THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
1
1
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
2
The Impact of Inflation and Real Interest Rates in Saudi
Arabia on the SAR/SDR Exchange Rates
By: Mishaal Abdulaziz Alkhudair
A Dissertation submitted in Partial Fulfillment of the
Requirement of the Bachelor in Finance
College of Business
Effat University
Thesis Advisor: Shabir Hakim
2015
2
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
3
Table of Contents
Abstract ...........................................................................................................................................3
1. Introduction ................................................................................................................................4
2. Literature Review ......................................................................................................................6
2.1 Purchasing Power Parity ................................................................................................................ 6
2.2 Interest rate parity conditions ......................................................................................................... 8
2.3 Other factors influencing the exchange rate ................................................................................... 9
2.4 Factors influencing Parity conditions ........................................................................................... 12
3. Data and Methodology ............................................................................................................13
3.1 The Data ....................................................................................................................................... 13
3.2 Models .......................................................................................................................................... 14
3.3 Descriptive Statistics .................................................................................................................... 15
3.4 Model Testing .............................................................................................................................. 20
3.5 Hypothesis Testing ....................................................................................................................... 21
4. Findings.....................................................................................................................................23
5. Conclusion and Suggestions ...................................................................................................27
Reference ......................................................................................................................................28
3
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
4
Abstract
Many factors influence the exchange rate; this study will attempt to examine two of these factors,
inflation and interest rates. This study is concerned with assessing the extent and the nature of
the impact of inflation and interest rates in Saudi Arabia on the SAR/SDR exchange rate, to
which the study hypothesized that there would be a negative relationship between inflation and
the exchange rate and a positive relationship between the interest rates and the exchange rate. A
regression analysis was used to on the monthly SAR/SDR exchange rate, inflation, and interest
rates for the period ranging from January 2004 to September 2015. The study has found that
there is a significant relationship between the SAR/SDR exchange rate and inflation, while there
is no significant relationship between the exchange rate and interest rates, further more the
finding have also confirmed the hypotheses of the study. The finding could have important
implications that would assist in making well-informed and more prudent investment and
financing decisions.
4
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
5
1. Introduction
Exchange rates are an important part of any economy. They allow countries to exchange
goods and services amongst each other. Exchange rates influence the supply and demand of
internationally traded goods as well as Foreign Direct Investment (FDI) flows (Shapiro, 2013).
There are many countries that have decided to fix their currencies to an anchor currency. Among
the reasons for that is to shield against major swings in the exchange rate as well as to lessen the
effect of inflation. Another important reason for fixing the exchange rate is to try to sustain
competitive prices especially when the anchor currencies are also their main trade partners. Some
countries try to depreciate their currencies for fear of the loss of trade competitiveness
(Krugman, 1994). In many cases, when allowed to float, exchange rates can reflect a certain
economy’s health.
Many currencies are pegged to the US dollar; one of the reasons would be is that the US
dollar is the world’s reserve currency. Among the countries that pegged their currency to the US
dollar is Saudi Arabia. The Saudi Arabian Riyal (SAR) was practically pegged, but not yet
officially, to the USD since 1986 at a rate of 1$ = SAR 3.75. The rate had become official in
2003. (SAMA, 2015)
5
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
6
The importance of the exchange rate to economies lead to numerous quantitative studies
conducted to forecast it or assess the factors that influence it. Among the methods used by many
researchers is the regression analysis, which is what this study has used. Alizadeh, Nassir, and
Masoudi, (2014) is an example of such a study. They use a regression analysis to test the
relationship between the exchange rates for ASEAN countries and interest rate differentials.
Not knowing the impacts of economic variables on the exchange rate in Saudi Arabia can
result in investors and traders making less ideal decisions. The knowledge would be helpful to
people engaging in foreign trade and investments in Saudi Arabia to know the influence of some
of these variables.
The purpose of this study is to investigate the impact of inflation and interest rates in
Saudi Arabia on the Saudi riyal exchange rate. The study will use a regression analysis on
monthly data of the SAR/SDR exchange rate, inflation rates, and interest rates obtained from the
IMF, ranging from the period January 2004 to September 2015. The method will be employed
with inflation and interest rates being the influencers (independent variables) on the SAR/SDR
exchange rate (dependent variable).
The significance of the study is that by better understanding the influences on the riyal
exchange rate, decision makers in Saudi Arabia will be able to make well-informed investment
decisions. This is especially true for investors and traders constantly engaging in foreign trade,
the findings could provide insight and possible foresight on the movement of the exchange rates,
which could have major impacts on investors’ returns and financing methods.
