fact that when there is no difference in nominal interest rates between two countries, their exchange rates should not have any alternations. Consequently, the nominal interest rate differentials would be exactly the same as the appreciation or depreciation of the currency for the country with higher nominal interest rate.
MATERIALS AND METHODS
This study illustrates the theory of International Fisher Effect and also describes the description of this theory as comprehensive as possible according to the purpose of this research. Moreover, a general outline of IFE which was depicted can clarify the method employed in statistical analysis.
A statistical test will be used to check the relationship between nominal interest
rates and exchange rate movements. Nominal interest rates of ASEAN members in
a period of past ten years will be found through the data bases. Then, the nominal
interest rate differentials relative to Malaysia as the home country will be
measured in this period. Consequently, exchange rate differentials of the other
nine countries relative to Malaysia will be measured in the same period of time.
These figures will be analyzed by employing a Regression model to identify if
IFE holds or not. The regression method which will be applied is Ordinary Least
Squares (Coppock and Poitras, 2000). OLS will show
if the estimates are the same as what happened in reality. To support, this
regression examines if exchange rate changes can be forecasted be using nominal
interest rate differentials among ASEAN countries and assuming Malaysia as the
home country in the period of time chosen.
So, this research investigates the relationship between annual data of nominal interest rates of the ASEAN countries exchange rates from 2002 to 2012. The exchange rates chosen here are floating rates because this study has assumed that exchange rates fluctuate without governments intervention.
The countries which are scrutinized are ASEAN countries including Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, Philippines, Singapore, Thailand and Vietnam. The home country chosen is Malaysia, therefore the nominal interest rate differentials and exchange rate changes will be measured relative to Malaysian nominal interest rate and Malaysian currency (Ringgit Malaysia, RM), respectively.
Regression model: Based on the EMH theory, exchange rate is moved based on the information effect on it. The foreign currency expected future value at time t+1 with the available information at time t can be shown as:
So, the future spot rate at time t+1 should be equal to Eq. 2:
In case of a discrepancy in the amount of the real future spot rate and expected future spot rate, μt+1 is considered to show the error which may affect Eq. 2 as follows:
By referring to the concept of International Fisher Effect, Φt can be replaced by nominal interest rate differentials.
The following equation shows the regression model:
According to the regression model used, the null hypothesis would be true if:
The null hypothesis estimates that Alfa must equal to zero. This is inferred that exchange rates ought to be constant for two countries provided that they have similar nominal interest rates. Null hypothesis pinpointed amount of 1 for β. This implies that the same percentage change in nominal interest rates have to occur for exchange rate between the home country and foreign country.
Data: The data is collected from the World Bank database for all ASEAN members. This data includes the nominal interest rates, exchange rates and inflation rates for 10 countries in ASEAN community from 2002 until 2012. All the exchange rates are based on US dollar. Having Malaysia as the home country in this research, these currencies values are divided by the Malaysian currency (RM) per USD for each year and then the percentage change of each country currencies are measured relative to RM from 2001 until 2012. In addition, the nominal interest rate differentials are calculated with respect to Malaysia nominal interest rates for the chosen period of time. These nominal interest rates should reflect the amount of depreciation or appreciation of the ASEAN countries currencies relative to RM as proposed by International Fisher Effect Theory.
RESULTS AND DATA ANALYSIS
This study aimed to investigate the empirical validity of International Fisher
Effect among ASEAN members in a period of ten years. This was done by analyzing
the relationship of nominal interest rate differentials and exchange rate changes
between each country and Malaysia as considered as the home country in the long
run. Regression method was used for this analysis and in the following the results
are presented. The null hypothesis was considered to be α = 0 and β
= 1 for the linear regression model. The results of the analysis are summarized
over the Table 1-9.
The extremely low R-squared for some cases has led to rejection of the IFE
to forecast exchange rate movements. The lowest R-squared was for Malaysia-Vietnam
case with only 0.2% for the overall performance of the model. Then, Malaysia-Thailand
regression resulted to solely 2.6% R-squared amount for empirical testing of
IFE in the long run. Malaysia-Myanmar with 3% R-squared also illustrated that
IFE cannot predict Myanmar currency changes.
|| Regression output for Malaysia-Brunei
|| Regression output for Malaysia-Cambodia
|| Regression output for Malaysia-Indonesia
|| Regression output for Malaysia-Laos
|| Regression output for Malaysia-Myanmar
|| Regression output for Malaysia-Philippines
|| Regression output for Malaysia-Singapore
|| Regression output for Malaysia-Thailand
Having R-squared of just above 5% for Malaysia-Cambodia is an evidence for
weak performance of the linear regression model used for this case.
|| Regression output for Malaysia-Vietnam
Malaysia-Philippines R-squared had 6.7% of the total changes to the Philippines
currency caused by nominal interest rate differentials between this country
and Malaysia which is not acceptable to forecast this countrys currency.
Besides, for all of these pairs the null hypothesis was rejected due to not
having acceptable amounts for α and β in the 5% significance zone.
