







































Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

Indian Journal of Finance and Banking; Vol. 4, No. 1; 2020 
                                       ISSN 2574-6081   E-ISSN 2574-609X 

Published by Centre for Research on Islamic Banking & Finance and Business, USA 

134 

 

Impact of Foreign Exchange Reserve, Exchange Rate and Crude Oil Price on 
Dhaka Stock Exchange Index: An Empirical Evidence from Vector Error 

Correction Model 
 
 

Uttam Golder 
Lecturer 

Department of Finance and Banking 
Jashore University of Science and Technology 

Jashore-7408, Bangladesh 
E-mail: uttamgolder@gmail.com 

 
Md. Nazrul Islam 
Assistant Professor  

Department of Finance and Banking 
Jashore University of Science and Technology 

 Jashore-7408, Bangladesh 
E-mail: nazrul@just.edu.bd  

 
Md. Shahidullah Kayser 

 Assistant Professor 
Department of Finance 
Jagannath University 

 Dhaka-1100, Bangladesh 
E-mail: kayser.shahid@outlook.com 

 
Abstract 
The supreme thrust of the present analysis is to explore the influences of foreign exchange reserve, exchange rate, and crude 
oil price on the stock index of the Dhaka stock exchange (DSE) of Bangladesh. Moreover, this study evaluates the identity of 
any unpremeditated relationship among the variables from the viewpoint of an emerging country like Bangladesh. Through 
using monthly time-series data, this study tries to discover the evidence of a long-run affiliation among the variables by using 
Johansen’s Cointegration test and Vector Error Correction Model (VECM). Besides, the Granger Causality technique is 
introduced to examine the casualty among variables where the empirical results show a causal linkage between the Dhaka stock 
exchange index, foreign exchange reserve, and exchange rate, moving only in one way from Dhaka stock exchange index to 
foreign exchange reserve and exchange rate. In contrast, no causal link was identified between Dhaka stock exchange indexes 
and crude oil prices. Lastly, Impulse Response Function suggests a permanent effect of all selected macroeconomic factors on 
the Dhaka stock exchange index in the long run and Variance Decomposition Analysis settles that, the reform in Dhaka stock 
exchange index can be caused by the innovation in foreign exchange reserve, exchange rate, and crude oil price. 

 
1. Introduction  
Stock market indices are useful overall scores that show a clear historical record of market prices and thus allow investors, 
advisors, and the public to measure what the present state of stock price trend happens to be. It is also noticeable that company 
stock price may not always in the identical path as per the market index, rather it can be opposite to the trend. Indices are 
subject to fluctuation, so it is of great importance to determine the factors that play an imminent role in deciding the trend  
(Khan & Yousuf, 2013). Among some significant macroeconomic factors that affect market index, this paper cautiously 
picked up foreign exchange reserve, exchange rate, and crude oil price as the explanatory variables and explored the relationship 
pattern among them. 

     In general, one country differs from another in different points of view, such as its population, geography, level of 
education, development status. It is expected that Bangladesh will be a developing country from the least developed country 
(LDC) in 2024, where three indicators require proper positioning of gross national income, index of the human asset, and 
economic vulnerability. For Bangladesh, it is a difficult challenge, and the most pressing need to meet this challenge is to 
achieve economic viability. In this case, the foreign exchange reserve can play a significant character in maintaining the 
confidence of a country's economic market. However, nowadays, it is a debating issue whether excess foreign exchange reserve 
is good for the economy of any country or creates some opportunity and social costs for the country (Mezui & Duru, 2013).  

     The exchange rate is another problematic issue that often leaves policymakers in a dilemma. No country in the world 
is self-sufficient; therefore, one country has to relay on another for making both import and export, which requires the 

  
Keywords:    Crude Oil Price, DSE Index, Exchange Rate, Foreign Exchange Reserve, VECM. 

 
    

mailto:uttamgolder@gmail.com
mailto:nazrul@just.edu.bd
mailto:kayser.shahid@outlook.com


Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

www.cribfb.com/journal/index.php/ijfb                         Indian Journal of Finance and Banking                             Vol. 4, No. 1; 2020 

135 

 

exchange of currencies. Indeed, the increase of the exchange rate (devaluation of the currency) discourages the importer, just 
as it encourages the exporter and vice versa. Therefore, different countries decrease or increase their exchange rate according 
to their different policies, which in sequence influences the capital market in different ways through the production and sale 
of the products.   

     Without energy, the whole world is untouchable, and hence every country has to rely on crude oil for irrigation in 
agriculture and production in the industrial sector. Not all the countries in the world have oil reserves, and for this reason, 
many countries have to manage their production depending on the oil imported from other countries. The real situation of 
the stock market of countries in which countries conduct agricultural operations and industrial activities with imported oils 
is generally different from those countries which do not need to import oil.   

     Previously, numerous macroeconomic variables were used by many researchers to unveil the connection of these 
variables with the stock returns in different stock exchanges of different countries, including the Dhaka stock exchange (DSE) 
of Bangladesh. The impact of these variables is not same always due to the different contexts and nature of the market 
(Chauque & Rayappan, 2018). Several theories are applied to explore the intensity of the linkage among the variables. The 
time frame was also an essential factor to be considered as the researchers pondered different time range to conduct their 
study. However, authors of this paper are highly persuaded to inspect the influence of foreign exchange reserve, exchange rate, 
and crude oil price on the stock index and also inspect the cointegration among the variables in the context of DSE, Bangladesh. 
Although many other scholars have already explored the influence of the chosen variables in different stock markets (Hasan, 
2018; Kibria et al., 2014; Zaidi, Ahmed, & Siok, 2017) the nature and context of Dhaka Stock Exchange (DSE) vary from 
their observed markets which is the ultimate motivation of the authors in conducting this study. 