The research questions of the study are as follows:
Q1: How much do real interest rates in Saudi Arabia affect the Riyal exchange rate?
Q2: How much does inflation in Saudi Arabia affect the Riyal exchange rate?
6
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
7
The study hypothesizes that inflation in Saudi Arabia has an inverse relationship with the
Riyal exchange rate due to the depreciation of the currency. While interest rates in Saudi Arabia
have a direct relationship with the Riyal exchange rate due to the Fisher Effect.
The scope of the study is focused on the riyal exchange rate for the period ranging from
January 2004 to September 2015.
2. Literature review
The literature is rich with information regarding influences on exchange rates and
international parity conditions. In an attempt to review the relevant literature, four major themes
are presented: the Purchasing Power Parity, International Fisher Effect for different countries,
other factors that influence currencies and their exchange rates, and the factors that influence the
international parity conditions. These factors have been presented in the literature in different
contexts, and this study mainly focuses on the factors that more directly influence exchange
rates.
2.1 Purchasing Power Parity:
One of the most widely used methods of forecasting or establishing new par values for
currencies is the Purchasing Power Parity (PPP). It was first thoroughly stated by the Swedish
economist Gustav Cassel in 1918. The PPP identifies the relationship between domestic and
foreign price levels. This relationship results from the concept of arbitrage, where price
differences in different places or countries are reconciled due to trading activities that capitalizes
on such discrepancies. The PPP in its absolute version states that prices of goods and services
7
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
8
everywhere should be the same when expressed in one currency. The PPP in its relative version
is the version that is most commonly used in forecasting currency values. This version states that
the exchange rates of two currencies will move to adjust for changes in each country’s price
levels. That means if inflation in home currency is higher than in foreign currency, this will
result in the depreciation of the home currency value. (Shapiro, 2013).
The PPP has been studied in many countries and has had varying results. Among the
studies that have found the PPP to hold was one done by Paul and Motlaleng (2008). They
examined the PPP hypothesis in Botswana on the Pula-Dollar exchange rate, using cointegration.
They have found that the PPP in both its absolute and relative versions is validated for the years
1992 through 2002. Most of the studies that supports the PPP hypothesis found that it held
mostly in the long run, and had major deviation in the short run such as the case with the finding
of Cheug and Lai (1993), when they tested 3 countries over the period 1914-1989. As for the
studies that have found that the PPP does not hold was in the case of the Canadian dollar-US
dollar exchange rate. This study, done by Al-Zyoud (2015), in an attempt to examine the longrun relationship between the US dollar and the Canadian dollar, tested the PPP hypothesis for the
their exchange rate. Al-Zyoud (2015) employed a cointegration test on monthly data on the
period ranging from 1995 to 2008. The results indicated that the PPP does not hold in either the
absolute or the relative version in the case of the Canadian dollar-US dollar exchange rate. Mark
(1991), after testing the exchange rates between the following currencies: US dollar, deutsche
mark, Japanese yen, and British pound, also found that PPP does not hold in the long run.
Some studies have shown that PPP for a particular country can hold for some of its
exchange rates but not its other. In a study done by Tang and Liew (2010) to examine the PPP
hypothesis in Cambodia on nine of its exchange rates, different results have been observed for
8
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
9
different model testing methods. In one method they have used structured breaks in the unit root
testing, while in the other they did not. In both methods they have used the unit root analysis for
the period ranging from 2001 to 2008. By using unit root tests without structure breaks, they
have found that in all nine of Cambodia’s exchange rates in the study including; the US dollar,
British pound, Euro, Indonesian rupiah, Japanese yen, Malaysian ringgit, Philippine peso,
Singaporean dollar, and the Thai baht, the PPP hypothesis is rejected. But when the authors
included unknown structured breaks in the unit root tests, the results showed that the PPP
hypothesis is validated for four of the nine exchange rates mentioned, which include; the Euro,
Indonesian rupiah, Malaysian ringgit, and Singaporean dollar. Even though the Cambodian riel
and Cambodia’s monetary policies in the time period in Tang and Liew’s (2010) analysis had
been under less conventional circumstances, such as Cambodia’s high dollarization as well as
having two exchange rate systems (one of which an official rate that is a managed float used for
external transaction, and a parallel rate used for most other transactions), Kungher and Lenz
(1993) saw a similar pattern of result in their study that is concerned with the Duetsche mark.
