# Impact of Macroeconomic Variables on Stock Prices Essay

This survey tried to research long tally relationship and short tally relationships between ISE10 index and five macroeconomic variables i.e. consumer monetary value index, industrial production index, existent effectual exchange rate, money supply, and three months treasury measures rate. The survey applied Johansen cointegration technique and VECM on monthly informations from July 2004 to June 2008 for analysing ISE10 index. Three long run relationships were found between macroeconomic variables and stock monetary values.

The consequences showed that ISE10 index was negatively related with consumer monetary value index, existent effectual exchange rate and money supply while positively related industrial production index and three months treasury measure rate in the long tally. The VECM analysis the consequences of vector mistake rectification theoretical account ( VECM ) depicted that the accommodations in ISE10 were due to all three mistake rectification footings i.e. ecm1, ecm2, and ecm3.

The ISE10 index was comparatively more exogenic in relation to other variables because 69 per centum of its discrepancy was explained by its ain daze even after 24 monthsCardinal words: Stock monetary values, Cointegration, VECM, Macroeconomic variables,Discrepancy decompositions

## 1: Introduction

The managed and good structured stock markets encourage and mobilize the nest eggs and trip the investing undertakings which lead to economic activities in a state. Islamabad Stock Exchange ( ISE ) became to the full operational on August, 1992. It is one of the three exchanges of Pakistani stock market. ISE10 index of Islamabad stock exchange was introduced in July, 2004. There were 248 listed companies and the market capitalisation was Rs.1943.65 one million millions on December 16, 2008. ISE10 index which reflects overall public presentation of listed companies started with 2716.

0 points in July, 2007 and reached all clip highs of 3334.38 points on April17, 2008 and declined to 2749.64 points ( Annual Report, Islamabad stock exchange, 2008 ) .Mandelker and Tandon ( 1985 ) , Chatrath et Al. ( 1997 ) , Groenewold et Al. ( 1997 ) , Alagidede ( 2008 ) , Ratanapakorn and Sharma ( 2007 ) , and Humpe and Macmillan ( 2009 ) explored the relationship between rising prices and stock monetary values.

Mandelker and Tandon ( 1985 ) investigated relationship between existent stock returns and expected rising prices, and unexpected rising prices and revealed that relationship between stock returns and expected rising prices were negatively related. Chatrath et Al. ( 1997 ) , found negative relationship between stock returns and inflationary tendencies in India. Groenewold et Al. ( 1997 ) investigated relationship between stock returns and expected rising prices in Australian economic system which was found negative in old surveies. The consequences showed that there was an indirect relationship between stock returns and rising prices. Alagidede ( 2008 ) investigated whether stock market supply hedge against rising prices for six African states ( Egypt, Kenya, Morocco, Nigeria, South Africa, and Tunisia ) because this issue got enormous attending in the economic sciences and finance literature. The writer tested Fisher ‘ Hypothesis[ 1 ]for these states.

In Kenya merely, the Fisherian hypothesis was non rejected. Ratanapakorn and Sharma ( 2007 ) reported a positive relationship between stock monetary values and rising prices in US while Humpe and Macmillan ( 2009 ) , showed negative impact of rising prices on stock monetary values.Some surveies established a nexus between stock monetary values and economic growing and found that motions in stock monetary values by and large reflect existent economic activities ( Fama, 1981 ; Chen et al. , 1986 ; Nishat and Shaheen, 2004 ; Ratanapakorn and Sharma, 2007 ; Cook, 2007 ; Shabaz et al.

, 2008 ; Humpe and Macmillan, 2009 ) . Harmonizing to Fama ( 1981 ) there was strong relationship between stock monetary values and gross national merchandise, and stock returns and industrial production. Chen et Al. ( 1986 ) found a powerful association between the economic activity and the stock market. Nishat and Shaheen ( 2004 ) found that industrial production had a positive and strong impact of on stock monetary values in Pakistan. Granger causality trial showed stock monetary value affected industrial production. Ratanapakorn and Sharma ( 2007 ) investigated the long tally relationship between US stock returns and industrial production.

