Group Project
Pooja1806
3.pdf
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
20
Financial Development and Economic Growth: Evidence From Jordan Economy
Zakia A. Mishal, Yarmouk University, Irbid, Jordan
Abstract
Whether financial market development causes, or is caused by, economic growth is an unsettled issue. This subject has greater relevance for Jordan, given the tremendous improvements in banking systems and stock market activities in the last decade. Causality tests were conducted by co-integration testing. The results provide evidence of a stable long-run equilibrium relationship between financial markets’ development and economic development. The causality test results showed a bi-directional causality between economic growth and banking system developments. Moreover, the results demonstrate that economic growth leads to the growth of stock market not vice versa.
Keywords: Economic Growth, Banking Sector Development, Stock Market Development JEL Classifications: E 44, G 10, O 16, O 50
Introduction
Several attempts have been made to investigate the role of the financial sector in
economic growth. The issue has been of great interest, generating a considerable amount of debate among economists for many years. In addition, the growing importance of stock markets recently opened a new avenue of research into the relationship between financial development and economic growth, which focuses on the effects of stock market development over and above the effects of the banking system (Levine & Zervos, 1998). Undoubtedly, stock markets are expected to increase economic growth by increasing the liquidity of financial assets, making global and domestic risk diversification possible, promoting wiser investment decisions, and solving institutional financial problems by increasing shareholders’ interest/value. Thus, the stock market is important for investors and policy makers because of the benefits it provides to the economy, and because it is often cited as a barometer of business direction.
Given the controversy that surrounds this issue, it seemed relevant to further research in order to identify the mechanism through which financial markets influence economic growth. Various channels have been suggested: mobilizing domestic savings, allocating capital proficiency, and diversifying risks (Caporale, Howells, & Soliman, 2005).
Stock markets and banks clearly are substitute sources for corporate finance, because, when a firm issues new equity, its borrowing needs from the banking system decline; it is then possible that stock market development may hamper economic growth if it happens at the expense of banking system development (Fry, 1997; Mayer, 1988;). On the other hand, increased stock market capitalization may be accompanied by an increase in the volume of bank business. If not providing an increase in new lending, financial intermediaries may provide complementary services to issuers of new equity such as underwriting. Thus, at the aggregate level, the
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
21
development of the stock market is likely to go hand in hand with the development of the banking system (Arestis, Demetriades, & Luintel, 2001).
The underlying objective of the present study was to explore the long-run relationship between financial markets’ development and the real sector growth in Jordan. If there has been a long-run relationship, what is the exact direction of the relationship? This issue in the Middle East has not been analyzed thoroughly despite tremendous growth in recent years. The Middle East countries have gone through the process of upgrading their stock markets and developing their regulations. Jordan is one of the most open to foreign investors. The study conducted by Sedik and Petri (2006) indicated that the Amman Stock Exchange (ASE) compares favorably with many other markets in the region in terms of investment restrictions, regulatory environment, and transparency. These characteristics make the ASE a good representative of other emerging stock markets. In this regard, the results reported in this study may be relevant for other emerging markets of similar characteristics and stage of development. In addition, there is an advantage in looking at a specific country when including the stock market development in the empirical model, since using a cross-country study might face the problems of a huge reduced sample size.
Overview of the Literature
An extensive volume of literature and research work has emerged attempting to highlight the role of the financial market in growth. Levine (1997, 2002) provided a comprehensive survey about the subject. In his theoretical study, Singh (1997) examined the importance of stock market development for the economic growth of developing countries. Some studies, such as King and Levine (1993a, b); Levine and Zervos (1996, 1998); and Liang and Reichert (2007) indicated that in most cases stock market improvements promote growth dramatically, especially in developed countries. Levine and Zervos (1996) also made a distinction between the financial services offered by credit and equity markets and suggested that they may complement each other. (Other studies found the same results using the Granger Causality approach and applying data for a single country (Dep & Mukherjee, 2008; Guru-Gharan, Rahman, & Parayita, 2009; Shahbaz, Ahmed, & Ali, 2008;; Somoye, Akintoye, & Oseni, 2009). Alkhathlan (2009) for Saudi Arabia, and Mazur and Alexander (2001) for New Zealand found that the banking system had only a strong, positive effect on economic growth. Agrawalla and Tureja (2007, 2008) for India found that stock market development caused long-term growth in output. For Zimbabwe, Oyama (1997) found that money supply (M2) and market interest rate explained changes in stock prices.