The authors, examining 15 of the Duetsche mark’s exchange rates using a multivariate
cointegration method on data from its floating period from 1973 to 1990, to test the validity of
the PPP in the long-run, has found that it is valid for six European currencies which included
pound, Lira, Norwegian krone, Schilling, Escudo, and Peseta, but rejected the US dollar,
Canadian, dollar, Belgian franc, and Danish krone.
2.2 Interest Rate Parity Conditions:
The Fisher Effect (FE) and the International Fisher Effect (IFE) are other important parity
conditions. Economist Ivring Fisher first introduced the FE in the late 1930s. This concept states
9
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
10
that nominal interest rates increase as inflation increases. This means that, if there is no
government intervention, the nominal interest rate differential between two countries is equal to
their inflation rate differential. In other words, if inflation in home currency is high relative to
foreign currency then the home country should have higher interest rates. The IFE is a concept
that combines the PPP and FE, which demonstrates the relationship between interest rates and
inflation in two countries. This theory states that if a country has low interest rates relative to
other countries, this will result in the appreciation of the currency of that country. (Shapiro,
2013).
Some studies have concluded that the Fisher Effect (EF) and International fisher effect
(IFE) are valid in the long run. Incekara, Demez, and Ustaoğlu (2012) have found that, in the
case of Turkey, after analyzing data from the years 1989 through 2011, the EF holds in the longterm. Also, Aliber and Stickney (1975) have shown in their study that the EF and IFE also hold
in the long run, after testing 48 countries for a period of eleven years, from 1961 to 1971. Kane
and Rosenthal’s (1982) findings concurred for the case of countries with well-developed capital
markets. Macdonald and Nagayasu (2000), after studying 14 countries, during floating exchange
rate periods, also supports the claim that the IFE holds in the long run.
Even though there are studies that support the validity of the IFE theory in the long run,
there are others studies Cumby and Obstfeld (1981) that suggest it may be an insufficient
indicator of currency movement. Another study that had similar findings was conducted by
Alizadeh, Nassir, and Masoudi, (2014), to investigate the relationship of exchange rates and
interest rate differentials using the International Fisher Effect (IFE) framework. The authors
investigated these relationships in ASEAN country members, while considering Malaysian
ringgit as the home currency in the analysis. The data that they have used in their regression
10
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
11
analysis is for a ten-year period (from 2002 to 2012). The authors found that in the case of
Malaysia and Indonesia the IFE theory partially holds. As for the rest of the eight ASEAN
country members, the theory fails to hold.
2.3 Other Factor Influencing Exchange Rate:
There are many factors influencing the movement of currency values as well, some of
them include supply and demand of currencies, monetary policies, balance of payment problems,
and speculation (Khalway, 2000). They may be the reason behind differences in the results
among researches regarding these parity conditions. Some of these economic factors are worth
mentioning here.
Strong economic growth leads to currency appreciation (Shapiro, 2013). This growth is
due to the increase in foreign investment flows, which leads to increased demand of the growing
country’s currency. This appreciation will continue until the exchange rate reaches equilibrium.
This positive relationship between economic growth and currency value is illustrated in Bussierè,
Claude, and Lopez’s (2014) paper that included 68 countries for the period from 1960 through
2011.
The balance of payment of a country has an effect on the exchange rate (Samuelson &
Nordhaus, 2005). This effect can be seen in the instances where there is a deficit in the current
account; it means that the country is importing more than it is exporting. This deficit would
result in the depreciation of the exchange rate relative to their trade partners, due to lack of
demand for the importer’s currency brought about by their lack of exporting goods or services.
The literature has shown that the balance of payment and exchange rate are interrelated, they
influence one another, but have common factors influencing them both, such as inflation. A
11
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
12
study done by Kappler, Reisen, Schularick, and Turkisch (2012) showed that if a currency
appreciates that would have an adverse affect on its balance of payment, specifically decreasing
the current account, and this effect is more prominent in emerging and developing markets.
These relationships can be seen in the concept of export competitiveness. This concept sates that
a weaker currency would increase exports competitiveness, meaning increase in said country’s
exports thus increasing its current account. Many exporting nations have showed strong
concerns, bordering on obsession, about their export competitiveness (Krugman, 1994). A
number of countries, including China, devaluated their currency to maintain their export levels
against their appreciating currencies. It is apparent that this relationship is an important factor
with regard to exchange rate or currency analysis.
Interest rates and inflation are two interrelated factors that impact the exchange rate.
