The Lahore Journal of Economics

14 : 1 (Summer 2009): pp. 115-137

An Empirical Investigation of the Causal Relationship among

Monetary Variables and Equity Market Returns

Arshad Hasan * and M. Tariq Javed **

Abstract

This study explores the long-term dynamic relationship between

equity prices and monetary variables for the period June 1998 to June
2008. Monetary variables include money supply, treasury bill rates, foreign
exchange rates, and the consumer price index. The data have been
examined using multivariate cointegration analysis and Granger causality
analysis. Johansen and Juselius’ multivariate cointegration analysis
indicates the presence of a long-term dynamic relationship between the
equity market and monetary variables. Unidirectional Granger causality is
found between monetary variables and the equity market. In the case of
money supply, a positive relationship supports the liquidity hypothesis.
Impulse response analysis indicates that the interest rate shock has a
negative impact on equity returns in the Pakistani equity market.
Exchange rates also have a negative impact on equity returns in the short
run. However inflation has little impact on returns in the equity market.
Variance decomposition analysis suggests that the interest rate, exchange
rate, and money supply shocks are a substantial source of volatility for
equity returns. The contribution of a monetary shock to the equity returns
ranges from 4% to 16% over different time lags. Similarly, the VECM also
confirms the presence of a short-term relationship between monetary
variables and equity returns. This state of affairs demands that monetary
variables be considered an important factor in determining stock market
movements. Policymakers should be more vigilant and careful in designing
monetary policies as it has a direct impact on cash inflows into the capital
market and on the stability of the capital market.

JEL Classification: E31, G12.

Keywords: Monetary variables, equity, causality, Pakistan.

*
Muhammad Ali Jinnah University, Islamabad, Pakistan.
**
Quaid-e-Azam University, Islamabad, Pakistan.
116 Arshad Hasan and M. Tariq Javed

1. Introduction

The causal relationship between monetary variables and equity

returns has been one of the most debated topics in finance during the last
few decades. Equity prices are the most closely observed asset prices in an
economy and are considered the most sensitive to economic conditions; high
volatility or abnormal movements in equity prices from fundamental values
can have adverse implications for the economy. Thus, it becomes imperative
to understand the relationship and dynamics of monetary variables and
equity market returns.

A number of studies have been conducted to investigate the potential

response of equity prices to a change in monetary variables. Jaffe and
Mandelker (1976) Fama and Schwert (1977), Nelson (1976) Chan, Chen and
Hsieh (1985), Chen, Roll and Ross (1986), Burnmeister and Wall (1986),
Burmeister and MacElroy (1988), Chang and Pinegar (1990), Defina (1991)
Kryzanowski and Zhang (1992), Chen and Jordan (1993), Sauer (1994), and
Rahman, Coggin and Lee (1998) explore the relationship between inflation
and equity prices. Kryzanowski and Zhang (1992), Sauer (1994), and
Mukherjee and Naka (1995) explore the relationship between the foreign
exchange rate and equity market returns. Burmeister and MacElroy (1988)
study the relationship between short-term interest rates and equity market
returns. Studies that explore the relationship between money supply and
equity market returns include Friedman and Schwartz (1963), Hamburger and
Kochin (1972) and Kraft and Kraft (1977) Beenstock and Chan (1988), Nozar
and Taylor (1988), Sauer (1994), and Mukherjee and Naka (1995).

Financial liberalization and globalization provide further impetus for

exploration of the subject, especially in the context of emerging markets such
as Pakistan. This interrelationship has an economic rationale as discounted
cash flow techniques for asset pricing assume that stock prices reflect
expectations about futures cash flows. These expectations about cash flows are
based on the expected performance of the corporate sector; the performance
of the corporate sector is influenced by changing patterns in monetary
variables. Therefore, any innovation in monetary variables will affect corporate
profits and will ultimately reflect asset prices. If an asset pricing mechanism is
efficient and reflects precisely the fundamentals of the corporate sector, then
equity prices can serve as a leading indicator of future dimensions of
economic activity. The efficient market hypothesis also provides that prices
instantly adjust to the arrival of new information and the current prices of
securities reflect all the information available about the security. Thus asset
prices generally react sensitively to the arrival of new information. This
response is neither equal nor homogeneous across all economic changes, and
it becomes imperative to investigate the interactions among monetary factors
 Causal Relationship among Monetary Variables and Equity Returns 117

and equity prices since it will provide the foundations for the formulation of
monetary policy in the country.

The influence of monetary variables on equity returns has attracted

considerable attention in both developed and developing countries. Many
studies have been conducted to find the long-term equilibrium relationship
between stock returns and monetary variables for the USA, Japan, and other
industrially developed countries. This study focuses on Pakistan as a rapidly
growing market in South Asia. The Pakistani equity market has shown
tremendous growth in the last few years: The KSE-100 index rose from
1,773 index points in January 2000 to 15,125 in March 2008. This
phenomenal growth has also attracted foreign investors and portfolio
investment has increased four-fold. In the current economic scenario, it is
necessary to explore the relationship between monetary variables and equity
returns so that the current dynamics of monetary policy can be examined.

This study examines the long-run dynamic relationship between

equity prices and four monetary factors, including the money supply,
treasury bill rates, foreign exchange rates, and inflation rates for the period
June 1998 to June 2008, using a multivariate cointegration analysis. We also
explore the short-term dynamics of equity prices using the VECM. These
variables are important as prior studies provide evidence that they have a
significant relationship with equity prices in several developed and emerging
markets. The study also investigates how the equity market responds to
innovations in monetary variables by using an impulse response function and
variance decomposition analysis.