     This study aims to know whether the foreign exchange reserve, exchange rate, and the crude oil price have a long-term 
consequence on DSE index. The status of a causal linkage between the variables is another exploratory question that leads to 
determining the aim of the paper. Another query that comes to the authors' investigative mind is that, if the background of 
the stock market changes, then what will be the difference in the result comparing to the past literature? 
The researchers, therefore, find interest in re-examining the long-run influence of foreign exchange reserve, exchange rate and 
crude oil price on the stock index in a different setting by exerting Johansen's Cointegration test and VECM. Besides, the 
paper observes the presence of any causal affiliation among the dependent and explanatory variables via the Granger Causality 
approach. 

     This study has some practical implications in the real field and is equally essential for students, academicians, and 
policymakers of government. The adverse effects of foreign exchange reserve and crude oil prices are shown here with the 
finger-pointing. In short, the results of the study can help one to easily perceive how an importer and exporter can be benefited 
or suffered due to adverse fluctuations in the exchange rate. 
 
2. Literature Review  
2.1 Foreign Exchange Reserve and Stock Index 
Foreign exchange reserve includes the deposits of foreign currency and bonds. However, in the broader sense, it is a country’s 
holdings of gold, SDR, and reserve of IMF for meeting its short to medium term international financial liabilities 
(Maheshwari, Upamannyu, Bhakuni, & Saban, 2013). Generally, this reserve is used to settle the payment of imported goods 
and services and repayment of international loans taken by individuals and government. Moreover, multinational companies 
send their royalty and profit to their parent firms in international currencies. The importer can easily open L/C with a 
minimum cost because the foreign exporters consider the importing country less risky and are assured of receiving their money 
in time. It stimulates the stock index of the importing nation as it can import raw materials and capital goods for its 
production, which increases the firm’s profit by selling produced goods and thereby flourishing stock index. 

     There is a maxim, “nothing excess is good.” Though adequate reserve is considered blessings for any country, the 
excess reserve also has some adverse effects. If an import-oriented country can repay its import payment for the next three 
months, it is expected that the country has a standard reserve. The most troublesome negative impact of the excess reserve is 
the opportunity cost (Mezui & Duru, 2013). The central bank has to hold reserve either in direct currency or in extremely 
liquid financial assets that can be converted into cash with a little cost. 

     Furthermore, direct currency does not earn any interest, and the extremely liquid financial asset earns low yield. If the 
excess reserve were possible to invest in another productive sector, it would profit more, and the economy would be more 
flourished. Besides, the excess reserve is shaped when import payment is decreased continuously due to the decline in the 
import of raw materials and machinery used in the local industry. 
 
2.2 Exchange Rate and Stock Index 
The interrelation between exchange rate and stock performance has long continued a topic of debate. Several scholars consider 
the exchange rate as a double-edged blade as it converts one currency into another and displays an incongruous effect at the 
same time. There are two popular theories of the exchange rate, namely the flow-oriented and stock-oriented models (Fauziah, 
Moeljadi, & Ratnawati, 2015). Flow oriented theory (Dornbusch & Fischer, 1980) postulates that the exchange rate influences 
the nature of competitiveness of a business in international markets by affecting the rate of interest, profit, production, and 
mostly on the worth of the share of the respective firm. The influence of the movement of the exchange rate is different from 
the organization, which is operated in the international market from that of the local market. If the firm is export-oriented, 
an upsurge in the exchange rate (devaluation of local currency) will appreciate the exporter, and higher growth in export will 



Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

www.cribfb.com/journal/index.php/ijfb                         Indian Journal of Finance and Banking                             Vol. 4, No. 1; 2020 

136 

 

lead to higher profit and, ultimately, the share price will increase. Khan, Khan, Ahmad, and Bashir (2018) studied the monthly 
stock price data of 15 firms of Pakistan from the period of 2008 to 2012, and they confirmed a positive influence of the 
exchange rate on stock yields through using the OLS regression model. 

     Similar findings were shown by Khalid and Khan (2017) who found a positive impact of the exchange rate by analyzing 
time series data from 1991 to 2017 in Pakistan. However, it will create a burden for import dominated industry by increasing 
the cost of imported raw material and thereby increasing the cost of production. It decreases the earnings of the firm, which 
also makes a decline in the stock price. Ndlovu, Faisal, Resatoglu, and Türsoy (2018) inspected the affiliation between stock 
yield and macroeconomic indicators in the context of South Africa from 1981 to 2016 by using a vector error correction 
model. They concluded that the exchange rate possesses a negative impact on share values. A downturn in the exchange rate 
(encouragement of local currency) will encourage importer, and thus the cost of production will be lower. It will create a high 
demand for domestic products, and the firm will earn more profit, which results in an upsurge in the share price. However, 
this arrangement will hamper the earnings of export-oriented firm and thus decreases the price of stocks.   

     Stock oriented theory (Frankel, 1993) states that there is a high capital inflow if the stock return is increased. Investor’s 
insight into the capital market is very much crucial for the fluctuation of the stock price. If the price of the stock is increased, 
it will make a positive perception about the domestic stock market, and individuals will tend to invest in the local market by 
selling their foreign stocks and thus there will be a high demand for the local currency which will decrease the exchange rate 
by appreciating the worth of the domestic currency. 

 
2.3 Crude Oil Price and Stock Index 
Changes in the international oil price are measured as a vital issue for analyzing the change of the stock market index (Giri & 
Joshi, 2017). The influence of crude oil prices on the stock index is mixed that depends on the ground that whether the state 
is an oil-exporting or importing one. If the nation is an exporting one, a rise in oil price will upsurge cash inflow, and thereby 
the earnings of the company will be expanded, which will be reflected in the stock index. Consequently, income from exporting 
oil will increase public expenditure and create new investment opportunities. As per the findings of many researchers, the oil 
price possess a positive reaction on the stock index (Shafi, Hua, Idrees, & Nazeer, 2015). The effect of the crude oil price 
was also found positive in China (Hosseini, Ahmad, & Lai, 2011). 