These factors are related in a way that changes in one would lead to inverse changes in the other
(inverse relationship). If interest rates increase in a particular country, its currency would
appreciate due to increased demand for said country’s currency. That is a result of increased
demand for that country’s bank accounts (time deposit accounts) and bonds, in order to reap the
benefits of these higher interest rate returns (Shapiro, 2013). This increased currency value or
strength would translate into a decrease in inflation relative to other countries. This decrease in
relative inflation would also mean that the currency would appreciate against other currencies
thus appreciating its exchange rate (the PPP hypothesis). Looking at the interest rate and
inflation relationship from another perspective, one would notice the Fisher Effect (FE). As
mentioned earlier the FE states that the real interest rate is equal to the nominal interest rate
excluding inflation. Therefor, this relationship states that an increase in inflation will result in the
decrease of the real interest rate (Shapiro, 2013), showing the relationship in reverse as well.
12
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
13
Other factors that have an influence on the exchange rate includes, political and economic
risks. Throughout history, there are many examples on how adverse political and economic
events have negative impact on the economy and, in turn, the currency value. These adverse
political events would drive away foreign investment, thus stunting economic growth and
depreciating the currency. Lim’s (2003) findings indicated, after analyzing 25 countries over the
expanse of 8 years, that there is a significant impact of political risk on the nominal exchange
rate.
2.4 Factors Influencing Parity Conditions:
Other studies have attempted to investigate the PPP and IFE hypotheses in different
exchange rate regimes and during transition from one regime into another. Qureshi (1993) found
that the PPP holds under a fixed as well as flexible exchange rate, but there were significant
deviations of PPP in the short run for the flexible exchange rate in the study. In the case of
transitioning from a fixed currency regime to a flexible one, Koráb and Kapounek, (2013) have
found that the adjustment of the interest rate and inflation differentials to meet the IFE depends
on the degree of rigidity of the fixed exchange rate prior to the transition to a floating one. The
IFE held mainly for the group of countries in the study that had rigid exchange rate policies prior
to shift to free floating.
One can notice from the in literature concerned with parity conditions for different
countries, results and conclusion vary a lot. When these parity conditions do hold, they seem to
do so only in the long run. In an effort to study Interest Rate Parity in developing and emerging
countries, Ito and Chinn (2007), found that the variation in finding regarding the validity of
parity conditions for certain countries is due to different factors within each respective country.
13
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
14
These factors include their inflation rate and its volatility, capital account openness (Samuelson
& Nordhaus, 2005), legal factors, and the nature of the exchange rate regime as noted by Qureshi
(1993) and Koráb and Kapounek, (2013).
Economic openness, the degree to which an economy is engaged in international trade
(Samuelson & Nordhaus, 2005), may be a major factor to consider, when trying to understand
why parity conditions for some countries do not hold. Balassa (1964) suggested that in order for
the PPP to hold, certain factors must exist. Among them, the absence of trade restrictions; that
means that the degree of economic openness is important for such parity condition to hold.
Multiple studied have concluded that the PPP hypothesis does not hold for the United States
exchange rates, such as the studies done by Al-Zyoud (2015), Mark (1990), Tang and Liew’s
(2010), and Kungher and Lenz (1993). The United States is a majorly self-sufficient economy,
that being so it has among the lowest degrees of economic openness (Samuelson & Nordhaus,
2005). This relatively low degree of economic openness in the US economy may be the reason
behind the PPP hypothesis not being validated for it.
The focus of this study is the impact of both interest rates and inflation in KSA on the
Saudi Arabia riyal. To the best if the best knowledge of the researcher, no such study has been
conducted. Hence, this study contributes to the literature by examining the important relationship
between inflation and interest rates in the context of Saudi Arabia.
3. Data and Methodology
3.1. The Data:
This study will attempt to test the effect of inflation and interest rates on the Saudi
Arabian riyal (SAR) exchange rate using data obtained from the IMF website. The SAR will be
14
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
15
tested against the Special Drawing Rights (SDR). The study used monthly data for the period
starting from January 2004 to September 2015, using the following variables:

SAR/SDR exchange rate.

Inflation rates based on Saudi Arabian Consumer Price Index (CPI)

Interest rates in Saudi Arabia.
The SDR is an international reserve asset. The IMF created the SDR in 1969 for the
purpose of supporting the Bretton-Woods system of fixed exchange rates. After the collapse of
the Britton-Woods system the SDR was redefined as a basket of currencies. It comprises of four
key currencies, which are: the euro, Japanese yen, Pound sterling, and UD dollar. The SDR
serves and a supplementary reserve asset, and a unit account of the IMF and a few other
international organizations. The SDR is not a currency in the conventional sense, but SDR units
can be exchanged for the currencies in the SDR basket. (IMF, 2015).