The writers found that stock returns were possibly affected by alteration in end product degree via impact of end product on profitableness. Shahbaz et Al. ( 2008 ) analyzed whether there exist a relationship economic growing between and development of stock market in instance of developing economic system such as Pakistan. Humpe and Macmillan ( 2009 ) explored long tally relationship between stock monetary values and the industrial production by utilizing cointegration technique in US and found US stock monetary values were influenced by the industrial production positively. The impact of foreign exchange rate on stock monetary values was documented by several economic experts during the last two decennaries. Assorted consequences were found among industrial states by Aggarwal ( 1981 ) and Soenen and Hennigar ( 1988 ) . Aggarwal ( 1981 ) in set uping positive relationship between the exchange rate and US stock monetary values. Soenen and Hennigan ( 1988 ) found negative correlativity between the two variables.

Mookerjee ( 1987 ) analyzed money supply and stock returns in France, United States, Japan, Italy, Canada, Germany, United Kingdom, the Netherlands, Belgium and Switzerland ; and Jeng, et Al. ( 1990 ) explored relationship between money supply and stock monetary values in Belgium, Czechoslovakia, France, Hungary, Japan, Poland, Sweden, Britain, Canada and United States. Nishat and Shaheen, ( 2004 ) found negative but undistinguished relationship between money supply and stock monetary values and Ratanapakorn and Sharma, ( 2007 ) explored positive relationship between stock monetary values and money supply in US. While ; Humpe and Macmillan, ( 2009 ) found negative impact of money supply on NKY225 in JapanSome surveies reported positive impact of involvement rate on stock returns while ; some surveies explored negative relationship between these two variables e. g. Ratanapakorn and Sharma, ( 2007 ) reported positive relationship between S & A ; P 500 and treasury measure rate in US and Humpe and Macmillan, ( 2009 ) found negative impact of exchequer measure rate on SP55 in US.The remainder of the paper is as follows.

In subdivision 2 we provide informations beginnings and methodological analysis to research long tally and short tally relationships between stock monetary values and macroeconomic variables and subdivision 3 gives empirical consequences. In the last, decision is explained in subdivision 4.

## 2: Data AND METHODOLOGY

Monthly clip series informations was used in researching the relationship between the macroeconomic variables such as consumer monetary value index, existent effectual exchange rate, three month measures rate, industrial production index, money supply ( M2 ) , and ISE10 ( Index associating to Islamabad stock exchange ) . The chief informations beginnings were monthly bulletins of State Bank of Pakistan, Annual studies of Islamabad stock exchange, The Business Recorder ( Pakistani fiscal newspaper ) , Publications of the Federal Bureau of Statistics, and International Financial Statistics ( IFS ) . The survey used the information from July, 2004 to June, 2008 to research the impact of macroeconomic variables on ISE10 index.

The description of variables used in this research survey was given as under:LISE10 = Log of ISE10LCPI = Log of Consumer monetary value indexLIP = Log of Index of industrial productionLREER = Log of Real effectual exchange rateLM2 = Log of money supply ( Broader money )LTTBR = Log of three months treasury measures rate

## 2.1: Stationary Checks

Many of variables studied in macroeconomics, pecuniary economic sciences and fiscal economic sciences were not stationary clip series ( Hill et al. , 2001 ) . If a clip series was stationary, so dazes were considered transitory. On the other manus, mean or the discrepancy or both the mean and the discrepancy of a non-stationary clip series depends on clip. The discrepancy depends on clip and attack to infinity as clip goes to eternity ( Asteriou and Hall, 2006 ) .Augmented Dickey Fuller trial ( Dickey and Fuller, 1981 ) , Phillips – Perron trial ( Phillips and Perron, 1988 ) , and KPSS ( Kwiatkowski, Phillips, Schmidt. and Shin, 1992 ) unit root trials were applied to prove the stationarity of the above mentioned series.