Other studies used data from several countries and found that equity markets play a significant role in the banking sector’s economic growth (Arestis, et al, 2001; Kassimatis & Spyrou, 2001; Liang & Reichert, 2007). The studies by Caporale et al. (2005), and Adjasi and Biekpe (2009) indicated that investment productivity is the channel through which stock markets enhance growth. Thus, any theoretical indication of a link between improvements in financial markets and faster economic growth remains ambiguous. The current study attempted to examine whether a relationship exists between financial market development and output growth in the case of developing a small open economy such as Jordan.
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
22
Research Methodology
The empirical analysis in the present study was based on multivariate Granger causality tests within an error correction framework. The first step in the analysis was to test the macroeconomic series for stationarity through unit root tests. The stationarity of the series was investigated by employing the unit root tests developed by Dickey and Fuller (1979, 1981), and Phillips and Perron (1988). The joint use of both tests attempted to overcome the common criticism that unit root tests have limited power in finite samples to reject the null hypothesis of nonstationarity. Thus, the Augmented Dickey-Fuller (ADF) t-tests and Phillips and Perron (1988) Z(tα) tests are used to test for the presence of a unit root for the individual time series and their first differences.
The next step for co-integration. The advantage of carrying out co-integration testing was to provide was testing evidence of the existence of a stable long-run equilibrium relationship between the macroeconomic variables and financial market variables, which was interesting from a theoretical perspective. In addition, the advantage of carrying out co-integration testing was that the causality tests were preceded by co-integration testing because the existence of co- integration has implications on the way in which causality testing is carried out (Granger, 1988; Toda & Phillips, 1993). The present study used Johansen Cointegration Tests and Vector Error Correction Model (VECM) to avoid potential misspecification biases that might result from the use of a more conventional VAR modeling technique; if the variables used in the VAR model were cointegrated, then the model may have been misspecified because it excluded an additional channel of influence resulting from a long-term equilibrium relationship among these variables (Engle & Granger, 1987). Finally, the causality tests among the co-integrated variables were undertaken.
Data and Measurement
The current study focused on Jordan’s economy spanning a period of 31 years. The model was estimated using yearly data for the period 1978 to 2009. The reason for choosing an annual data was the unavailability of quarterly or monthly data for GDP for a sequent long time frame. The variables that were used are as follows:
1. Economic development: measured by growth rate of nominal GDP (Y). 2. Stock market development: measured by three proxies: The first proxy was market-
capitalization ratio (size proxy) defined by the ratio of market capitalization to GDP (MCY). The second proxy was value-traded ratio (activity proxy) defined by the ratio of trading volume to GDP (VTY), (this ratio did not directly measure the cost of buying and selling shares; it averaged the cost of equity transactions as a share of national output over a long time frame. There would be a less trading if it is costly to buy and sell equities. However, this ratio was particularly suitable to capture the stock market liquidity). The third proxy was stock prices return defined by the growth rate of share price index weighted by market capitalization (SR). Stock prices reflect the marginal productivity of capital; thus increases in stock prices would result in an increase in the marginal productivity of capital, which would be linked directly to an increase in investment activities. Fama (1981) and Barro (1990) also explained changes in stock prices and considered these changes as an important component of variation in the market
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
23
value of capital. Thus, changes in stock prices would cause changes in the market value of capital, in addition to changes in output.