3.2. Models:
To investigate the relationship between the exchange rate and the mentioned variables,
the study used the following models:
The univariate models:
These models will be used to assess the impact of each variable on the exchange rate
individually. The first model (Model 1), includes inflation alone as the independent variable,
which is the variable that acts as a possible predictor for the exchange rate. The expected
relationship is that inflation has a negative effect on exchange rates, because it depreciates the
currency relative to other currencies (Shapiro, 2013).
Model 1
𝑌𝑖 = 𝛽0 + 𝛽1 𝑋𝑖 + 𝜇𝑖
15
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
16
Where:
𝑌𝑖 = Monthly exchange rate (SAR/SDR)
𝑋𝑖 = Monthly inflation
𝜇𝑖 = Error term
The second model (Model 2) contains interest rates as the independent variable. The
study hypothesizes that the interest rates (specifically the nominal interest rates) has a positive
relationship because of the International Fisher Effect, which states that changes in interest rates
differentials between to countries is equal changes in the exchange rate differential.
Model 2
𝑌𝑖 = 𝛽0 + 𝛽1 𝑋𝑖 + 𝜇𝑖
Where:
𝑌𝑖 = Monthly exchange rate (SAR/SDR)
𝑋𝑖 = Monthly interest rates
𝜇𝑖 = Error term
The Multivariate model:
This model shall include both of the inflation and the interest rates, to see if a
combination of the two variables have a more significant impact on the exchange rate.
Model 3
𝑌𝑖 = 𝛽0 + 𝛽1 𝑋1𝑖 + 𝛽2 𝑋2𝑖 + 𝜇𝑖
Where:
𝑌𝑖 = Monthly exchange rate (SAR/SDR)
𝑋1𝑖 = Inflation
𝑋2𝑖 = Interest rate
𝜇𝑖 = Error term
16
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
17
3.3. Descriptive statistics:
This section is dedicated to describing the statistics of each of the three variables in the
study. The sample period for all the variables is from January 2004 to September 2015. All the
variables are monthly data.
Graph 1
17
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
18
The first variable we will be looking at is the monthly SAR/SDR exchange rate. As
shown in the table (variable #1) during 25% of the observations of the SAR/SDR exchange rate
was at 5.54534 or lower, and 75% of the time the exchange rate was at 5.81898 or lower. The
highest that the exchange rate has been during the period in this study was 6.16688, and the
lowest it has been was 5.17309. The average value of the exchange rate for that period was
5.68773. The sample has a standard deviation (𝜎 2 ) of 21.098%, which indicates that most of the
data points are spread out somewhat far from the mean. The exchange rate data skewness was 0.18116, which means that the distribution of the data in the sample is approximately
symmetrical, as can be view in histogram and distribution curve graph 1. The kurtosis, which
relates to the shape of the distribution of the data in the sample, was 2.7357, which says that it
has a mesokurtic distribution, very close in shape to the normal distribution (graph 1,).
18
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
19
Graph 2
The descriptive statistics for the second variable, inflation, is illustrates in table (variable
#2) and graph 2. 25% of the observations inflation was at 0.1% and lower and 75% of the
observations it was 0.406% and lower. The highest the inflation rate had been during the period
of the study was a maximum of 1.922% and at its lowest in the period was at a minimum of 0.575%. The average inflation rate for the period was 0.314%. The data has a standard deviation
(𝜎 2 ) of 0.375%, this indicate that the data points are clustered relatively close to the mean. As
can be observed in histogram and distribution curve (graph 2), the skewness of the inflation data
distribution was 1.35539, which is moderately skewed, with the tail of the distribution
moderately to the right. The distribution curve is leptokurtic (high kurtosis of 6.40939), which
means the data in the distribution has a relatively high peak and its tail is longer (graph 2).
19
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
20
Graph 3
Interest rates are the third variable in the study and the statistical descriptions are
illustrated in the table (variable #3) and graph 3. Through the study period interest rates had been
at 0.25% and lower for 25% of the observations and 3.0% and lower 75% of the observations.
The maximum value the Saudi interest rates have reached was 5.0% while the minimum value
was 0.25. The Saudi interest rates have reached its lowest value since the beginning of 2009 and
remained at the value until the end of the study period. The standard deviation (𝜎 2 ) for interest
20
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
21
rates in this sample was 174.93%, which indicates that the data points are spread out quite far
from the mean. One can see from the graph that data distribution is moderately asymmetrical,
which gives a skewness value of 0.94036 (moderately skewed to the right). The graph is also
platykurtic (kurtosis > 3), which means the central peak of the distribution curve is low and
broad.