     On the contrary, some others suggested that in the oil-importing state, there is an inverse relationship. The country 
whose production activities largely depend on imported oil, an upsurge in oil price, makes a significant contraction in its 
economic progress and decreases the stock index (Khan & Yousuf, 2013). Extensive shipping and production costs may lower 
the demand for the produced goods and lower the company profits, which can disincline the investor to purchase the share 
of the particular firm, and consequently, the stock index decreases (Miller & Ratti, 2009). Similar findings were reported in 
the stock exchange study, showing an adverse outcome of the oil price on stock return covering the year 1979 to 2014 (Giri 
& Joshi, 2017). 

 
3. Methodology 
3.1 Characterization of Data 
This study has been conducted adopting monthly historical data ranged from July 2008 to October 2019. The index of 
Dhaka Stock Exchange is extracted from the authorized database of the company and the other two macroeconomic variables 
out of three, e.g., foreign exchange reserve and exchange rate were collected from the monthly publications (Monthly 
Economic Trends) of the central bank of Bangladesh (Bangladesh Bank, 2019). The data of the latter variable (crude oil 
price) was taken from the IFM’s primary commodity prices (International Monetary Fund, 2019). For handling extreme 
values, all the variables have been converted into their natural log procedure and ensured the steadiness of the variables. 
However, the variables are symbolized by LOGDSEI (natural log of all share price index of Dhaka Stock Exchange), 
LOGFER (natural log value of foreign exchange reserve), LOGER (natural log of the exchange rate of BDT for US dollar), 
and LOGCOP (natural log of per barrel crude oil price in US dollar).  
 
3.2 Conceptual Outline 
This review emphasizes on the estimation of any long-run affiliation between the index of DSE and particular macroeconomic 
factors as well as defining the causal relationship among them. For fulfilling the objectives, the econometric model (1) was 
used:  

𝐿𝑂𝐺𝐷𝑆𝐸𝐼𝑡 = 𝛽0 + 𝛽1𝐿𝑂𝐺𝐹𝐸𝑅𝑡 + 𝛽2𝐿𝑂𝐺𝐸𝑅𝑡 + 𝛽3𝐿𝑂𝐺𝐶𝑂𝑃𝑡 + 휀𝑡                                      (1) 

     Here, the variables have already been defined. 휀𝑡 indicates error terms. 𝛽0 is constant, and 𝛽1, 𝛽2, and 𝛽3 signify long 
term parameters.   

     At first, unit root methods were carried to confirm if the data is stationary. The regression outcomes will not be valid 
unless the time series data is stationary. There are several methods for testing the stationarity of the data, and each of these 
has been using widely in the modern econometric arena. However, this study used both Augmented Dickey-Fuller (ADF) 
(Dickey & Fuller, 1981)and Phillips-Peron (PP) (Phillips & Perron, 1988) unit root methods to verify the stationarity of the 
data and to cross-check the results. After that, the optimum number of lags was selected for conducting Johansen’s 
Cointegration test, and VECM as those determine whether there is any long-run linkage between the index of Dhaka stock 
exchange and foreign exchange reserve, exchange rate, and crude oil price.      



Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

www.cribfb.com/journal/index.php/ijfb                         Indian Journal of Finance and Banking                             Vol. 4, No. 1; 2020 

137 

 

     Maximum likelihood process is applied for examining the existence of a cointegrating vector for non-stationary time 
series data (Johansen & Juselius, 1990). However, the Johansen-Juselius technique is indirectly implemented in Vector Auto-
regression to assess the integrating association (Masuduzzaman, 2013) which is based on the equation (2): 

𝒛𝒕=𝑨𝟏𝒛𝒕−𝟏 + 𝑨𝟐𝒛𝒕−𝟐 + ⋯ ⋯ ⋯ + 𝑨𝒑𝒛𝒕−𝒑 + 𝜺𝒕                                 (𝟐) 

Where, 𝒛𝒕 is a “n” vector of I(1) variables, 𝜺𝒕 is an innovation vector. However, VAR can be rewritten as follows:   

∆𝒛𝒕 = ∏ 𝒛𝒕−𝟏 +  ∑ Γ𝒊

𝒑−𝟏

𝒊=𝟏

∆𝒛𝒕−𝒊 + 𝜺𝒕 

Where, 

∏ = ∑ 𝐴𝑖

𝑝

𝑖=1

− 𝐼 

 

 
and Γ𝑖 =  − ∑ 𝐴𝑗

𝑝

𝑗=𝑖+1

 

 

     Here, ∏ matrix discloses disequilibrium adjustment and Γ matrix indicates an adjustment of short-run dynamic. 
Johansen & Juselius (1990) propose two tests to check hypothesis viz.   

𝑇𝑟𝑎𝑐𝑒 𝑇𝑒𝑠𝑡 = 𝜆𝑡𝑟𝑎𝑐𝑒 = −𝑇 ∑ 𝑙𝑛(1 − �̂�𝑗)

𝑘

𝑗=𝑟+1

 

𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝐸𝑖𝑔𝑒𝑛 𝑉𝑎𝑙𝑢𝑒 𝑇𝑒𝑠𝑡 =  𝝀𝒎𝒂𝒙 =  −𝑇 𝑙𝑛(1 − �̂�𝑟+1) 

Where, T= Size of sample and �̂�𝑗= Eigenvalues. 

Equation (3) shows the estimation of VECM with Dhaka stock exchange index as the target variable: 

∆ log 𝐷𝑆𝐸𝐼𝑡 = 𝛼0 + 𝛼1𝑡 ∑ ∆𝑙𝑜𝑔𝐷𝑆𝐸𝐼𝑡−1

𝑝

𝑖=1

+ 𝛼2𝑡 ∑ ∆𝑙𝑜𝑔𝐹𝐸𝑅𝑡−1

𝑝

𝑖=1

+ 𝛼3𝑡 ∑ ∆𝑙𝑜𝑔𝐸𝑅𝑡−1

𝑝

𝑖=1

+ 𝛼4𝑡 ∑ 𝑙𝑜𝑔𝐶𝑂𝑃𝑡−1

𝑝

𝑖=1

+ 𝛿1𝐸𝐶𝑇𝑡−1 + 𝜇1𝑡                                                                                (3) 

     Where, 𝛼1 to 𝛼4 denote elasticity of the short run, 𝐸𝐶𝑇𝑡−1 depicts error correction term, and 𝛿1 is the coefficient of 
ECT which depicts the swiftness of modification in the equilibrium if there is any shock in the system.  