3. Banking system development was measured by the ratio of domestic bank credit to nominal GDP (BY). In addition, the market lending interest rate (I) was used, since it affects the borrowing needs from the banking system. These credit-based indicators were more likely to exhibit a stable long-run relationship with output than deposit-based ones (Arestis, 2001). The data on market capitalization, trading value, and share price index were collected from the Amman Stock Exchange (ASE), while that of GDP, bank credit, market lending interest rate are collected from several issues of the central bank of Jordan (CBJ). Table 1 shows the variables definitions and abbreviations, and Figure 1) represents the
variables in the study during the study period. Table 1. Variable Definitions Definition Calculation Abbreviation Annual growth rate of nominal GDP log (Yt) – log (Yt-1) Y Market capitalization ratio1 (first difference) mcyt – mcyt-1 MCY Value traded ratio2 (first difference) vtyt – vtyt-1 VTY Annual growth rate of share price index3 log (spi)4t – log(spi)t-1 SR Bank credit ratio5 (first difference) byt – byt-1 BY Market lending interest rate (first difference) 6mlit – mlit-1 I Notes: 1Defined by the ratio of market capitalization to GDP. 2Defined by the ratio of trading volume to GDP. 3Stock prices return defined by the growth rate of share price index weighted by market capitalization. 4spi is the stock price index weighted by market capitalization. 5Defined by the ratio of domestic bank credit to nominal GDP.
Empirical Results
The first step of the empirical part of this study was subjecting the data to diagnostic tests; such as unit root tests and Johansen co-integration test. The main purpose of these tests was to search for the model that best fits the data set. The empirical results reported for Jordan were based, as mentioned before, on annual observations for the period 1978 to 2009. All the data were expressed in logarithms.
Response Analysis
For further evidence on the relationships between the growth rate of GDP and the
financial markets, an impulse response function was employed on the multivariate VECM to
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
24
trace the effect of a one-time shock to one of the innovations on current and future values of the endogenous variables. The impulse function is designed to identify the dynamic effects of an exogenous and temporary shock in one variable, say bank credit ratio, on another variable, say the GDP growth. The short-term response is obtained from the first step of the impulse response analysis. It measures the immediate impact on the growth of GDP when an exogenous shock in bank credit ratio occurs. On the other hand, the long-run response is obtained by allowing all variables to respond over time to a shock in bank credits in the first step. To obtain the long-run effects, the number of steps for the impulse response analysis was set at 12 periods. Figure 2 shows responses of GDP growth, market capitalization ratio, value traded ratio, and stock returns to Cholesky One S.D. change in bank credit ratio innovation. As expected, following an expansionary bank credit, the GDP growth decreased as shown in Figure 2. However, this effect was not significant quantitatively, since the output growth converged gradually back toward its long-run equilibrium level.
Table 2 reports the augmented Dickey-Fuller and Phillips-Perron test statistics for the levels and first difference of all nominal variables in this study. According to the results shown in Table 2, the ADF and PP tests for unit roots suggested that the variables considered in this study are all non-stationary in their levels but stationary in first difference.
The second step is testing for co-integration among the variables using the Johansen's (1988) methodology, that is, the trace (trace) and the maximum Eigen value (max) statistics. In general, if two series are found to be co-integrated, then the inference of a long-run equilibrium relation between them is sufficiently robust, except for a stationary disturbance with finite variance. Moreover, in the presence of co-integration, the long-run elasticity of GDP, with respect to the other variables (or vice versa), can be estimated without specifying any dynamics and without an a priori determination of causality, since both variables are endogenous and can be treated symmetrically (Ahmad, 2001).
Since the results derived from these tests were sensitive to the selection of the lag length, two criteria for lag order selection were used: AIC (Akaik Information Criterion), and SC (Schwarz -Information Criterion). The test results suggested using a lag length (which has white noise residuals) of two lags. Subsequent analysis, therefore, proceeded with the use of VAR with lag lengths k = 2. Given that there were six variables in the model (n = 6), there could be a maximum of five co-integrating vectors; thus, r would be equal to 0,1, 2, 3, 4, or 5.