Each of the variables in the study showed different levels of deviation within the sample
period. The monthly interest rates showed the highest deviation from the mean while inflation
had the lowest deviation. Inflation and interest rates being the variables, whose effects would be
tested on the exchange rate, should yield interesting results.
3.4. Model Testing:
The models of the study are tested using regression also called Ordinary Least Square
(OLS). The response (dependent) variable is the monthly SAR/SDR exchange rate, and the
explanatory (independent) variables are the monthly inflation rates and interest rates. The
explanatory power of the model will be tested using the coefficient of determination (𝑅 2 ), which
is the ratio of the sum of squared deviations of the predicted values of 𝑌𝑖 (𝑌̂𝑖 ), from their mean
value, and the sum of squared deviations of 𝑌𝑖 from its mean (Stock & Watson, 2007). In the
context of this study 𝑌𝑖 is the values exchange rate and 𝑌̂𝑖 is the predicted values of the exchange
rate.
𝑅2 =
𝐸𝑆𝑆
𝑇𝑆𝑆
Where:
𝐸𝑆𝑆 = The squared deviations of the predicted values 𝑌̂𝑖 from their mean (Explained Sum
of Squares).
21
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
22
𝑇𝑆𝑆 = The squared deviations of the values 𝑌𝑖 from its mean (Total Sum of Squares).
The higher the 𝑅 2 the better the data fits the model, and the more it can explain or predict the
deviations of the exchange rate.
3.5. Hypothesis testing:
T-statistic is used to test the hypotheses about the coefficients in the regression. In the
case of the univariate models the coefficients in the equations 𝛽1, determine the slope of the
regression line. If the regression line is flat that means that the regression coefficient (𝛽1 ) is
equal to zero. That would mean that independent variable could have no effect on the dependent
variable, and that would result in accepting the null hypothesis (𝐻0 ). The alternate hypothesis
(𝐻1 ) would be the one being accepted if 𝐻0 was not. Accepting 𝐻1 would mean that coefficient
𝛽1 is not equal to zero, which suggests that the independent variables have an effect on
dependent variable. The t-test will determine whether one should accept or reject the null
hypothesis. (Stock & Watson, 2007)
𝐻0 : 𝛽1 = 0 𝑜𝑟 𝛽1,0
𝐻1 : 𝛽1 ≠ 0 𝑜𝑟 𝛽1,0
Where 𝛽1,0 is some specific value.
The t-statistic is calculated as follows:
𝑡=
𝛽̂1 − 𝛽1,0
𝑆𝐸(𝛽̂1 )
Where:
𝛽̂1 = The estimator.
𝑆𝐸(𝛽̂1 ) = The standard error of the estimator.
22
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
23
𝛽1,0 = Hypothesized value.
The critical value will be compared with the calculated t-statistic at significance level of
5%, to determine where to reject or accept the null hypothesis. Which means that if the absolute
value of the t-statistic is greater that 1.96 in a two-sided-test (value obtained from the z-table) the
null hypothesis is rejected. (Stock & Watson, 2007)
The F-statistic will be used to test joint hypothesis about the regression coefficients for
the multivariate model. The F-statistic combines the two t-statistics 𝑡1 and 𝑡2 , resulting from
each coefficient in the regression, 𝛽1 and 𝛽2. In the case of this model the null hypothesis has
two restrictions 𝛽1 = 0 and 𝛽2 = 0. The F-statistic is the average of the square t-statistics. (Stock
& Watson, 2007)
1 𝑡12 + 𝑡22 − 2𝑝̂ 𝑡1, 𝑡2 𝑡1 𝑡2
𝐹= (
)
2
1 − 𝑝̂𝑡1, 𝑡2
The F-statistic tests the joint hypothesis that all the coefficients in the regression are equal to
zero. The null and alternate hypotheses are as follows:
𝐻0 : 𝛽1 = 0, 𝛽2 = 0,
𝐻1 : 𝛽𝑗 ≠ 0, 𝑎𝑡 𝑙𝑒𝑎𝑠𝑡 𝑜𝑛𝑒 𝑗,
To determine whether to reject or accept 𝐻0 , the F-statistic will be compared to the F critical
value (from the F-distribution) at a significant level of 5%. Accepting 𝐻1 if the F-statistic value
is greater than the F critical value.
4. Findings
23
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
24
This section is dedicated to present and interpret the findings. The regression analyses
have yielded the following numerical and graphical outputs. They will be in the following order:
the two univariate models (model. 1) and (model. 2), and the multivariate (model. 3).