The study will help understand the dynamics of equity market

activities in an emerging market by identifying monetary variables that affect
the equity market, and quantifying the impact of changes in monetary
variables on equity market movements. This will enable investors and
portfolio managers to make effective investment decisions. It will also
facilitate policymakers in the formulation of policies that will not only
encourage more capital inflows into the capital market but also provide it
with stability.

The paper is arranged in four sections. Section II briefly surveys the

empirical literature on the relationship among macroeconomic variables and
capital markets. Section III explains the methodology adopted and data
employed. The empirical results are discussed in Section IV and Section V
concludes the paper.
118 Arshad Hasan and M. Tariq Javed

2. Literature Review

The efficient market hypothesis provides that asset prices respond to

the arrival of new information. This response is stronger in the case of
certain economic events whereas in other cases it may be weaker. Empirical
studies try to identify factors that have a significant influence on equity
prices: monetary factors are no exception.

The relationship between equity market returns and exchange rates

has attracted the attention of academics and researchers during the last
decade due to significant changes in the financial world. This period is
known for the emergence of new capital markets, elimination of barriers to
capital flows and foreign exchange restrictions, and the adoption of flexible
exchange rate arrangements in emerging and transition countries. While
these attributes have opened the door for investment opportunities, they
have simultaneously increased the volatility of exchange rates and
contributed significantly to the overall risk associated with investment
decisions and portfolio diversification. Interaction between foreign exchange
and equity markets is now more complex and needs greater attention. It is
worth mentioning that no consensus exists among academics regarding the
presence of a relationship between stock prices and exchange rates, and the
direction of the relationship.

Bahmani and Sohrabian (1992) examine the relationship between the

exchange rate and equity market returns for the period 1963-1988 by
employing cointegration analysis and Granger causality analysis. The study
provides evidence of bidirectional causality in the short run. Yu (1997) finds
a bidirectional relationship between exchange rates and the equity market in
Japan and unidirectional causality flowing from changes in exchange rates to
changes in stock prices in Hong Kong. However no causality has been
observed in the daily time series of the Singapore market during 1983-1994.
Abdalla and Murinde (1997) also examine the relationship between the
exchange rate and equity prices in India, Pakistan, Korea and the Philippines
for the period 1985-1994 by employing cointegration analysis. The results
do not provide evidence of a causal relationship in Korea and Pakistan, but
do so of unidirectional causality between the exchange rate and equity
prices in India and the Philippines. In India, causality flows from exchange
rates to equity prices while in the Philippines, unidirectional causality runs
from the equity market to the exchange rate.

Muhammad and Rasheed (2003) explore the relationship between

exchange rates and equity prices in Pakistan, India, Sri Lanka, and
Bangladesh for the period 1994–2000. The results indicate that no
 Causal Relationship among Monetary Variables and Equity Returns 119

relationship exists between equity markets and foreign exchange rates in the
long or short run in India and Pakistan. However, bidirectional causality is
observed between exchange rates and equity markets in Bangladesh and Sri
Lanka. Stavárek (2005) examines the presence of causal relationships
between equity prices and effective exchange rates in Austria, France,
Germany, the UK, Czech Republic, Hungary, Poland, Slovakia, and United
States for the period 1970-2003. Results provide evidence of unidirectional
causality in the long run as well as short run. Results also indicate that this
causal relationship is stronger in developed markets, i.e., Austria, France,
Germany, the UK, and US. Moreover, the relationship is found stronger for
the period 1993–2003 than 1970–92.

Academics as well as professional observers have explored the

relationship between stock prices and various monetary variables that are
subjective to monetary policy. One such variable is money supply; initial
studies conducted in the 1960s and 1970s generally indicated a strong
leading relationship between money supply changes and equity prices.
However, subsequent studies have raised questions about the nature of this
relationship. They have confirmed the presence of a relationship between
money supply and stock prices but the timing of the relationship remains
debatable. Rozzef (1974) examines stock market efficiency with respect to
money supply by employing regression analysis and trading rule analysis and
finds that equity market returns do not lag behind money supply. The study
confirms EMH and provides that current equity returns incorporate all
information about historical as well as anticipated future changes in money
supply. Beenstock and Chan (1988) examine the relationship between equity
markets and a set of macroeconomic variables and provide evidence of a
positive relationship between equity returns and money supply and inflation.

The relationship between inflation, interest rates, and equity prices is

not direct and consistent. Equity prices are based on two factors: (i) discount
rate and (ii) expected cash flows. Interest rates and inflation affect both.
Different possibilities may exist with respect to expected equity prices. Equity
prices may be stable when an increase in interest rates is the result of an
increase in the rate of inflation and firms are able to increase prices in line
with cost increases. In such a situation, equity prices might not experience a
significant change as the negative effect of an increased discount rate is offset
by the increase in corporate earnings. However, equity prices may show some
negative trend when firms are not able to increase prices proportionately in
response to higher costs. However, the impact is most negative when the
required rate of return increases and expected cash flows decrease due to
inflation. The effect of interest rate changes on stock prices will depend on
what caused the change in interest rates and the effect of this event on
120 Arshad Hasan and M. Tariq Javed

expected cash flows on common stock. We cannot be certain whether this

change in cash flows will augment or offset the change in interest rates.
However, earlier studies like Nelson (1976), Mandelker (1976), Fama and
Schwert (1977), and Chen, Roll and Ross (1986) provide evidence of a
negative relationship between inflation and equity prices. The latter examines
the presence of a long-run relationship between equity prices and seven
macroeconomic variables for the US, which include monetary as well as real
sector variables. Monetary variables include inflation and interest rates. The
results provide evidence, during periods of a high volatility yield curve that
unanticipated inflation can explain expected returns.