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
25
.00
.05
.10
.15
.20
.25
1980 1985 1990 1995 2000 2005
annual growth of (GDP)
-1.0
-0.5
0.0
0.5
1.0
1.5
1980 1985 1990 1995 2000 2005
annual change of (MCY)
-0.8
-0.4
0.0
0.4
0.8
1.2
1.6
1980 1985 1990 1995 2000 2005
annual change of (VTY)
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
1980 1985 1990 1995 2000 2005
annual growth of (stock price index)
-.12
-.08
-.04
.00
.04
.08
.12
1980 1985 1990 1995 2000 2005
annual change of (BY)
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
1980 1985 1990 1995 2000 2005
annual change of interest rate (I)
Figure 1. The Variables Used in the Analysis During the Period 1978 to 2009
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
26
Table 2. The ADF and PP Tests for Unit Roots
Test statistic GDP BY MCY VTY I Spi Comment Levela. Variable t = 1978-2009 ADF 3.55(1) -1.53(1) -1.22(1) -1.52(1) -2.39(1) -0.76(1) Not I(0) PP 5.10(3) -1.84(3) -1.69(3) -1.99(3) -1.65(3) -092(3) Not I(0) First differencea ADF 2.54(1)b -
4.03(1)* - 2.61(1)***
- 4.44(1)*
- 3.17(1)**
- 3.76(1)*
I(1)
PP - 3.28(3)**
- 4.24(3) *
- 6.71(3)* - 6.10(3) *
- 2.95(3)***
- 5.57(3)*
I(1)
a with intercept and no trend. b non-stationary at the first difference. ***,**,* denote significance at the 1%, 5%, and 10% levels respectively. Note. ADF stands for Augmented Dickey-Fuller; PP for Phillips-Perrone. Numbers in brackets are number of lags used in the ADF test in order to remove serial correlation in the residuals, these lag lengths are chosen based on Akaike's Information Criterion (AIC) and Schwartz Bayesian Criterion (BIC). The truncated lag for PP tests was obtained based on a Newey-West adjustment with lag three for the sample period 1978-2009.
Table 3. Tests for Co-integration Using the Johansen Procedure
Model: Y = f(BC, SMC, REM, SPI) period sample 1978-2009 p=2a
Test statistics hypothesis
r=0b
r1
r 2
r 3 r 4
Trace test 209.4** 121.5** 57.5**
max test 87.9** 64.1** 28.5* a The lag length p was chosen based on AIC (Akaik information criterion), and SC (Schwarz information criterion). b r is the number of co-integrating vectors. The Trace test indicate 3 co-integration equations at both 1% and 5% , while Max eigen value test indicate 3 and 2 co-integration equations at 5% and 1% levels respectively. *, ** Indicates statistical significance at 1% and 5% critical levels. Critical values for choosing the number of co-integrating vectors are taken under the assumption that there is no deterministic trend in the data.
Results of co-integration rank tests for the model are presented in Table 3. The value of the trace test (trace ) indicated that the null hypothesis of three (r 3 ) co-integrating vectors can be rejected at the 1% and 5% levels. That is, it suggested the presence of three co-integrating vectors between annual growth of GDP (Y), annual change of market capitalization ratio (MCY), annual change of value traded ratio (VTY), annual growth rate of stock returns (SR), annual change of bank credit ratio (BY), and annual change in market lending interest rate (I) (see Table 3). At the same time, the Max-eigen value test indicated the existence of 3 co-integrating equations at both 1% and 5% levels, and 2 co-integrating equations at 1% level. Consequently, Jordanian growth rate of output, market capitalization ratio, bank credit ratio, stock returns,
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
27
value-traded ratio, and market lending interest rates were co-integrated. The estimated normalized coefficients of this co-integrating relationship (the s) were significantly different from zero for all the variables. These results suggested the existence of a long-run relationship between Y and the other variables in the study over the entire period 1978 to 2009.
Estimated Error Correction Models
The short-run dynamics, or the direction of causality between the variables in the co- integration equation, were examined by estimating an error correction model. Given the presence of co-integrating relationship, the Engle and Granger (1987) error correction specification could be used to test for Granger causality. The error correction representation for our model for six variable cases is written as follows (we ignore the other five equations to save space):
0tY
117
65432 1
1 ][
tk
ktkktkktkttt
m
EC
SRIBYVTYMCYY
Where Y, the annual growth rate of GDP; MCT, market capitalization ratio; VTY, value traded ratio; BY, bank credit ratio; I, market lending interest rate; and SR, the annual growth rate of share price index; ECt-1 is the vector error correction for the model, and Ɛ1 is uncorrelated disturbances, and m is the lag length.