Table. 1
Exchange rate
Scatter Diagram (Predicted Y, SAR/SDR exchange rate vs. Inflation)
6.1
5.9
5.7
5.5
5.3
5.1
-0.01
-0.005
0
0.005
Inflation
0.01
0.015
0.02
Figure 1
24
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
25
The regression output for the first univariate model is tabulated in table 1. The
coefficient of determination, which tests the explanatory power of the model, is relatively low
R2=0.11133. This means that the model explains 11.133% of the variance in the exchange rate.
This is due to the high variability of the actual data (exchange rate) from the predicted data
(predicted exchange rate) as seen in figure 1. In other words, inflation in this model may not
account for most of the exchange rate movement in the time series.
It can be inferred from the t-statistic and p-value (table. 1), there is strong evidence
against the null hypothesis 𝐻0 of 𝛽1 equaling zero. The t-statistic = 4.14 is greater than the
critical value at the significance level of 𝛼 = 0.1%, t-statistic > 2.575. This suggests that a type I
error in the hypothesis rejection is highly unlikely. This is also evident in the slope of the
regression line (figure. 1), that 𝛽1 (coefficient of inflation) in not equal to zero. Therefore, the
null hypothesis is not accepted and 𝛽1 is not equal to zero.
Table 2
25
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
26
Exchange rate
Scatter Diagram (Predicted Y, SAR/SDR exchange rate vs. Interest
rate)
6.1
5.9
5.7
5.5
5.3
5.1
0
1
2
3
Interest rate
4
5
6
Figure 2
The regression output for the second univariate models is tabulated in table 2. The R2 is
very low, indicating that the regression model explains only 1.996% of the variation in the data.
In other words the deviation explained by the regression is very low compared to the total
deviation. This means that interest rates in this model may not be able to explain or predict the
deviation in the exchange rate.
The t-statistic and p-value indicate that there is strong evidence for the 𝐻0 of a zero 𝛽1.
The 𝐻0 has been accepted, because the t-statistic (= -1.67) much smaller than the critical value at
a the significance level 𝛼 = 5%, -1.67 < 1.96. The value of coefficient 𝛽1 is very small and
close to zero. The slope of the regression line in figure 2 backs this; one can notice that it
relatively flat.
26
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
27
Table 1
The regression output for the multivariate model is tabulated in table 3. The R2 for this
model is relatively low, indicating that the model explains or predicts 13.99% of the deviations
in the data. It is slight improvement over the model 1 but still doesn’t account for the majority of
the deviations in the exchange rate.
The F-statistic, which is generally the average of the two squared t-statistics, tests the
hypothesis for the coefficients 𝛽1 and 𝛽2 jointly. The F critical value was found to be equal to
3.8891, which is smaller than the F-statistic value. This resulted in 𝐻0 not being accepted. It is
highly likely that the coefficients are not equal to zero, and the chance of committing a type I
error is highly unlikely.
One can notice in the finding that multivariate model may be a better predictor of
exchange rate movements than univariate models. That finding was interesting because the
model 2, the one containing the interest rate as a possible predictor, performed poorly compared
to the other univariate model 1. In the multivariate model the inclusion of the interest rate seems
to have marginally improved the model.
27
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
28
5. Conclusion and Suggestions
Exchange rates are dynamic and many factors have been found to have an influence on
them. They are an important and influential part of the economy, and need to be considering
when making decisions in investment and financing. The study focused on assessing the impact
of inflation and interest rates in Saudi Arabia on the SAR/SDR exchange rate. The study has
found that inflation in Saudi Arabia has an impact on the SAR/SDR exchange rate and may be an
important factor to consider when analyzing the exchange rate. Although the univariate inflation
model may not have been able to explain the majority of the deviation of the exchange rate, the
null hypothesis of a zero coefficient 𝛽 was not accepted at a significance level of 0.1%. This
suggests that there is an important relationship between inflation and the exchange rate. The
study has also found that Saudi interest rates may not have a significant impact on the Saudi
exchange rate. The explanatory power of the interest rate univariate model was very weak, and
null hypothesis of a zero 𝛽 was accepted. Therefore, there may not be a significant relationship
between Saudi interest rates and the Saudi exchange rate. Lastly, the study has found that
combining the two variables provides a slightly better prediction to the exchange rate movement,
which means the explanatory power of the multivariate model is slightly stronger, although it
does not account for most of the deviation of the exchange rate, the null hypothesis of a zero 𝛽
has been rejected at a significance level of 0.1%. This could suggest that jointly these variables
may have an important relationship with the exchange rate in the case of Saudi Arabia.