## 2.2: Cointegration Test and Vector Error Correction Model

Cointegration trial was used to place equilibrium or a long-term relationship among the variables. If there was a long-term relationship between variables, so divergency from the long-term equilibrium way was bounded and the variables were co-integrated. Johansen and Juselius ( 1990 ) process undertook the most of the jobs of Engle and Granger attack. The Johansen and Juselius ( 1990 ) attack was based on maximal likeliness estimations and gives maximal Eigen Value and Trace Value trial statistics for observing figure of cointegrating vectors. This process provides model for cointegration trial in the context of vector autoregressive attack. Johansen method was explained as follows:Where ; Ao is an ( n x 1 ) vector of invariables, xt is an ( n x 1 ) vector of non stationary I ( 1 ) variables, , K is the figure of slowdowns, Aj is a ( n x N ) matrix of coefficients and T is assumed to be a ( n x 1 ) vector of Gaussian mistake footings. The above vector autoregressive procedure was reformulated and turned into a vector mistake rectification theoretical account ( VECM ) in order to utilize Johansen and Juselius trial as under:Where ;and“ I ” is an ( n x N ) individuality matrix, and a?† is the difference operator.

The Trace and the Maximum Eigen Value trial was used to happen the figure of characteristic roots that were insignificantly different from integrity.

## 2.3: Discrepancy decomposition

The vector autoregressive ( VAR ) by Sims ( 1980 ) was estimated to happen short tally causality between macro economic variables and stock monetary values. To exemplify deduction of relationships among macro economic variables and stock indices, discrepancy decomposition was employed. In this survey, Bayesian VAR theoretical account specified in first differences obtained in equation ( 3 ) and ( 4 ) .Where ; Iµ ‘s are the stochastic mistake footings, called inventions or daze in the linguistic communication of VAR.

## 2.4: Model

To research long tally relationship between macro economic variables and ISE10 index, following econometric theoretical accounts was specified in the survey.

## L ISE10 = I?1 L CPI+ I?2 LIP+ I?3 LREER + I?4 L M2 + I?5 LTTBR + Iµt

To capture both the short-term kineticss between clip series and their long-runequilibrium dealingss following theoretical accounts were estimated.

## 3: Empirical consequences

## 3.1: Stationarity trial

The survey applied three different trials for look intoing the stationarity of the informations. All three trials were nem con in the consequences and indicated that all the series were found not stationary at degree but stationary at first difference as was shown in Table 1.

## Table 1: Unit of measurement Root Analysis

## Variables

## Augmented Dickey-Fuller trial statistic

## Phillips-Perron Test

## Statisticss

## Kwiatkowski-Phillips-Schmidt-Shin trial statistic

Null Hypothesis: Variableis Non-stationaryNull Hypothesis: Variableis Non-stationaryNull Hypothesis: VariableIs stationaryDegreeFirst DifferenceDegreeFirst DifferenceDegreeFirst Difference

## LISE10

-2.

43-5-93*-2.40-6.80*0.630.

18*

## LCPI

2.99-1.082.44-5.05*0.900.41**

## LIPI

-2.

51-5.54*-2.53-8.14*0.

760.11*

## LREER

-1.76-6.81*-1.73-6.81*0.

350.14*

## LM2

-0.14-2.68**-0.

13-10.08*0.910.02*

## LTTBR

-2.47-3.

81*-6.24-3.66*0.670.53Test critical values ( MacKinnon, 1996 )5 % Degree-2.

925169-2.9251690.46300010 % Degree-2.600658-2.

6006580.347000* implies that the coefficient is important at 0.05 per centum chance degree and ** implies important at 0.10 per centum chance degree

## 3.

2: Cointegration Analysis

In this survey, to happen the long tally relationship between the ISE10 and macroeconomic variables Johanson and Juselius ( 1988 ) cointegration technique was applied after corroborating the stationarity of the series.The consequences of stationarity analysis shown in the Table 1 illustrated that all the variables involved in the survey were integrated of order one. Hence, the Johansen and Juselius ( 1990 ) cointegration technique was used to research the long tally relationship between the macroeconomic variables i.