In order to make valid inferences on causality, all the variables must be stationary. Thus, the annual growth of GDP (Y) and stock return (SR), and the first differences of the variables (MCYt, VTYt, BYt, It), and the residuals (ECs) obtained from the co-integrating vector were included in the Granger causality test structure. The above structure focused on the short-run dynamics among Y, MCY, VTY, BY, I, and SR, and at the same time the long-run information, which was contained in the error correction vector (EC). For each variable in the system, at least one channel of Granger causality was active: either in the short-run through the joint tests of lagged-differences or in a statistically significant EC.
The coefficient of EC contained information about whether the past values of variables affected the current values of the variable under study. The size and statistical significance of the coefficient on the error correction term in each error correction model measured the tendencies of each variable to return to equilibrium. A significant coefficient implies that past equilibrium errors play a role in determining the correct outcomes. For example, if α7 in the above equation were significant, it could be concluded that Y responded to disequilibria in its relationship with the independent variables. The short-run dynamics were captured through the individual coefficients of the difference terms.
Co-integration tests suggested a long-run relationship, but they do not indicate the direction of this relationship. The significance of the error correction coefficient was determined, by the t-ratio given below the coefficient for ECt-1. In each specification, the magnitude of the error correction coefficient indicated the speed of adjustment of any disequilibrium toward a long-run equilibrium state. A statistically significant EC coefficient implied that past equilibrium errors played a role in determining current outcomes. The short-run dynamics were captured by the individual coefficients of the differenced terms. Even if the coefficients of the lagged changes in the dependent variables were not statistically significant, Granger causality could still exist as long as the coefficient of EC were statistically different from zero (Choudry, 1995).
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
28
Table 4. Regression Results Based on Johansen Error Correction Procedure
Independent variables
GDP Development
Equation Eq.1
Stock Market Development Equations
Eq.2 Eq.3 Eq.4
Bank Market Development
Equations Eq.5 Eq.6
ΔY ΔMCY ΔVTY ΔSR ΔBY Δi aECt-1 -0.014
(-0.68) -0.162 (-3.4 )*
-0.05 (-1.24 )
-0.54 (-2.66 )**
-0.037 (-2.96)*
-0.041 (-0.17)
ΔYt-1 -0.56 (-1.73)***
-3.40 (-4.6 )*
-2.00 (-3.37 )*
-6.85 (-2.19)**
-0.026 (-0.13)
-0.762 (-0.20)
ΔYt-2 0.525 (1.35)
-3.45 (-3.88 )*
-1.82 (-2.55 )**
-6.46 (-1.72)***
-0.733 (-3.13)*
-2.24 (-0.49)
ΔMCYt-1 -0.127 (-0.33 )
-3.10 (-3.53 )*
-0.844 ( -1.20)
-5.89 (-1.59)
-0.612 (-2.65)**
-3.30 (-0.73)
ΔMCYt-2 -0.134 (-0.60 )
-0.87 (-1.71)***
-0.534 (-1.32 )
-0.96 (-0.45)
-0.23 (-1.72)***
-3.48 (-1.34)
ΔVTYt-1 0.125 (1.10 )
-0.704 (-2.7)**
-1.14 (-5.40 )*
-0.54 (-0.48)
0.024 (0.34)
0.64 (0.47)
ΔVTYt-2 0.125 (1.12 )
-0.25 (-0.96)
-0.365 (-1.78)***
-1.19 (-1.10)
-0.07 (-1.00)
1.15 (0.87)
ΔSRt-1 0.088 (0.81 )
1.25 (5.10)*
0.78 (3.92)*
-1.72 (-1.65)***
0.19 (2.86)**
1.10 (0.85)
ΔSRt-2 -0.026 (-0.28 )
0.80 (3.73)*
0.623 (3.65)*
0.37 (0.41)
0.145 (2.59)**
1.41 (1.284)
ΔBYt-1 -0.212 (-0.30 )
0.62 (0.39)
-0.130 (-0.10)
2.86 (0.42)
1.032 (2.45)**
7.21 (0.88)
ΔBYt-2 0.94 (2.31 )**
-3.47 (-3.71)*
-0.372 (-0.50)
-7.45 (-1.90)***
-0.55 (2.23)**
1.11 (0.23)
ΔIt-1 -0.01 (-0.39 )
0.094 (1.60)
-0.055 (-1.17 )
0.32 (1.27)
0.016 (1.04)
0.17 (0.56)
ΔIt-2 0.046 (2.12 )**
-0.105 (-2.10)**
-0.040 ( -0.94)
-0.037 (-0.18)
-0.013 (-0.98)
-0.17 (-0.65)
Constant -0.005 (-0.41 )
-0.08 (-2.70)**