Hamao (1988) uses the methodology proposed by Chen, Roll and

Ross (1986) for the Japanese economy and reveals that variations in expected
inflation and unexpected variations in the risk premium and the term
structure of interest rates influence equity returns significantly. However,
variations in macroeconomic activities are found to be weakly priced in
Japan in comparison with variations priced in the US.

Mukherjee and Naka (1995) explore the long-term relation between

equity prices in the Japanese stock market and six macroeconomic variables,
i.e., money supply, industrial production, exchange rate, inflation, long term
government bond rates, and the call money rate by employing monthly data
for the period 1/71 to 12/90. They employ a vector error correction model
(VECM) to investigate the relationship among these variables and provide
evidence of a positive relationship between equity prices and money supply,
exchange rate, and industrial production. However, the study is mixed with
reference to interest rates and inflation. Zhao (1999) explores the possibility
of long-run relationships among industrial production, inflation, and equity
prices in the Chinese capital market by employing monthly data for the
period 1/1993 to 3/1998. The study reveals a significant negative relationship
between equity prices and inflation, and also shows that industrial production
significantly influences equity prices in the Chinese economy while the
direction of this relationship is negative.

Nishat and Rozina (2001) analyze causal relationships between the

Karachi Stock Exchange Index and inflation, industrial production, narrow
money, and the money market rate by employing a vector error correction
model for the period 1/1973 to12/2004. Results indicate the presence of two
cointegrating equations among macroeconomic variables. Industrial
production and inflation are identified as the largest determinants of equity
prices in the Karachi Stock Exchange. Industrial production has a positive
relationship with equity prices whereas inflation is negatively associated with
 Causal Relationship among Monetary Variables and Equity Returns 121

stock prices. Granger causality is found flowing from macroeconomic variables

to stock price, as is industrial production.

Maysami and Koh (2000) examine long-term dynamic interactions

between the Strait Times Index (STI) and macroeconomic variables for the
period 1988 to 1995 by employing a vector error correction model (VECM).
The variables are seasonally adjusted money supply, industrial production
index, foreign exchange rate, retail price index (inflation), domestic exports,
and interest rates. Results indicate a cointegrating vector among returns on
the Strait Times Index (STI) and money supply growth, inflation, term
structure of interest rates, and changes in exchange rates. This study
investigates the long-term dynamic relationship among S&P 500, Nikkei 225
and STI by using cointegration analysis and finds that the equity markets of
the US, Japan and Singapore are co-integrated.
Hussain and Mahmood (2001) investigate the long-run causal
relationship between equity prices and macroeconomic variables for the
period 7/1959 to 6/1999 by employing a vector error correction
framework. Annual data for gross domestic product, consumption and
investment is analyzed and it is concluded that a long-run relationship
exists between equity prices and macroeconomic variables. Results also
reveal the presence of unidirectional causality flowing from macro variables
to stock prices. However, the equity market is not found to influence
aggregate demand so its movement cannot be termed as a leading
indicator of economic activity.
Mishra (2004) investigates the long-run dynamic causal relationship
between equity market and macroeconomic variables in India including the
foreign exchange rate, interest rate and demand for money for the period
1992 to 2002 by employing the vector auto regression (VAR) technique.
This paper applies the Granger causality test to monthly data to examine the
direction of the relationship and finds evidence of unidirectional causality
flowing from the foreign exchange rate to the interest rate and demand for
money. However, no Granger causality is found between equity returns and
exchange rate returns.
Akmal (2007) investigates the relationship between equity market
prices and inflation in Pakistan for the period 1971-2006 by employing the
autoregressive distributed lag (ARDL) approach to observe cointegration
among variables and provides evidence that equity returns are hedged
against inflation in the long run.
122 Arshad Hasan and M. Tariq Javed

3. Data Description and Methodology

We investigate the long-term dynamic interaction between the

Pakistani equity market and monetary variables by employing monthly data
for the period 6/1998 to 6/2008. The monetary variables we use include
money supply, consumer price index, interest rate and exchange rate. The
preference for monthly data is in line with earlier work done by Chan and
Faff (1998) to explore the long-run relationship between macroeconomic
variables and capital markets.
Stock Market Returns
Stock market returns have been calculated by using following equation

Rt = ln (Pt / Pt-1)
Where Rt is return for month ‘t’; and Pt and Pt-1 are closing values
of KSE- 100 Index for month ‘t’ and ‘t-1’ respectively.

Money Growth Rate

Narrow money (M1) is used as a proxy for money supply. The money
growth rate has been calculated by using the log difference for narrow money
(M1)
Money Growth Rate = ln (Mt / Mt-1)

Change in Interest Rate

Treasury bill rates are used as a proxy for the interest rate. Change
is measured by log difference to T bill rates.
Change in the Interest Rate = ln (TBt / TBt-1)

Change in Foreign Exchange Rate

The change in the foreign exchange rate is measured by employing
the end-of-month US $/Rs exchange rate and the change in value is worked
out through log differencing, i.e.,
Change in Foreign Exchange Rate = ln (FERt / FERt-1)

Where FER is the Foreign Exchange Rate US $/Rs

 Causal Relationship among Monetary Variables and Equity Returns 123

Inflation Rate
The consumer price index (CPI) is used as a proxy for inflation. The
CPI is a broad-based measure for calculating the average change in prices of
goods and services during a specific period.
Inflation Rate = ln (CPIt / CPIt-1)