     However, the Cointegration test only estimates whether the variables are correlated or not, but it does not specify any 
information about the path of their causality (Hossain, Hossain, & Sadi, 2013). Granger causality test identifies whether the 
historical values of one variable significantly affect in determining the forthcoming value of another variable. This study tries 
to inspect whether the selected macroeconomic factors help predict the fluctuation of the DSE index. Besides, it attempts to 
inspect whether the selected macroeconomic variables are affected by the fluctuations of the DSE index. Finally, impulse 
response function and variance decomposition analysis were applied to reveal some insights about the variables. For validating 
the selected model and ensuring its accuracy, this study applies heteroskedasticity and autocorrelation test. 

 
4. Results and Arguments 
The outcomes of the unit root method set out in table 1 demonstrate that both in the ADF and PP unit root test, all the 
variables are non-stationary at the level. However, the variables are stationary at first difference both in ADF and PP unit root 
method, indicating the refutation of the null hypothesis that evidenced the nonexistence of unit root issue in variables at I(1). 

Table 1. Results of Unit Root Test 

 
Variables 

 

 
From 

ADF PP 

C CT C CT 

t-stat t-stat t-stat t-stat 

LOGDSEI I (0) -2.4214 -2.1311 -2.4297 -2.1436 

I (1) -11.4345*** -11.4997*** -11.4345*** -11.4997*** 

LOGFER I (0) -2.0635 -2.2474 -1.7199 -1.4113 

I (1) -3.4092** -3.8031** -16.9305*** -17.0957*** 

LOGER I (0) -1.2528 -1.8446 -1.1579 -1.6675 



Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

www.cribfb.com/journal/index.php/ijfb                         Indian Journal of Finance and Banking                             Vol. 4, No. 1; 2020 

138 

 

Source: Researchers' calculation 

     Table 2 exhibits the outcomes of lag selection norms where the FPE and AIC recommend lag 3 as optimum; however, 
SC and HQ signpost lag 2 as optimum. Moreover, the sequential modified LR test statistic indicates 10 as the optimum 
number of lags. This study adopted the AIC for selecting the optimum number of lag and used lag 3 as optimum lag for 
further studies.  

Table 2. Choosing of Lags 

Lag LogL LR FPE AIC SC HQ 

0 585.5250 NA 1.15e-09 -9.230556 -9.140515 -9.193975 

1 1416.823 1596.620 2.76e-15 -22.17179 -21.72159 -21.98889 

2 1492.146 139.8859 1.08e-15 -23.11343 -22.30306* -22.78420* 

3 1515.961 42.71574 9.54e-16* -23.23748* -22.06695 -22.76193 

4 1527.525 20.00750 1.03e-15 -23.16706 -21.63637 -22.54519 

5 1536.454 14.88128 1.16e-15 -23.05482 -21.16397 -22.28663 

6 1554.249 28.52830 1.13e-15 -23.08331 -20.83230 -22.16879 

7 1563.920 14.89060 1.27e-15 -22.98286 -20.37168 -21.92201 

8 1582.394 27.27064 1.24e-15 -23.02212 -20.05078 -21.81496 

9 1606.342 33.83190 1.12e-15 -23.14829 -19.81678 -21.79480 

10 1627.403 28.41507* 1.06e-15 -23.22861 -19.53694 -21.72880 

Note: * designates lag order selected by the criterion.  

Source: Researchers' calculation 

     The evaluations of the Johansen Cointegrating approach are depicted in table 3 to check the long-run connection of 
the variables. This test is much responsive to slight changes to the lag length, so this study uses one lag less than the optimum 
lag length selected by the information criteria. 
 
Table 3. Results of the Johansen Cointegrating Test 

CE(s)  Trace Statistics 

(𝝀𝒕𝒓𝒂𝒄𝒆) 

5% Critical Value Max Eigen 

Statistic (𝝀𝒎𝒂𝒙) 

5% Critical Value 

r=0*  62.46252 47.85613 34.52087 27.58434 

r≤1  27.94165 29.79707 14.86852 21.13162 

r≤2  13.07313 15.49471 9.420578 14.26460 

r≤3  3.652552 3.841466 3.652552 3.841466 

 Note: r denotes the number of cointegrating linkages, CE(s) indicates cointegrating equations and * 
signifies refusal of the hypothesis at the 0.05 level.  

Source: Researchers' calculation 

     The result identifies that the null hypothesis (no cointegration between variables) cannot be accepted up to the level 
of zero both in trace statistics and max Eigen statistics at a 5% level of significance. It infers at least one cointegrating equation 
that exists between the index of the DSE and other selected macroeconomic factors.  

Table 4 shows the coefficient of the long-run model with t statistics.   
 
Table 4. Long-Run Model 

Exploratory Variables Coef. Std. Err. t-stat 

Constant -5.925886 - - 

Foreign Exchange Reserve (LOGFER) -0.439875 0.10268 -4.28395*** 

Exchange Rate (LOGER) 2.688054 0.78237 3.43576*** 

Crude Oil Price (LOGCOP) -0.514410 0.10774 -4.77445*** 

Note: ***, ** and * represent 1%, 5% and 10% level of significant respectively. 

Source: Researchers' calculation 

I (1) -7.3832*** -7.3678*** -7.3351*** -7.3188*** 

LOGCOP I (0) -2.3052 -2.4450 -2.3404 -2.3415  

I (1) -7.7746*** -7.7378*** -7.4110*** -7.3694*** 

Note: *** and ** represent 1% and 5% level of significant respectively, C and CT refer to constant and constant plus trend 
respectively. 



Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

www.cribfb.com/journal/index.php/ijfb                         Indian Journal of Finance and Banking                             Vol. 4, No. 1; 2020 

139 

 

     Based on cointegrating outcomes, this study derives the following long-run relationship among variables: 

𝐿𝑂𝐺𝐷𝑆𝐸𝐼 =  −0.439875𝐿𝑂𝐺𝐹𝐸𝑅 + 2.688054𝐿𝑂𝐺𝐸𝑅 − 0.514410𝐿𝑂𝐺𝐶𝑂𝑃 − 5.925886 

     Foreign exchange reserve is identified to have a significantly negative (p<0.01) relationship with the Dhaka stock 
exchange index. This result is coherent with the theory of opportunity and social cost (Mezui & Duru, 2013) which postulates 
the negative impact of the excess reserve on the economy of a country. Bangladesh is an agrarian country, and till 31st October 
2019, it has a total reserve of 32437.7 million U.S dollar (Bangladesh Bank, 2019) and by using this, it can meet it's up to 
seven months import payment which is four months extra than usual. Nevertheless, Bangladesh Bank does not invest those 
excess reserves in any productive sector; thus, the opportunity cost is swelling day by day. Generally, most of the foreign 
currency of Bangladesh comes from remittance, and most of this through whom is earned is the working class. The local 
beneficiary uses that remittance in different unproductive sectors, e.g., building the house, purchasing land, flat and consumer 
goods. As a result, this excess reserve is not using for productive purposes and thus creating opportunity cost. Due to lower 
import payment, the excess reserve is increasing, which is not good news for the economy of Bangladesh, thus reducing the 
country's production and enfeebling the stock market simultaneously. However, these findings are conflicting with (Abakah 
& Abakah, 2016; Akinlo, 2015; Hasan, 2018;  Ray, 2012) where they concentrated only on reserve and did not pay attention 
to excess reserve and recognized that foreign exchange reserve has a noteworthy positive influence on stock index. 

     The Exchange rate occupies a positive and significant (p<0.01) association with the index of the Dhaka stock 
exchange, which supports the flow-oriented theory of Dornbusch and Fischer (1980). Currently, an upsurge in the exchange 
rate of Bangladeshi taka (devaluing the money) in contradiction of the U.S. dollar encourages native exporters; thus, an 
expansion of export upsurges company’s profit. Higher profit hints to an increase in the value of a firm and thereby increase 
the stock index. This finding is consistent with (Giri & Joshi, 2017; Keat, Ling, Yi, & Yee, 2017; Kibria et al., 2014). 
However, an opposite outcome is identified by (Ali, 2013; Chauque & Rayappan, 2018; Hsing, 2014; Khan & Khan, 2018; 
Khan & Yousuf, 2013) who elucidate that increase of exchange rate (currency devaluation) forces the import dominated 
industry to raise the price of the products. This action negatively influences the cash inflow and thereby decreases the firm's 
profit, which negatively impacts the stock index. Besides, the stock oriented theory of Frankel (1993) states that if some 
stocks give a high return, investors tend to purchase those stocks by selling foreign stock, which also appreciates local currency, 
and thus exchange rate demonstrates a negative relationship with stock index.   

     The result ascertains a significant long-run negative association (p<0.01) between Dhaka stock exchange and crude 
oil price. This finding is certainly consistent with (Dhaoui & Kheraief, 2014; Giri & Joshi, 2017; Kamande, 2015) where it 
has been concluded that oil-importing country faces high importing cost due to raises of crude oil price. As Bangladesh is an 
oil-importing country, the increased price of crude oil in the global market puts pressure on the country’s economy. If the 
price of crude oil in the international market rises, the losses in the domestic market increases. The price has to be adjusted in 
the local market, which has a negative impact on all types of products. Besides, crude oil is used not only in transportation, 
agriculture, and production of goods as factory fuel but also in electricity generation. It may lead to upsurges in the price of 
both agricultural and manufacturing products of the country and negatively influences the stock market by decreasing the 
stock index. However, this result is inconsistent with (Aigbovo & Izekor, 2015; Khan & Yousuf, 2013) who investigated a 
long-run positive linkage between crude oil price and stock index where the business cycle and the global economic boom in 
energy, industrial and material sectors were thought to be the primary motives.  

     Table 5 represents the results of the error correction mechanism. As this study has a coefficient of error correction 
term -0.099064, it signposts a 9.91 percent speed of adjustment, which is significant at a 1 percent level. Finally, it postulates 
that if there is an exogenous shock, it will make a 9.91 percent adjustment per month to reach in the long-run equilibrium. 

Table 5. Results of VECM 

Exploratory Variables Coef. Std. Err. t-stat 

Speed of Adjustment -0.099064 0.037143 -2.667100*** 

∆𝐿𝑂𝐺𝐷𝑆𝐸𝐼𝑡−1 0.001053 0.088684 0.011875 

∆𝐿𝑂𝐺𝐷𝑆𝐸𝐼𝑡−2 0.051320 0.087986 0.583270 

∆𝐿𝑂𝐺𝐹𝐸𝑅𝑡−1 0.062330 0.203014 0.307024 

∆𝐿𝑂𝐺𝐹𝐸𝑅𝑡−2 0.090643 0.203210 0.446059 

∆𝐿𝑂𝐺𝐸𝑅𝑡−1 -0.695179 1.063916 -0.653415 

∆𝐿𝑂𝐺𝐸𝑅𝑡−2 2.175796 1.044064 2.083967** 

∆𝐿𝑂𝐺𝐶𝑂𝑃𝑡−1 0.082852 0.078487 1.055608 

∆𝐿𝑂𝐺𝐶𝑂𝑃𝑡−2 -0.067637 0.075673 -0.893807 

Constant -0.000498 0.003602 -0.138181 

Note: *** and ** represent 1% and 5% and level of significant respectively.  