e. LCPI, LIP, LREER, LM2, and LTTBR and ISE10 index. In the first measure, appropriate slowdown length was determined by utilizing Schwarz Bayesian Criteria ( SBC ) which showed that the appropriate slowdown length was equal to one. In order to look into the figure of long run relationships between the macroeconomic variables and ISE10 index, both Trace statistic and Maximal Eigen statistic were used. Using Pantula rule, the theoretical account with ‘Unrestricted intercept and no tendency ‘ was selected. The consequences for both Trace statistic and Maximal Eigen statistic were reported in Table 2 and Table 3 severally.Both trials i.

e. the Trace statistic and the Maximal Eigen statistics recognized three cointegrating vectors, hence, the survey used three cointegrating vectors in order to set up the long-term relationships among the variables.

## Table 2: Unrestricted COINTEGRATION RANK TEST ( TRACE )

Hypothesized

## A

TraceStatistic0.05Critical ValueProb. **No.

of CE ( s )EigenvalueNone *0.792187.13995.7540.

000At most 1 *0.682118.12869.8190.000At most 2 *0.

56767.72147.8560.000At most 30.36928.90829.7970.037At most 40.

18110.62815.4950.235At most 50.0411.8443.8410.175Trace trial indicates 3 cointegrating eqn ( s ) at the 0.

05 per centum Probability degree

## A

## A

## A

## A

* denotes rejection of the hypothesis at the 0.05 per centum Probability degree

## A

## A

## A

## A

**MacKinnon-Haug-Michelis ( 1999 ) p-values

## A

## A

## A

## A

## Table 3: Unrestricted Cointegration Rank Test ( Maximum Eigen value )

Hypothesized

## A

Max-EigenStatistic0.05Critical ValueProb. **No. of CE ( s )EigenvalueNone *0.79269.

01040.0780.000At most 1 *0.68250.40833.8770.000At most 2 *0.

56736.81327.5840.

003At most 30.36920.27921.1320.

066At most 40.1818.78414.2650.305At most 50.

0411.8443.8410.175Max-eigenvalue trial indicates 3 cointegrating eqn ( s ) at the 0.

05 per centum Probability degree

## A

## A

## A

## A

* denotes rejection of the hypothesis at the 0.05 per centum Probability degree

## A

## A

## A

## A

**MacKinnon-Haug-Michelis ( 1999 ) p-values

## A

## A

## A

## A

## 3.3: Long Run Relationship

After standardization the first cointegrating vector on LISE10, normalized cointegrating coefficients were estimated as reported in Table4

## Table 4: NORMALIZED COINTEGRATING COEFFICIENTS

## LISE10

## LCPI

## Lip

## LREER

## LM2

## LTTBR

## 1

4.999-5.15217.5431.277-2.562

## S E

-6.

548-0.967-3.789-3.409-0.

616

## t-value

-0.7635.328-4.630-0.3754.160The first normalized equation was estimated as below ;LISE25 = – 4.999LCPI + 5.

152LIP – 17.543LREER – 1.277 LM2 + 2.562LTTBR aˆ¦ ( 6 )Harmonizing to the first normalized equation 6, stock monetary values ( LISE10 ) showed insignificantly negative relationship with consumer monetary value index ( LCPI ) in the long tally. The negative relation between stock monetary values and consumer monetary value index was consistent with the consequences of Humpe and Macmillan ( 2009 ) for US informations. However, findings were at discrepancy Abdullah and Hayworth ( 1993 ) and Ratanapakorn and Sharma ( 2007 ) . Normalized equation depicted that there was a important positive relationship between stock monetary values and industrial production ( LIP ) .

The similar consequences were reported by many research workers ( Fama, 1981 ; Chen et al. , 1986 ; Abdullah and Hayworth, 1993 ; Eva and Stenius, 1997 ; Ibrahim and Yusoff, 2001 ; Nishat and Shaheen, 2004 ; Ratanapakorn and Sharma, 2007 ; Cook, 2007 ; Shabaz et al. , 2008 ; Humpe and Macmillan, 2009 ) . The LISE10 index was influenced by existent effectual exchange rate ( LREER ) negatively.