Trend in Log of Macroeconomic Series

15
Index
10
M1
5 TBill
XRate
0
CPI
1 11 21 31 41 51 61 71 81 91 101 111 121
-5

There are several techniques for testing the long-term dynamic

interaction between prices in equity markets and macroeconomic variables.
In this study, we emphasize testing the relationship between monetary
variables and the Pakistani equity market, via:

• Descriptive statistics
• Correlation matrix
• Cointegration tests
• Granger causality test
• Impulse response analysis
• Variance decomposition analysis

The stationarity of data is tested using unit root tests. The null
hypothesis of a unit root is tested using the Augmented Dickey-Fuller (ADF)
Test and Phillips-Perron Test. The ADF test examines the presence of a unit
root in an autoregressive model. A basic autoregressive model is Zt = αZt-1 +
ut, where Zt is the variable studied, t is the time period, α is a coefficient,
and ut is the disturbance term. The regression model can be written as ΔZt
= (α - 1)Zt-1 + ut = δZt-1 + ut, where Δ is the first difference operator. Here,
testing for a unit root is equivalent to testing δ = 0.

The ADF tests assume that the error terms are statistically
independent and have a constant variance. This assumption may not be true
124 Arshad Hasan and M. Tariq Javed

of all the data used, and so the Phillip-Perron test is used to relax the above
assumptions and permit the error disturbances to be heterogeneously
distributed. This can be represented mathematically by

Zt= αo + α1 Zt-1 + αt {t- T/2} + ut

Test statistics for the regression coefficients under the null

hypothesis that the data are generated by Zt = Zt-1 + ut, where E(ut) = 0.

If a time series is nonstationary but becomes stationary after

differencing, then it is said to be integrated of the order one i.e. I (1). If
two series are integrated of order one, there may exist a linear combination
that is stationary without differencing. If such a linear combination exists
then such streams of variables are called cointegrated.

Cointegration tests are divided into two broader categories: (i)

residual-based tests, and (ii) maximum likelihood-based tests. Residual-based
tests include the Engle-Granger (1987) test while maximum likelihood-based
tests include the Johansen (1988, 1991) and Johansen-Juselius (1990) tests.
During this study, we apply the Johansen and Juselius test to determine the
presence of cointegrating vectors in a set of nonstationary time series data.
The null hypothesis is that there is no cointegration among the series. The
vector autoregressive (VAR) approach is employed to test multivariate
cointegration. This assumes that all the variables in the model are
endogenous. The Johansen and Juselius procedure is employed to test for a
long-run relationship between the variables. Johansen and Juselius suggest two
likelihood ratio tests for the determination of the number of cointegrated
vectors. The maximal eigenvalue test evaluates the null hypothesis that there
are at most r cointegrating vectors against the alternative of r + 1
cointegrating vectors. The maximum eigenvalue statistic is given by,

λmax = - T ln (1 - λr+1)

Where λ r+1,…,λn are the n-r smallest squared canonical

correlations and T = the number of observations.

A trace statistic tests the null hypothesis of r cointegrating vectors

against the alternative of r or more cointegrating vectors. This statistic is
given by
λ trace = -T Σ ln (1 - λi)

In order to apply the Johansen procedure, lag length is selected on

the basis of the Akaike Information Criterion (AIC).
 Causal Relationship among Monetary Variables and Equity Returns 125

If cointegration is present in the long run, then the system of

equations is restructured by inserting an error correction term to capture
the short-run deviation of variables from their relevant equilibrium values.
This is necessary as the impact of financial development is generally more
apparent in the short run and disappears in the long run as the economy
expands and matures. According to Granger (1988), the presence of
cointegrating vectors indicates that Granger causality must exist in at least
one direction. A variable Granger causes the other variable if it helps
forecast its future values. In cointegrated series, variables may share
common stochastic trends so that dependent variables in the VECM must
be Granger-caused by the lagged values of the error correction terms. This
is possible because error correction terms are functions of the lagged
values of the level variables. Thus, evidence of cointegration between
variables itself provides the basis for the construction of an error
correction model (ECM). The ECM permits the introduction of past
disequilibrium as explanatory variables in the dynamic behavior of existing
variables and thus facilitates in capturing both the short-run dynamics and
long-run relationships between variables. The chronological Granger
causality between the variables can be explored by applying a joint F-test
to the coefficients of each explanatory variable in the VECM. The variance
decomposition of equity returns is based on an analysis of responses of the
variables to shocks. When there is a shock through the error term, we
study the influence of this shock on other variables of the system and thus
obtain information on the time horizon and percentage of the error
variance. The F test is in fact a within-sample causality test and does not
allow us to gauge the relative strength of the causality among variables
beyond the sample period.

In order to examine out-of-sample causality, we use variance

decomposition analysis which partitions the variance of the forecast error of a
certain variable into proportions attributable to shocks to each variable in the
system. Variance decomposition analysis presents a factual breakup of the
change in the value of the variable in a particular period resulting from
changes in the same variable in addition to other variables in preceding
periods. The impulse response analysis investigates the influence of a random
shock to a variable on other variables of interest. Impulse responses of returns
in various markets to a shock in oil innovations are also examined. Impulse
responses show the effect of shocks separately for different days whereas
variance decomposition analysis exhibits the cumulative effect of shocks.
126 Arshad Hasan and M. Tariq Javed

4. Empirical Results

Table-1 exhibits descriptive statistics. The average monthly returns

in percentage terms in the Karachi Stock Exchange are 2.2% which is
equivalent to an annualized return of 40.4%. This is one of the highest
returns offered by emerging equity markets. The maximum return in the
Karachi stock market in 1 month is 24.11% whereas the maximum loss in
one month is 27.8%. The average money supply growth rate is 1.67% per
month which is significantly high. Average inflation per month is 0.56%
whereas T bill rates appear to change at a rate of 0.25% per month. The
average decrease in the value of Pakistani currency is 0.35%. Percentage
changes in exchange rates range from a minimum of -7.62% to a maximum
value of 3.03%. However, significant volatility is observed in equity returns
and monetary variables, especially equity returns and interest rates.