Source: Researchers' calculation 

     The outcomes of table 6 show the effect of both heteroskedasticity and Breusch-Godfrey Serial Correlation LM test, 
where the prob. value of Chi-Square is 0.1017 in heteroskedasticity test that is more than 0.05; hence this study confirms the 



Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

www.cribfb.com/journal/index.php/ijfb                         Indian Journal of Finance and Banking                             Vol. 4, No. 1; 2020 

140 

 

presence of homogeneity in residuals. On the other hand, in the LM test, the prob. value of Chi-Square is 0.2589, which is 
more than 0.05; hence this study also confirms the absence of serial correlation in our model.   

Table 6. Results of Heteroskedasticity and Serial Correlation Test 

Heteroskedasticity Test: ARCH 

F-statistic 2.313901 Prob. F(2,128) 0.1030 

Obs*R-squared 4.571003 Prob. Chi-Square(2) 0.1017 

Serial Correlation Test 

F-statistic 1.254885 Prob. F(2,121) 0.2888 

Obs*R-squared 2.702616 Prob. Chi-Square(2) 0.2589 

Source: Researchers' calculation 

     The empirical results of table 7 demonstrate that there stands a causal bond between Dhaka stock exchange index, 
foreign exchange reserve and exchange rate, moving only in one way from Dhaka stock exchange index to foreign exchange 
reserve and the exchange rate which indicate that if there is an upsurge or decline in the index of DSE, it will affect in foreign 
exchange reserve and exchange rate. However, no causal linkage is identified between Dhaka stock exchange indexes and crude 
oil prices.  

    Table 7. Results of Pairwise Granger Causality Tests 

Causality Direction F-Statistic Prob. 

LOGFER ~ LOGDSEI 0.43723 0.6468 

LOGDSEI → LOGFER 15.9099*** 7.E-07 

LOGER ~ LOGDSEI 0.75989 0.4698 

LOGDSEI → LOGER 3.37123*** 0.0374 

LOGCOP ~ LOGDSEI 0.87809 0.4180 

LOGDSEI ~ LOGCOP 0.66622 0.5154 

Note: *** represents 1% level of significant. 

Source: Researchers' calculation 

     Figure 1 represents the shape of the IRF of the Dhaka stock exchange index to an innovation in foreign exchange 
reserve, exchange rate, and crude oil price. Along with their impacts, this figure forecasts 36 months or 3 years of data plotted 
on IRF. The figure shows, a one standard deviation innovation to its own stock index indicates a substantial decrease in the 
Dhaka stock exchange index primarily. However, this impact decreases with time and settles at a permanent level of 0.011 
units above the baseline. The impulse response of the Dhaka stock exchange index of 1 percent increased shocks coming from 
foreign exchange reserve leads to a permanent increase in the Dhaka stock exchange index of 0.024 from the 24th month, 
indicating a long-term positive relationship. A one standard deviation innovation to exchange rate initially decreases the Dhaka 
stock exchange index, and after the second month onward, it increases and settles at 0.001 units above the baseline at a 
permanent level from 15th month. Lastly, a one SD innovation to crude oil price causes the Dhaka stock exchange index 
primarily to fluctuate positively, and it goes to a permanent level from 15th month at 0.011 units above the baseline. So, in 
the long run, all the selected macroeconomic factors have a permanent consequence on the Dhaka stock exchange index.  

.00

.01

.02

.03

5 10 15 20 25 30 35

Response of LOGDSEI to LOGDSEI

.00

.01

.02

.03

5 10 15 20 25 30 35

Response of LOGDSEI to LOGFER

.00

.01

.02

.03

5 10 15 20 25 30 35

Response of LOGDSEI to LOGER

.00

.01

.02

.03

5 10 15 20 25 30 35

Response of LOGDSEI to LOGCOP

Response to Cholesky One S.D. (d.f. adjusted) Innovations

 

 Figure 1. Impulse Response Function (IRF) for Dhaka Stock Exchange Index 
Source: Researchers' computation 



Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

www.cribfb.com/journal/index.php/ijfb                         Indian Journal of Finance and Banking                             Vol. 4, No. 1; 2020 

141 

 

     Table 8 limns the consequences of variance decomposition analysis up to the 20 months, where it shows how much 
of the Dhaka stock exchange index’s individual innovation is explicated by movement in its individual variance along with 
foreign exchange reserve, exchange rate, and crude oil price. The results show 20 months of forecast where it demonstrates 
that most of the Dhaka stock exchange index variations are elucidated by itself. In the 5th month, the Dhaka stock exchange 
index explains 92.56 percent of variation by shocks to itself, 3.44 percent by foreign exchange reserve, 1.18 percent by the 
exchange rate, and 2.81 percent by the crude oil price. In the 10th month, the Dhaka stock exchange index explains 76.95 
percent of variation by shocks to itself, 14.41 percent, 1.19 percent, and 7.44 percent by foreign exchange reserve, exchange 
rate, and crude oil price respectively. So, it can be marked that in the long run, the oscillation in the Dhaka stock exchange 
index can be caused by the shock in foreign exchange reserve, exchange rate, and crude oil price.  