Furthermore, the findings confirm the hypotheses of the study, of the SDR/SAR
exchange rate having a negative relationship with the inflation rate and a positive relationship
with the interest rate. That is evident in the equation in the first regression output where the
coefficient of the variable of the inflation rate is positive, indicating that if inflation increases the
28
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
29
value of the SAR (SAR/SDR exchange rate) would decrease. As for the conformation of the
hypothesized relationship between the SAR/SDR exchange rate and the interest rate, it is also
evident in the equation of the second regression output. The coefficient of the variable of the
interest rate is negative indicating that if the interest rates decrease that would result in the
decrease of the SAR/SDR exchange rate.
In light of this conclusion, noticing that the inclusion of one more factor in the regression
model improved its explanatory power and its p-values, future research on the SAR exchange
rate should be conducted with more economic variables to attempt to paint a bigger picture of the
dynamics of the SAR exchange rate.
Reference:
Al-Zyoud, H. (2015). An empirical test of purchasing power parity theory for Canadian dollarUS dollar exchange rates. International Journal of Economics and Finance, 7(3), 233240. Retrieved from http://search.proquest.com/docview/1668027541?accountid=130572
Aliber, R.Z. and C.D. Stickney, 1975. Accounting measures of foreign exchange exposure: The
long and short of it. Accounting Review., 50:44-57.
Alizadeh, H. N., Nassir, A. M., & Masoudi, S. Y. S. A. (2014). An empirical investigation of
international fisher effect on 10 asean countries. Journal of Applied Sciences, 14(6), 531536. Retrieved from http://search.proquest.com/docview/1501915090?accountid=130572
Cumby, R.E. and M. Obstfeld, 1981. A note on exchange rate expectations and nominal interest
differentials: A test of the fisher hypothesis. Journal of Finance, 36: 697-703.
Incekara, A., Demez, S., & Ustaoğlu, M. (2012). Validity of Fisher Effect for Turkish economy:
Cointegration Analysis. Procedia - Social and Behavioral Sciences, 396-405.
Kane, E. and L. Rosenthal, 1982. International interest rates and inflationary expectations.
Journal of International Money Finance, 1:97-100.
Kappler, M., H. Reisen, M. Schularick, and E. Turkisch, E. (2012), “The Macroeconomic Effects
of Large Exchange Rate Appreciations”, Open Economies Review.
Khalwaty, T., 2000. Inflation and solutions. 1st Edn, Gramdia Pustaka Utama, Indonesia.
29
THE IMPACT OF INFLATION AND REAL INTEREST RATES IN SAUDI ARABIA ON THE
SAR/SDR EXCHANGE RATE
30
Koráb, P., & Kapounek, S. (2013). International fisher effect under exchange rate regime shifts:
Evidence from 10 examples. Society and Economy, 35(4), 451-469.
doi:http://dx.doi.org/10.1556/SocEc.35.2013.4.2
Krugman, P. (1994), “Competitiveness: a Dangerous Obsession”, Foreign Affairs, 73 (2).
Kugler, P. and C. Lenz. (1993), ‘Multivariate Cointegration Analysis and the Long-Run Validity
of PPP’, Review of Economics and Statistics, Vol. 75, pp.180-184.
Paul, M. T., & Motlaleng, G. R. (2008). The pula-dollar exchange rate and the purchasing power
parity in botswana. The Journal of Developing Areas, 41(2), 205-231. Retrieved from
http://search.proquest.com/docview/220447850?accountid=130572
Price-based Measurement of Financial Globalization: A Cross-Country Study of Interest Rate
Parity by Hiro Ito* Portland State University and Menzie Chinn** University of
Wisconsin and NBER June 30, 2007
Qureshi, A. (1993). Purchasing power parity under fixed and flexible exchange rates in
developing countries. A Indian Economic Journal, 3(40), 51-51.
Shapiro, A. (2013). Multinational financial management (10th ed.). New York: J. Wiley & Sons.
Stock, J., & Watson, M. (2007). Introduction to econometrics (2nd ed.). Boston:
Pearson/Addison Wesley.
Tang, T. C., & Liew, V. K. (2010). Purchasing power parity in a transition economy-cambodia:
evidence from bilateral exchange rates. The Asia Pacific Journal of Economics &
Business, 14(1), 55-73. Retrieved from
http://search.proquest.com/docview/817183812?accountid=130572
30