This implied that along with the addition in exchange rate or depreciation in domestic money, there was a negative consequence on production due to increase monetary values of imported natural stuff finally returns of the houses lessenings and stock monetary values were depressed. Similar determination were reported by Soenen and Hennigan ( 1988 ) . The relationship between stock monetary value and money supply was found negative but undistinguished. The negative nexus between the two variables was consistent with the survey of Humpe and Macmillan ( 2009 ) for Japan.

The survey found that stock monetary values and three month exchequer measures ( LTTBR ) had a positive but undistinguished relation with LISE10 in the long tally. The consequence was consistent with the survey of Ratanapakorn and Sharma, ( 2007 ) for US three months treasury measures rate but contrary to the survey of Humpe and Macmillan ( 2009 ) for US who found negative relationship between US stock market ( S & A ; P500 ) and treasury measures rate.

## 3.4: Vector Error Correction Model

Error rectification mechanism was applied to capture the short tally kineticss of the theoretical account.The consequences of vector mistake rectification theoretical account were reported in Table 5.

The coefficients of ecm1 ( -1 ) , ecm2 ( -1 ) , and ecm3 ( -1 ) showed the velocity of adjustment velocity of the dependent variables to the long tally equilibrium place in a period. As all three mistake rectification footings were important, therefore the consequences of vector mistake rectification theoretical account ( VECM ) depicted that the accommodations in LISE10 were due to all three mistake rectification footings i.e. ecm1, ecm2, and ecm3.

## DLISE10= – O.

004 + 0.243DLISE10 ( -1 ) + 2.369DLCPI ( -1 ) + 0.1797DLIP ( -1 ) + 0.

715DLREER ( -1 ) – 0.304DLM2 ( -1 ) – 0.105DLTTBR ( -1 ) – 0.

642 Vecm1 ( -1 ) – 1.595 Vecm2 ( -1 ) +0.194Vecm3 ( -1 ) aˆ¦aˆ¦..

( 7 )

## Table 5: VECTOR ERROR CORRECTION ESTIMATES

## Variables

## D ( LISE10 )

## D ( LCPI )

## D ( LIP )

## D ( LREER )

## D ( LM2 )

## D ( LTTBR )

## Vecm1 ( -1 )

-0.642( -4.45 )0.027( 0.

075 )-0.042( -0.34 )0.012( 0.43 )0.

037( 1.11 )0.154( 2.20 )

## CointEq2

-1.595( -3.70 )0.105( 2.

65 )-0.392( -1.07 )0.124( 1.

97 )0.136( 1.37 )0.901( 4.32 )

## CointEq3

0.

194( 3.10 )0.003( 0.60 )-0.086( -1.61 )0.

030( 3.23 )0.001( 0.045 )0.007( 0.23 )

## D ( LISE10 ( -1 ) )

0.243( 1.48 )0.

011( 0.76 )0.287( 2.06 )-0.038( -1.59 )-0.

064( -1.69 )-0.252( -3.18 )

## D ( LCPI ( -1 ) )

2.

369( 1.096 )0.067( 0.34 )2.349( 1.27 )-0.972( -3.

07 )-0.402( -0.81 )-1.235( -1.18 )

## D ( LIP ( -1 ) )

0.179( 0.

85 )-0.013( -0.66 )-0.116( -0.65 )-0.047( -1.51 )0.

066( 1.36 )-0.023( -0.22 )

## D ( LREER ( -1 ) )

0.715( 0.60 )-0.084( -0.78 )-1.

117( -1.11 )0.246( 1.

428 )-0.024( -0.08 )1.

129( 1.98 )

## D ( LM2 ( -1 ) )

-0.304( -0.45 )0.096( 1.55 )0.147( 0.25 )-0.

007( -0.07 )-0.464( -2.99 )-0.116( -0.

35 )

## D ( LTTBR ( -1 ) )

-0.105( -0.39 )-0.014( -0.55 )0.

209( 0.92 )-0.054( -1.37 )0.010( 0.17 )-0.

101( -0.78 )

## C

-0.004( -0.15 )0.

007( 2.81 )-0.022( -0.94 )0.011( 2.