Table 8. Variance Decomposition of LOGDSEI 

Period S.E. LOGDSEI LOGFER LOGER LOGCOP 

1 0.031080 100.0000 0.000000 0.000000 0.000000 

2 0.042843 98.42375 0.326756 0.209064 1.040433 

3 0.051253 97.02071 1.247897 0.432370 1.299023 

4 0.057761 94.87791 2.300499 0.906790 1.914802 

5 0.062980 92.56092 3.441694 1.182578 2.814805 

6 0.067530 89.87707 5.007936 1.285298 3.829694 

7 0.071705 86.93744 6.904513 1.304643 4.853406 

8 0.075712 83.69416 9.218202 1.282202 5.805436 

9 0.079590 80.36040 11.72187 1.243664 6.674061 

10 0.083417 76.95072 14.41125 1.194920 7.443114 

11 0.087172 73.62932 17.09730 1.143225 8.130148 

12 0.090899 70.38924 19.78931 1.089306 8.732140 

13 0.094568 67.32482 22.37082 1.036709 9.267651 

14 0.098209 64.41157 24.86704 0.985325 9.736061 

15 0.101792 61.69762 27.21236 0.936924 10.15309 

16 0.105340 59.15295 29.43719 0.891030 10.51883 

17 0.108829 56.79925 31.50659 0.848450 10.84571 

18 0.112276 54.60755 33.44992 0.808625 11.13391 

19 0.115663 52.58553 35.24971 0.771888 11.39288 

20 0.119003 50.70796 36.93158 0.737727 11.62273 

Source: Researchers' calculation 
5. Conclusion 
This study attempts to visualize the empirical linkage between the Dhaka stock exchange index and three designated 
macroeconomic factors, namely foreign exchange reserve, exchange rate, and crude oil price with monthly time series data 
ranging from July 2008 to October 2019. ADF and PP unit root tests were applied to ensure the same integrating order of 
the variables and for cross-checking of the outcomes. The Johansen Cointegration test was applied and found a long-run 
equilibrium linkage between the Dhaka stock exchange index, foreign exchange reserve, exchange rate, and crude oil price. It 
was observed that both foreign exchange reserves and crude oil prices have a significant negative consequence on the Dhaka 
stock exchange index in the long run, and the exchange rate was found to have a positive influence on the Dhaka stock 
exchange index, which was also statistically significant. This result confirms that the Dhaka stock exchange index can be 
projected by past data. The coefficient of ECT exposed that a 9.91 percent of disequilibrium in the long-run model is rectified 
per month as the Dhaka stock exchange index goes back to its equilibrium. The results of the Granger Causality test divulged 
a causal association running in one direction from the Dhaka stock exchange index to foreign exchange reserve and exchange 
rate, indicating that if there is any change in the index of DSE, it will affect in foreign exchange reserve and exchange rate. 
The impulse response function supported the long-run model of the study, and the outcomes of variance decomposition 
specified that the Dhaka stock exchange index is determined by foreign exchange reserve, exchange rate, and crude oil price.   

     From the above discussion, it can be said that the Bangladesh government, policymakers, and economists should be 
cautious in adopting any new economic policy as these selected macroeconomic factors have a significant consequence on the 
capital market of Bangladesh. For keeping the stock market afloat, the authorities need to be aware of economic progress, 
sound fiscal policy, and, above all, proper management of different macroeconomic factors that influence the stock market. 
Since the stock index has a deep connection with the exchange rate, and both importers and exporters are affected by it, the 
government should try to stabilize the exchange rate. The authority needs to adopt a policy that will stabilize the value of 
Bangladeshi taka against the US dollar, and at the same time, prevent unforeseen fluctuations in the stock market.  Bangladesh 
is an oil-importing country, so it is time to think about how to reduce the dependence on crude oil and increase domestic 
energy production and become more dependent on renewable energy. The use of locally produced and renewable energy can 
improve the dynamics of the stock market by reducing the cost of production. This paper also discusses the detrimental 



Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

www.cribfb.com/journal/index.php/ijfb                         Indian Journal of Finance and Banking                             Vol. 4, No. 1; 2020 

142 

 

aspects of excess reserve that may create social and opportunity cost. The government should keep enough money from the 
foreign exchange reserve, and the rest should be used for filling the funding gap of the infrastructure of the country. 

     However, this study has uncovered some new avenues of future research for investigators. Only the detrimental aspects 
of the excess reserve are discussed, but no guidance has been given on how these excess reserves can be used in some productive 
sectors. Regarding the exchange rate, both positive and negative aspects of currency appreciation and devaluation are 
mentioned, but no guideline mentioned on what degree of appreciation and devaluation may be useful for the economy of a 
country like Bangladesh. So, there is a considerable scope to cover this area in the future.  

 
References 
Abakah, E. J. A., & Abakah, M. K. (2016). Foreign Exchange Reserve and Its Impact on Stock Market: Evidence from 

Ghana. Journal of Finance and Economics, 4(5), 136–141. https://doi.org/10.12691/jfe-4-5-2 
Aigbovo, O., & Izekor, A. O. (2015). The impact of macroeconomic variables on stock market index in Nigeria. African 

Journal of Management Sciences, 1(1), 18–40. 
Akinlo, O. O. (2015). Impact of Foreign Exchange Reserves on Nigerian Stock Market. The International Journal of 

Business and Finance Research, 9(2), 69–76. Retrieved from 
https://econpapers.repec.org/RePEc:ibf:ijbfre:v:9:y:2015:i:2:p:69-76 

Ali, M. B. (2013). The Impact Of Selected Micro And Macroeconomic Variables On The Stock Prices Of Emerging Stock 
Market : A Case Of Dhaka Stock Exchange ( DSE ). Dhaka. 

Bangladesh Bank. (2019). Monthly Economic Trend. https://www.bb.org.bd/econdata/index.php 
Chauque, D. F. F., & Rayappan, P. A. (2018). The Impact of Macroeconomic Variables on Stock Market Performance: A 

Case of Malaysia. Edelweiss Applied Science and Technology, 2(1), 100–104. 
https://doi.org/10.33805/2576.8484.122 

Dhaoui, A., & Kheraief, N. (2014). Empirical Linkage between Oil Price and Stock Market Returns and Volatility: Evidence 
from International Developed Markets. Economics Discussion Papers. 

Dickey, D. A., & Fuller, W. A. (1981). Likelihood Ratio Statistics for Autoregressive Time Series with a Unit Root. 
Econometrica, 49(4), 1057–1072. https://doi.org/10.2307/1912517 

Dornbusch, R., & Fischer, S. (1980). Exchange Rates and the Current Account. American Economic Review, 70(5), 960–
971. Retrieved from https://econpapers.repec.org/RePEc:aea:aecrev:v:70:y:1980:i:5:p:960-71 

Fauziah, F., Moeljadi, M., & Ratnawati, K. (2015). Dynamic Relationship Between Exchange Rates and Stock Prices in Asia, 
2009-2013. Journal of Economics Finance and Accounting, 2(1), 124–134. 