Table-1: Descriptive Statistics

Money Change in Change in
Returns Inflation
growth rate T bill rate X rate
Mean 0.0220 0.0167 -0.0025 -0.0035 0.0056
Median 0.0219 0.0091 0.0000 -0.0006 0.0047
Std Dev 0.0912 0.0422 0.0985 0.0121 0.0070
Skewness -0.3055 0.0422 0.0121 0.0985 0.0070
Minimum -0.2780 -0.0646 -0.4242 -0.0762 -0.0088
Maximum 0.2411 0.3481 0.3200 0.0307 0.0303

Weak correlation is observed between the equity return and monetary

variables. The money growth rate is positively correlated with returns that are
in line with results drawn by Maysami and Koh (2000). A possible reason for
this is that an increase in money supply leads to an increase in liquidity that
ultimately results in the upward movement of nominal equity prices. Interest
rates are negatively correlated with equity returns, which is in line with
economic rationale but this relationship is weak. An increase in interest rates
leads to an increase in discount rates in the economy. Since the price of
equity shares is theoretically equal to the present value of cash flows, higher
discount rates lead to a reduction in prices. Similarly, the interest rate parity
theory is also confirmed by our results as the interest rate is negatively
correlated with exchange rates. However, the results indicate a weak
correlation among variables as evident from Table-2.
 Causal Relationship among Monetary Variables and Equity Returns 127

Table-2: Correlation Matrix

Money Change in Change
Returns Inflation
growth rate T bill rate in X rate
Returns 1.0000
Money growth rate 0.0241 1.0000
Change in T bill rate -0.1429 -0.0198 1.0000
Change in X rate 0.1219 0.1455 -0.1974 1.0000
Inflation -0.1698 -0.0145 0.2557 -0.2029 1.0000

Correlation analysis is a relatively weaker technique. The causal

nexus among monetary variables has been investigated by employing
multivariate cointegration analysis. Cointegration analysis tells us about the
long-term relationship among equity returns and set of monetary variables.
Cointegration tests involve two steps. In the first step, each time series is
scrutinized to determine its order of integration. To meet this requirement,
unit root tests designed by Dickey and Fuller (1979) and Phillips and Perron
(1988) have been employed. In the second step, the time series is analyzed
for cointegration by using the likelihood ratio test, which includes (i) trace
statistics and (ii) maximum Eigen value statistics.

A financial time series is said to be integrated to order one i.e, I (1),

if it becomes stationary after differencing once. If two series are integrated
to order one and a linear combination of these is stationary without
requiring differencing, then the data streams are cointegrated.

Our first step is to test the stationarity of the index series. For this
purpose, the ADF test for unit roots has been used at level and first
difference. Table-3 exhibits the results of the Dickey-Fuller (ADF test),
which clearly show that the time series is not stationary at level but that the
first differences of the logarithmic transformations of the series are
stationary. Thus, the series is integrated to the order of one I (1).
128 Arshad Hasan and M. Tariq Javed

Table-3: Unit Root Analysis

ADF- Level ADF- Ist Diff PP- Level PP- Ist Diff
Ln Index -2.1686 -12.015 -2.0872 -12.2821
Ln Money supply -1.8832 -10.245 -1.9545 -10.2284
Ln T bill rate -1.6981 -3.6063 -1.3595 -7.8162
Ln X rate -2.3659 -6.6074 -3.1003 -6.4168
Ln CPI 2.9023 -8.6160 2.6215 -8.6190
1% Critic. Value -4.0363 -4.0363 -4.0363 -4.0363
5% Critic.Value -3.4477 -3.4477 -3.4477 -3.4477
10% Crit.l Value -3.1489 -3.1489 -3.1489 -3.1489

The Dickey-Fuller test requires that the error terms be statistically

independent and data homoskedastic. However, in certain cases these
assumptions may not be true for some data, and so we use another important
technique, the Phillips-Perron test, to test the stationarity of the time series.
Table-3 also displays the results of the Phillips-Perron test, which confirms
the results of the ADF test. Thus, we can conclude that the series is I (1).

Having met these prerequisites, we can now perform cointegration

analysis. The maximum likelihood-based Johansen (1988, 1991) test and
Johansen-Juselius (1990) procedure is used to determine the presence of
cointegrating equations in a set of nonstationary time series. A trace statistic
has been used to test the null hypothesis of r cointegrating vectors against
the alternative of r or more cointegrating vectors. Table-4 exhibits the
results of the multivariate cointegration test for the entire sample period.

Table-4: Multivariate Cointegration Analysis Trace Statistic

Hypothesized Eigen Trace Critical

No. of CE(s) value Statistic Value0.05 Prob.
None * 0.21 71.27 69.82 0.04
At most 1 0.17 44.20 47.86 0.11
At most 2 0.09 22.60 29.80 0.27
At most 3 0.07 12.17 15.49 0.15
At most 4 0.03 3.52 3.84 0.06
The trace test indicates one cointegrating equation at the α = 0.05.
 Causal Relationship among Monetary Variables and Equity Returns 129

Table-5 fails to reject the null hypothesis of no cointegration

between the equity indices and monetary variables for the period 6/1998 to
6/2008 in the Pakistani equity market. The trace test indicates the presence
of one cointegrating equation at the 0.05 level. Therefore, the result
provides evidence of a long-term relationship between monetary variables
and equity prices. However, it must be noted here that the Johansen
cointegration tests do not account for structural breaks in the data.