73 )0.023( 3.54 )0.054( 4.05 )R-squared0.420.420.230.

360.320.74F-statistic3.022.921.212.261.8911.

52( ) shows ‘t ‘ values of “ T ” statistics* show the coefficient significantly different from nothing at 0.01 per centum chance degree** show the coefficient significantly different from nothing at 0.05 per centum chance degree*** show the coefficient significantly different from nothing at 0.10 per centum chance degree

## 3.

5: Discrepancy Decompositions

In order to analyze the proportion of calculating mistake discrepancy in 24-months, the vector autoregressive ( VAR ) was estimated. The Variance decomposition provided farther grounds on the relationships of the variables under survey and grade of exogeneity among the variables. Table 3.6 showed that the LISE10 index was comparatively more exogenic in relation to other variables i.e. LCPI, LREER, LM2, and LTTBR because 69 per centum of its discrepancy was explained by its ain daze even after 24 months.

LCPI explained 10.34 percent impact on stock monetary values. Inventions in other macroeconomic variables i.e. LIP, LEER LM2, and LTTBR explained forecast discrepancy 6.66 per centum, 1.

31 per centum, 4.17 per centum, and 8.44 per centum severally for LISE10. The value of discrepancy prognosis mistake explicated by all macroeconomic variables increased along with the transition of clip. The grade of exogeneity of LCPI was greater than other variables including LISE10.

## Table 6: VARIANCE DECOMPOSITIONS

## VDC of

## Calendar months

## S.E.

## LISE25

## LCPI

## Lip

## LREER

## LM2

## LTTBR

## LISE25

10.08100.000.000.000.

000.000.00120.1070.

0510.466.650.874.

067.91240.1069.

0810.346.661.

314.178.44

## LCPI

10.012.8097.200.000.

000.000.00120.

039.4434.6819.1527.

715.123.89240.078.7026.0722.6126.489.127.02

## Lip

10.060.006.8593.150.000.000.00120.081.949.0885.330.182.950.52240.082.4910.1580.742.643.240.73

## LREER

10.010.143.3511.8784.640.000.00120.020.232.6113.0472.971.329.83240.020.844.5613.4665.893.5711.68

## LM2

10.022.218.032.816.6480.310.00120.056.6020.6919.8117.1934.191.52240.108.1622.7323.1224.8115.565.63

## LTTBR

10.055.310.0011.030.640.4482.57120.113.845.657.2323.087.4652.74240.135.4410.6713.0128.126.7536.00Cholesky Ordering: LISE10 LCPI LIP LREER LM2 LTTBR

## 4: Decision

This survey investigated long tally and short tally relationships between ISE10 Index and five macroeconomic variables in Pakistan. All the series used in this analysis was found not stationary at degrees but stationary at first difference. Three long run relationships were found between macro economic variables and ISE10 index. In the long tally, Industrial production index, and three month exchequer measure rate affected stock returns positively. While, rising prices, existent affectional exchange rate, money supply showed negative impact on stock returns in the long tally.As all three mistake rectification footings were important, therefore the consequences of vector mistake rectification theoretical account ( VECM ) depicted that the accommodations in LISE10 were due to all three mistake rectification footings i.e. ecm1, ecm2, and ecm3. The consequences of Variance Decomposition revealed that ISE10 index explained about 69 per centum of its ain prognosis mistake discrepancy while CPI, IP, REER, M2, and TTBR explained 10.34 per centum, 6.66 per centum, 1.31 per centum, 4.17 per centum, and 8.44 per centum severally for LISE10.The survey proposed that appropriate pecuniary steps should be adopted by pecuniary directors to command rising prices so that the volatility of the stock markets can be minimized. Increase in Industrial production can play important positive function in development of the capital markets of Pakistan. Therefore, it was recommended that governments should explicate such a policy which supports stock monetary values through the publicity of industrial production. The long tally positive impact of exchange rate on ISE10 index suggested that for the development of stock market in Pakistan, exchange rate should be managed carefully maintaining in position the snaps of exports and imports which will take to stableness in stock market.