Frankel, J. (1993). Monetary and Portfolio-Balance Models of Exchange Rate Determination. In Economic Interdependence 
and Flexible Exchange Rates. Retrieved from http://www.hks.harvard.edu/fs/jfrankel/Monetary&PB Models 
ExRateDetermtn.pdf 

Giri, A. K., & Joshi, P. (2017). The impact of macroeconomic indicators on Indian stock prices: An empirical analysis. 
Studies in Business and Economics, 12(1), 61–78. https://doi.org/https://doi.org/10.1515/sbe-2017-0005 

Hasan, M. T. (2018). Impact of Foreign Exchange Reserve on Stock Market Development: Evidence from Dhaka Stock 
Exchange. Journal of Business Studies, 39(2), 19–30. 

Hossain, M. K., Hossain, A., & Sadi, R. (2013). An Examination of the Relationship between Stock Market and Economic 
Growth: A Study in Malaysia. Journal of Transformative Entrepreneurship, 1(2), 124–133. 

Hosseini, S. M., Ahmad, Z., & Lai, Y. W. (2011). The Role of Macroeconomic Variables on Stock Market Index in China 
and India. International Journal of Economics and Finance, 3(6), 235–243. 

Hsing, Y. (2014). Impacts of Macroeconomic Factors on the Stock Market in Estonia. Journal of Economics and 
Development Studies, 2(2), 23–31. 

International Monetary Fund. (2019). Retrieved from https://www.imf.org/en/Research/commodity-prices 
Johansen, S., & Juselius, K. (1990). Maximum Likelihood Estimation and Inference on Cointegration — With Applications 

To The Demand For Money. Oxford Bulletin of Economics and Statistics, 52(2), 169–210. 
https://doi.org/10.1111/j.1468-0084.1990.mp52002003.x 

Kamande, M. N. (2015). Macroeconomic Variables and Stock Market Return in Nairobi Securities Exchange. Nairobi. 
Keat, L. Y., Ling, N. Y., Yi, S. C., & Yee, T. L. (2017). The Relationship between Macroeconomic Variables and Stock 

Market Performance in Thailand. Thailand. 
Khalid, W., & Khan, S. (2017). Effects of Macroeconomic Variables on the Stock Market Volatility: The Pakistan 

Experience. International Journal of Econometrics and Financial Management, 5(2), 42–59. 
https://doi.org/10.12691/ijefm-5-2-4 

Khan, J., & Khan, I. (2018). The Impact of Macroeconomic Variables on Stock Prices: A Case Study of Karachi Stock 
Exchange. Journal of Economics and Sustainable Development, 9(13), 15–25. 
https://doi.org/10.5901/mjss.2012.v3n3p295 

Khan, M. M., & Yousuf, A. S. (2013). Macroeconomic Forces and Stock Prices: Evidence from the Bangladesh Stock Market. 
Khan, M. T., Khan, A., Ahmad, A., & Bashir, U. (2018). Impact of macroeconomic factors on stock returns of KSE 100 

Index. Journal of Business and Tourism, 04(01), 133–145. 
Kibria, U., Mehmood, Y., Kamaran, M., Arshad, M. U., Perveen, R., & Sajid, M. S. (2014). The Impact of Macroeconomic 

Variables on Stock Market Returns: A Case of Pakistan. Research Journal of Management Sciences, 3(8), 1–7. 



Copyright © CC-BY-NC 2020, CRIBFB | IJFB 

 

www.cribfb.com/journal/index.php/ijfb                         Indian Journal of Finance and Banking                             Vol. 4, No. 1; 2020 

143 

 

Maheshwari, A., Upamannyu, N., Bhakuni, P., & Saban, A. (2013). Impact of wpi, exchange rate, gold rate, forex and market 
capitalization on NSE returns. International Monthly Refereed Journal of Research in Management & Technology, 
(Special Issue), 20–29. 

Masuduzzaman, M. (2013). Impact of Macroeconomic Variables on the Stock Market Returns in Bangladesh: Does a 
Meaningful Impact Exist? Ministry of Finance Finance Division, 1–17. 

Mezui, C. A. M., & Duru, U. (2013). Holding Excess Foreign Reserves Versus Infrastructure Finance : What Should Africa 
Do ? 

Miller, J. I., & Ratti, R. A. (2009). Crude oil and stock markets: Stability, instability, and bubbles. Energy Economics, 31(4), 
559–568. https://doi.org/https://doi.org/10.1016/j.eneco.2009.01.009 

Ndlovu, B., Faisal, F., Resatoglu, N. G. R., & Türsoy, T. T. (2018). The Impact of Macroeconomic Variables on Stock 
Returns: A Case of the Johannesburg Stock Exchange. Romanian Statistical Review, 87–104. 
https://doi.org/10.18032/kaaba.2017.30.1.33 

Phillips, P. C. B., & Perron, P. (1988). Testing for a Unit Root in Time Series Regression. Biometrika, 75(2), 335–346. 
https://doi.org/10.2307/2336182 

Ray, S. (2012). Foreign Exchange Reserve and its Impact on Stock Market Capitalization : Evidence from India. 2(2), 46–
61. 

Shafi, K., Hua, L., Idrees, Z., & Nazeer, A. (2015). Oil prices & stock market : Evidence from KSE & BSE. American Journal 
of Business, Economics and Management, 3(2), 40–44. 

Zaidi, I., Ahmed, R. M. A., & Siok, K. S. (2017). Examining the relationship between economic growth, energy consumption 
and Co2 emission using inverse function regression. Applied Ecology and Environmental Research, 15(1), 473–
484. https://doi.org/10.15666/aeer/1501_473484 

 
 
Copyrights  
Copyright for this article is retained by the author(s), with first publication rights granted to the journal. This is an open-
access article distributed under the terms and conditions of the Creative Commons Attribution license 
(http://creativecommons.org/licenses/by/4.0/). 