According to the representation theorem, if two variables are

cointegrated then Granger-causality must exist in at least one direction. The
results of Granger causality are reported in Table-5. Rejection of the null
hypothesis at 5% indicates that there exists unidirectional Granger causality
between the money growth rate and equity returns at the 5% level.
Similarly, unidirectional Granger causality also exists between the interest
rate, inflation, exchange rate and equity returns. This indicates that
monetary variables are Granger-causing equity returns. Treasury bill rates
are also Granger-causing exchange rates. These results are consistent with
Nishat (2001) who indicates that inflation and equity returns are negatively
related to each other.
130 Arshad Hasan and M. Tariq Javed

Table-5: Granger Causality Test

Null Hypothesis Obs F-Statistic Probability

M1 Growth does not Granger Cause Returns 117 2.865 0.040
Returns does not Granger Cause M1 Growth 0.566 0.639
T bill rate does not Granger Cause Returns 117 3.511 0.018
Returns does not Granger Cause T bill rate 0.906 0.441
Change in X rate does not Granger Cause Returns 117 6.191 0.001
Returns does not Granger Cause Change in X rate 0.099 0.960
CPI does not Granger Cause Returns 117 2.980 0.035
Returns does not Granger Cause CPI 0.395 0.757
T bill rate does not Granger Cause M1 Growth 117 3.546 0.017
M1 Growth does not Granger Cause T bill rate 1.938 0.128
Change in X rate does not Granger Cause M1 117 0.481 0.696
M1 does not Granger Cause Change in X rate 0.146 0.932
CPI does not Granger Cause M1 Growth 117 2.078 0.107
M1 Growth does not Granger Cause CPI 0.376 0.770
Change in X rate does not Granger Cause T bill rate 117 1.113 0.347
T bill rate does not Granger Cause Change in X rate 3.087 0.030
CPI does not Granger Cause T bill rate 117 0.924 0.432
T bill rate does not Granger Cause CPI 1.180 0.321
CPI does not Granger Cause Change in X rate 117 1.203 0.312
Change in X rate does not Granger Cause CPI 1.668 0.178

Since a long-run association has been observed between equity

prices and monetary variables, we can explore the possibility of a short-run
relationship by using an ECM framework. The results of the ECM report
indicate that the error term is significant at α = 0.05 and 33% of
disequilibrium is adjusted within a lag of one period.

ecm = ln index -1.3963*M1 + 0.19526*LnTBILL -1.3288*LnXRATE -

0.70738*LnCPI

It is worth mentioning that the coefficients of the money supply, T-bill rate
and exchange rate are significant at α = 0.05.
 Causal Relationship among Monetary Variables and Equity Returns 131

The responses of equity returns have also been examined by using

impulse response analysis (IRF) in the VAR system and results are shown in
Figure-1. Impulse response functions capture the effect innovations in money
growth rate, T-bill rates, exchange rate and inflation on equity returns in the
Karachi stock market. Figure-1 shows the impulse response of equity returns
from a one standard deviation shock to monetary variables. The statistical
significance of the impulse response function has been examined at 95%
confidence bounds. These figures confirm that a one standard deviation
change in money supply leads to an increase in equity prices due to an
increase in liquidity; this result is consistent with Maysami and Koh (2000).
Similarly, a one standard deviation change in the T-bill rate leads to a
reduction in the price of equity due to increased discount rates. No
statistically significant impact has been observed with reference to a variation
in exchange rates. This is reasonable because Pakistan has had a managed
floating rate system and, during the last five years, exchange rates have been
managed within a small range by the State Bank of Pakistan through open
market operations. These results are in conformity with earlier work.

Fig.-1: Impulse Response Analysis

Response to Cholesky One S.D. Innovations

Impulse response functions display the response of an endogenous

variable over time to a given innovation. On the other hand, variance
decomposition analysis expresses the contributions of each source of
innovation to the forecast error variance for each variable. Thus, we have
132 Arshad Hasan and M. Tariq Javed

conducted a variance decomposition analysis to measure the degree to which

shocks to the equity market are explained by money supply, T-bill rates,
exchange rates and inflation. This also supports the pattern of linkages
between monetary variables and equity markets and enhances our insights
into the reaction of markets to system-wide shocks. It also helps identify the
pattern of response transmission over time. Table 6 exhibits the
decomposition of forecast error variance for the equity market that is
explained by monetary variables.

Table-6: Variance Decomposition Analysis

M1 Change in Change Change in

Period SE Returns
Growth T bills rate in X rate CPI
1 0.0838 100.0000 0.0000 0.0000 0.0000 0.0000
2 0.0927 81.8340 10.6955 6.2005 0.9531 0.3169
3 0.1005 71.9706 12.1346 7.2983 3.7298 4.8667
4 0.1026 69.7728 12.2185 7.0533 6.0189 4.9365
5 0.1078 64.9402 14.6710 6.4407 8.2748 5.6734
6 0.1127 63.7044 13.8983 7.2692 9.4382 5.6900
7 0.1159 62.3909 14.9289 7.3497 9.9227 5.4079
8 0.1195 61.0919 15.4836 7.1243 10.7386 5.5616
9 0.1228 60.2977 15.4341 6.9971 11.6140 5.6571
10 0.1259 59.0425 16.1902 6.9894 12.2330 5.5450

Variance decomposition analysis suggests that the money growth

rate, change in T-bill rate and change in exchange rate are considerable
sources of volatility in equity returns. The contribution of an exchange rate
shock to equity returns ranges from 6.2 to 7.3%. Similarly, the contribution
of changes in T-bill rates and inflation to the equity market is also
significantly high, ranging from 4 to 6%. The money growth rate also
contributed to equity market volatility during 1998-2008.