However, R-squared showed higher amounts of 12.2, 23 and 40.6% for Malaysia-Brunei, Malaysia-Laos and Malaysia-Singapore, respectively. Though, the null hypothesis for these paired were rejected as well and the amounts for the constant and coefficient of the regression model for all of them did not fall in the 5% significance level.
On the contrary, Malaysia-Indonesia was the only pair which had a different story. R-squared displayed for this regression was 34.1%. Furthermore, α = 1.258 was significant in 5% acceptance level. This means that if there was no nominal interest rate differential between Malaysia and Indonesia, the Indonesian rupiah would be depreciated by 1.258%. Moreover, β = 0.151 could be considered acceptable in 10% significance level. This renders the fact that with each 1% nominal interest rate differentials between Malaysia and Indonesia, Indonesian currency would be appreciated by 0.151% relative to Malaysian ringgit. Ultimately, International Fisher Effect can be partially accepted for Malaysia-Indonesia case empirically. However, this result shows that exchange rate movements of IDN rupiah/MR was not completely offset by the discrepancy in nominal interest rates between these two countries.
The above mentioned results exhibit the fact that Malaysian investors can make profit or losses in investing in other ASEAN members in comparison to local investment purposes. From these results, it is concluded that there are other factors such as political risk, currency risk, taxes, transaction costs and psychological barriers which have meaningful influences on the changes to these countries future currency spot rates.
In addition, it is difficult to make a decision about the changes to nominal interest rates changes. These changes can be because of either real interest rate differentials or inflation rates expectations. Any alternation of these two components of nominal interest rate has an opposite result. For instance, an increase in nominal interest rate in Indonesia relative to Malaysia because of the higher real interest rate causes Indonesian rupiah to appreciate while if the rise in nominal interest rate in Indonesia is due to high inflation rate expectations relative to Malaysia leads to depreciation of rupiah per each Malaysian ringgit. It is seen that IFE is solely valid partially for Malaysia-Indonesia; however, with further investigations this validation can be inapplicable in other time horizons.
By having a concise glimpse over the aforementioned statistical results, International
Fisher Effect appears to be valid for some countries pairs and only for some
time horizons selected. Hence, the nominal interest rate differentials would
not be applicable in predicting the changes in the relative future currency
changes of countries.
DISCUSSION AND RECOMMENDATIONS
Summary of major findings: The chief purpose of this study is to check the empirical evidence of International Fisher Effect among ASEAN countries from 2002 to 2012. According to the statistical method used as well as revealed results, this theory showed partial significant relationship between nominal interest rate differentials and the exchange rate change only for Malaysia-Indonesia case, whereas the difference in nominal interest rate for other eight ASEAN members with that of Malaysia did not offset the changes of their exchange rates relative to Malaysian Ringgit.
Implications of study: The findings of this study help investors to understand exchange rate behaviors via International Fisher Effect theory. Accordingly, the results clearly underscore that Malaysian investors require conducting in-depth practical scrutiny regarding other influential impacts on exchange rate movements, except nominal interest rates and inflation rates differentials, while deciding to invest in other ASEAN countries.
Limitations of study: IFE has some limitations which must be taken into attention. First, It is empirically proved that the difference between real interest rate and nominal interest rate in every country does not really measures expected inflation rate in that country. Base on this, Fisher Effect can be refuted. Then, inflation is not the only element which causes exchange rate movement in a particular country. Other factors such as government control and level of income are significantly influential in this regard. So, PPP theory is not practically useful in this occasion. Ultimately, when both components of IFE have these limitations, there would be more barriers for employing IFE in practice.
Suggestions for future studies: Although there are some problems in
the methodology employed here, IFE is tested in a specific geographical area
called ASEAN. So there is a possibility that the results will be only true for
this especially chosen zone. This problem can be mitigated if this theory is
tested in the other parts of the world. Another problem is that Malaysia is
selected to be the home country and hence the comparison is done between other
ASEAN countries relative to Malaysia. This causes that whatever analysed here
become particularly applicable for Malaysia. One can analyse this theory by
having other ASEAN countries as home country to settle this problem, however,
these results can also be applicable for the surveyed country. The last factor
which is an obstacle to the validity of the methodology is the time frame chosen.
The results will show if IFE will hold in this time period or not. This problem
will be solved by choosing other periods and reanalysing the data, but what
will be measured here can be used to show the validity of IFE theory only between
2002 and 2012. Therefore the results will vary by choosing different periods
of time for ASEAN countries. Notwithstanding all the suggested ideas, the authors
are going to conduct a comprehensive scrutiny regarding the empirical evidence
of this theory throughout nearly all the countries in the world, considering
Malaysia as the home country over the previous decade commencing from 2000.
By meticulously considering the results and statistical outcomes, it is evident that one cannot simply rely on a few macro-economical figures to predict the exchange rates fluctuations among countries, definitely in ASEAN region. The impact of the other factors; such as currency risks, transaction costs, psychological barriers, political risks and alike must be weighted as well as nominal interest rate differentials to offer a more realistic anticipation of foreign exchange rate movements in the long run.