-0.030 ( -1.40)
-0.170 (-1.50)
-0.0134 (-1.92)***
-0.025 (-0.18)
Adjusted R2 0.56 0.97 0.97 0.56 0.68 0.18 D.W Test 1.89 1.75 2.12 1.86 1.75 2.17 F-Testb 3.85 63.58 64.98 3.59 5.34 1.45 aECt-1 is the one period lagged error correction term from the co-integrating equation. Numbers in parentheses below ECt-1 are t-statistics for H0: αi = 0. b The F-statistics tests the joint significance of lagged values of the independent variables. The values in parenthesis are the t-test. * , **, and *** denotes significance at 1% , 5% , and 10% significance level respectively.
The results in Table 4 clearly show significant error correction terms for the variables, namely market capitalization ratio (MCY), stock price return (SP), and bank credit ratio (BY) only. That is, the variables market capitalization ratio; stock price return, and bank credit ratio
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
29
adjusted to disequilibrium from the long-run relationship, while GDP growth, value of shares traded ratio, and market lending interest rate did not significantly respond to deviations from the long-run relationship. That is because the lagged coefficients on the independent variables in the error correction model represented short-run Granger causality, while the coefficient on the error correction (EC) term in the error correction model reflected long-run Granger causality.
Causality Tests Results
The results of the estimated multivariate VECM (vector error correction model) are
presented in Table 5. The Causality test results could be summarized between output growth, banking system development, and stock market development in a more candid manner as presented in Table 6. It is clear from Table 6 that the banking system development is a vehicle of economic growth, through the bi-directional causality, which runs both sides between bank credit ratio (BY) and GDP growth (Y), and a uni-directional causality, which runs from lending interest rate to GDP growth. Beside that, the banking system development affects the stock market development. A bi-directional causality runs from BY to MCY (market capitalization ratio), and a uni-directional causality runs from lending interest rate to MCY. On the other hand, the study strongly contradicted the assumption that Jordanian’s stock market is a vehicle of economic growth. The results showed that the only effect of the stock market on the economy was through bank credits. The results showed the existence of a bi-directional causality runs between bank credit ratio (BY) and market capitalization ratio (MCY). As mentioned before, just the variables market capitalization ratio (MCY), stock price return (SR), and bank credit ratio (BY) adjusted to disequilibrium from the long-run relationship, which means the causality runs both ways for long-run and short-run relationships between stock market development (MCY and SR), and banking system development (through BY and I). In general, the results showed that stock market did not Granger cause the GDP growth in Jordan, while the banking system development did Granger cause the GDP growth. This result implies that the revival of stock market could not be taken as a leading indicator of the revival of the economy in Jordan. That is, stock market development (proxied by market capitalization ratio [size proxy], value traded ratio [activity proxy], and stock returns) cannot be considered as a barometer of business direction, or it cannot be relied upon to measure changes in the general economic activities in Jordan. Alkhathlan (2009) reported similar results for Saudi Arabian economy. The study results were also consistent with Agrawalla and Tuteja (2007, 2008) for India. Kassimatis and Spyrou, (2001), in their studies for five emerging countries, indicated that equity markets have a role to play only in liberalized economies.