5. Conclusion

This paper examines the lead lag relationship among stock prices and
four important monetary variables which include money supply, T-bill rates,
exchange rates, and inflation for the period 6/1998 to 6/2008 by using
multivariate cointegration analysis and the Granger causality test. The results
provide evidence on information transmission in equity markets and explain
 Causal Relationship among Monetary Variables and Equity Returns 133

the impact of changes in monetary variables on the stock market. Multivariate

regression analysis provides evidence of one cointegration vector, which is an
indicator of a long-term relationship among the variables concerned.

The Granger causality test indicates that the money growth rate
Granger-causes returns. This appears logical as an increase in money supply
leads to increased inflation, which translates into discount rates and
ultimately results in reduction of prices. Similarly, the T-bill rate and
inflation Granger-cause equity returns. These results are consistent with
Nishat (2001) who indicates that inflation and equity returns are negatively
related. The impulse response analysis shows that a one standard deviation
change in money supply leads to an increase in equity prices due to an
increase in liquidity; this is consistent with Maysami and Koh (2000).

Similarly, a one standard deviation change in the T-bill rate leads to

a reduction in the prices of equity due to increased discount rates. A
statistically significant impact has been observed with reference to variations
in exchange rates. In order to take an overall view of the volatility of
returns, we performed a variance decomposition analysis which revealed that
the money growth rate, change in T-bill rate and change in exchange rate
are considerable sources of volatility in equity returns. The contribution of
an exchange rate shock to equity returns ranges from 3 to 14%. Similarly,
the contribution of changes in T-bill rates to the equity market is also
significant. The money growth rate also contributed to equity market
volatility during 1998-2008. Similarly, the significant impact of inflation on
equity prices is captured in our findings.

We can conclude that monetary variables have a long-run as well as

short-run relationship with equity returns. The identification of the impact
of monetary variables on stock market behavior facilitates investors in
making effective investment decisions as by estimating expected trends in
exchange rates, interest rate, and money supply, investors can estimate the
future direction of equity prices and thus allocate their resources more
efficiently. Architects of monetary policy should keep in mind the impact of
changes in interest rates on the capital market in the form of a reduction of
prices. The central bank should consider the impact of money supply on
capital markets. Under the efficient market hypothesis, capital markets
respond to the arrival of new information, implying that macroeconomic
policies should be designed to provide stability to the capital market.
134 Arshad Hasan and M. Tariq Javed

References

Abdalla, I. S. A. Murinde, V., 1997, “Exchange Rate and Stock Price

Interactions in Emerging Financial Markets: Evidence on India,
Korea, Pakistan, and Philippines,” Applied Financial Economics, Vol.
7 : 25-35.

Bahmani-Oskooee, M. Sohrabian, A., 1992, “Stock Prices and the Effective

Exchange Rate of the Dollar,” Applied Economics, Vol. 24 : 459-
464.

Beenstock, M. and Chan, K., 1988, “Economic Forces in the London Stock
Market,” Oxford Bulletin of Economics and Statistics, Vol. 50 (1) :
27-39.

Burmeister, E. and K.D. Wall, 1986, “The Arbitrage Pricing Theory and
Macroeconomic Factor Measures,” The Financial Review, Vol. 21 (1)
: 1-20.

Burmeister, E. and M.B. McElroy, 1988, “Joint Estimation of Factor

Sensitivities and Risk Premia for the Arbitrage Pricing Theory,” The
Journal of Finance, Vol. 43 (3) : 721-735.

Chan, K.C., N.F. Chen and D.A. Hsieh, 1985, “An Exploratory Investigation
of the Firm Size Effect,” Journal of Financial Economics, Vol. 14 :
451-471.

Chang, E.C. and J.M. Pinegar, 1990, “Stock Market Seasonals and
Prespecified Multifactor Pricing Relations,” Journal Of Financial and
Quantitative Analysis, Vol. 25 (4) : 517-533.

Chen, N. F., Roll, R. & Ross, S.A., 1986. “Economic Forces and the Stock
Market,” Journal of Business, Vol. 59 : 383-403.

DeFina, R.H., 1991, “Does Inflation Depress the Stock Market?” Business
Review, Federal Reserve Bank of Philadelphia : 3-12.

Dickey, D. A. and Fuller, W. A., 1981, “Likelihood Ratio Statistics for

Autoregressive Time Series with a Unit Root,” Econometrica, Vol.
49 (4 ): 1057–72.
 Causal Relationship among Monetary Variables and Equity Returns 135

Engle, R.F. and Granger, C.W.J., 1987, “Cointegration and Error

Correction: Representation, Estimation and Testing,” Econometrica,
Vol. 55: 251-276.

Fama E. F. and Schwert, W.G., 1977, “Asset Returns and Inflation,” Journal
of Financial Economics, Vol. 5 : 115-146.

Granger, C.W.J., 1988, “Some Recent Developments in a Concept of

Causality,” Journal of Econometrics, Vol. 39 (1/2) : 199-211.

Hamao, Y., 1988, “An Empirical Investigation of the Arbitrage Pricing

Theory,” Japan And the World Economy, Vol. 1 : 45-61.

Hamburger, M. J. and Kochin, L. A., 1972, “Money and Stock Prices: The
Channels of Influence,” Journal of Finance, Vol. 27 (2) : 231-249.

Hassan, A. H., 2003, “Financial Integration of Stock Markets in the Gulf: A

Multivariate Cointegration Analysis,” International Journal of
Business, Vol. 8 (3).

Humpe, A., and Macmillan, P., 2007, “Can Macroeconomic Variables

Explain Long Term Stock Market Movements? A Comparison of the
US and Japan,” CDMA Working Paper No. 07/20.

Hussain. F., and Mahmood. T., 2001, “The Stock Market and the Economy
of Pakistan,” The Pakistan Development Review, Vol. 40 (2) : 107-
114.

Jaffe, J. & Mandelkar, G., 1976, “The Fisher Effect for Risky Assets: An
Empirical Investigation,” Journal of Finance, Vol. 31 : 447-456.

Johansen, S., 1988, “Statistical Analysis of Cointegrating Vectors,” Journal of

Economic Dynamic and Control, Vol. 12 : 231-254.

Johansen, S., 1991, “Estimation and Hypothesis Testing of Cointegrating

Vectors in Gaussian Vector Autoregressive Models,” Econometrica,
Vol. 59 : 1551-1580.

Johansen, S., 1992, “Determination of Cointegration Rank in the Presence

of a Linear Trend,” Oxford Bulletin of Economics and Statistics,
Vol. 54 (3) : 383-397.
136 Arshad Hasan and M. Tariq Javed

Johansen, S., and K. Juselius, 1990, “Maximum Likelihood Estimation and

Inference on Cointegration with Application to the Demand for
Money,” Oxford Bulletin of Economics and Statistics, Vol. 52 : 169-
209.

Jones, C. M. and Kaul, G., 1996, “Oil and the Stock Market,” Journal of
Finance, Vol. 51 : 463-491.

Kraft, J. and A. Kraft, 1977, “Determinants of Common Stock Prices: A

Time Series Analysis,” Journal of Finance, Vol. 32 (2) : 417-425.

Kryzanowski,L. and H. Zang, 1992, “Economic Forces and Seasonality in

Security Returns,” Review of Quantitative Finance and Accounting,
Vol.2 : 227-244.

Ma, C. K. Kao, G. W., 1990, “On Exchange Rate Changes and Stock Price
Reactions,” Journal of Business Finance and Accounting, Vol. 17 :
441–449.

Maysami, R. C. and Koh, T. S., 2000, “A Vector Error Correction Model of

the Singapore Stock Market,” International Review of Economics
and Finance, Vol. 9 : 79-96.

Mishra A.K., 2004, “Stock Market and Foreign Exchange Market in India:
Are They Related?” South Asia Economic Journal, Vol. 5 : 209-232.

Muhammad, N. Rasheed, A., 2003, “Stock Prices and Exchange Rates: Are
they Related? Evidence from South Asian Countries,” Paper
presented at the 18th Annual General Meeting and Conference of
the Pakistan Society of Development Economists, January 2003.

Mukherjee, T and A. Naka, 1995, “Dynamic Relations between

Macroeconomic Variables and the Japanese Stock Market: An
Application of a Vector Error Correction Model,” Journal of
Financial Research, Vol. 18 : 223-37.

Nelson, C. R., 1976, “Inflation and Rates of Return on Common Stocks,”

Journal of Finance, Vol. 31 (2) : 471-483.

Nishat M. and Rozina. S., 2004, “Macroeconomic Factors and Pakistani Equity
Market,” The Pakistan Development Review, Vol. 43 (4) : 619-637.
 Causal Relationship among Monetary Variables and Equity Returns 137

Nozar, H. and Taylor, P., 1988, “Stock Prices, Money Supply and Interest
Rates: The Question of Causality,” Applied Economics, Vol. 20 :
103-161.

Osterwald-Lenum, M., 1992, “A Note with Quintiles of the Asymptotic

Distribution of the Maximum Likelihood Cointegration Rank Test
Statistics,” Oxford Bulletin of Economics and Statistics, Vol. 50 :
361-377.

Phillips, P. and Perron, P., 1988, “Testing for a Unit Root in Time Series
Regression,” Biometrica, 75: 335-46.

Phylaktis, K. Ravazzolo, F., 2000, “Stock Prices and Exchange Rate

Dynamics,” Paper presented at the EFMA 2000 Meeting in Athens,
May 2000.

Rahman, S., T.D. Coggin and C.F. Lee, 1998, “Some Tests of the Risk-
Return Relationship Using Alternative Pricing Models and Observed
Expected Returns,” Review of Finance and Accounting, Vol. 11 : 69-
91.

Roll, Richard and Stephen A. Ross, 1980, “An Empirical Investigation of the
Arbitrage Pricing Theory,” Journal of Finance, Vol. 35 : 1073-1103.

Ross, S. A., 1976, “The Arbitrage Theory of Capital Asset Pricing,” Journal
of Economic Theory, Vol. 13 : 341-360.

Solnik, B., 1987, “Using Financial Prices to Test Exchange Rate Models: A
Note,” Journal of Finance, Vol. 42 : 141–149.

Stavárek, D., 2004, “Stock Prices and Exchange Rates in the EU and the
USA: Evidence of their Mutual Interactions,” Czech Journal of
Economics and Finance, Vol. 55.

Yu, Q., 1997, “Stock Prices and Exchange Rates: Experience in Leading East
Asian Financial Centres: Tokyo, Hong Kong and Singapore,”
Singapore Economic Review, Vol. 41 : 47–56.

Zhou, G., 1999, “Security Factors as Linear Combinations of Economic

Variables,” Journal of Financial Markets, Vol. 2 : 403-432.