Response Analysis
For further evidence on the relationships between the growth rate of GDP and the
financial markets, an impulse response function was employed on the multivariate VECM to trace the effect of a one-time shock to one of the innovations on current and future values of the endogenous variables. The impulse function is designed to identify the dynamic effects of an exogenous and temporary shock in one variable, say bank credit ratio, on another variable, say the GDP growth. The short-term response is obtained from the first step of the impulse response analysis. It measures the immediate impact on the growth of GDP when an exogenous shock in bank credit ratio occurs. On the other hand, the long-run response is obtained by allowing all
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
30
variables to respond over time to a shock in bank credits in the first step. To obtain the long-run effects, the number of steps for the impulse response analysis was set at 12 periods. Figure 2 shows responses of GDP growth, market capitalization ratio, value traded ratio, and stock returns to Cholesky One S.D. change in bank credit ratio innovation. As expected, following an expansionary bank credit, the GDP growth decreased as shown in Figure 2. However, this effect was not significant quantitatively, since the output growth converged gradually back toward its long-run equilibrium level.
Table 5. VEC Pairwise Granger Causalitya/Block Exogeneity Wald Test
Dependent variable: Δ GDP growth Variable Chi-sq df Probability Δ Bank credit ratio (BY) 5.53 2 0.062 Δ Lending interest rate (I) 4.66 2 0.097
Dependent variable: Δ MCY (Market Capitalization Ratio) Δ GDP growth 23.67 2 0.000 Δ value traded ratio (VTY) 7.90 2 0.019 Δ Stocks return (SR) 26.35 2 0.000 Δ bank credit ratio (BY) 14.5 2 0.000 Δ lending interest rate ( I) 7.10 2 0.029
Dependent variable: Δ VTY (Value Traded Ratio) Δ GDP growth 12.05 2 0.002 Δ Stocks return (SR) 15.82 2 0.000
Dependent variable: Stocks Return (SR) Δ GDP growth 5.18 2 0.074 Δ market capitalization ratio (MCY)
8.45 2 0.014
Dependent variable: Bank Credit ratio (BY) Δ GDP growth 13.40 2 0.001 Δ market capitalization ratio (MCY)
9.85 2 0.007
Δ Stocks return (SR) 8.31 2 0.015 Note. aThe table shows just the relations which has a significant statistics for the independent variables in the model.
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
31
Table 6. Summary of the Causality Tests between Output Growth, Banking System Growth and Stock Market Growth
Output development Banking system
development Stock market development Y ↔ BYa BY ↔ Ya SR ↔ MCYa
Y → MCYb BY ↔ MCYa SR → VTY Y → VTY I → MCY VTY → MCY Y → SR I → Y SR → BY MCY ↔ BY
a
a ↔ represent bi-directional causality. b → represent uni- directional causality
(Regression Results Based on Johansen Error Correction Procedure)
Figure 2. Response of (1) GDP Growth, (2) Market Capitalization, (3) Value Traded and Stock Returns to Cholsky One S. D. Change in Bank Credits Innovations. Figure 3 shows responses of bank credit ratio, market capitalization ratio, value traded
ratio, and stock market returns to Cholesky One S.D. change in GDP growth innovation.
-.05
-.04
-.03
-.02
-.01
.00
1 2 3 4 5 6 7 8 9 10 11 12
R e sp ons e of GDP g r owth
-.4
-.3
-.2
-.1
.0
.1
1 2 3 4 5 6 7 8 9 10 11 12
Response of Market capitaliza tion r at io ( MC Y)
-.25
-.20
-.15
-.10
-.05
.00
.05
1 2 3 4 5 6 7 8 9 10 11 12
Respons e of Va lue tr aded r a tio (VTY)
-.6
-.5
-.4
-.3
-.2
-.1
.0
.1
.2
1 2 3 4 5 6 7 8 9 10 11 12
Response of Stock ret ur ns ( SR)
FI GU R E 2: R esp o nse o f : ( 1) GDP growth, (2) Market capitalization, (3) Value t r ad ed , a n ( 4) S to ck ret u rn s t o Cholesky one S.D change in Bank credits innovat io n
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
32
Following a shock in GDP growth, bank credit jumped in period 1, as shown in Figure 3(1), and converged gradually back at period 12 towards its long-run equilibrium level. The stock returns growth (SR) also jumped and reached levels higher than its long-run norm, then converged gradually back towards its long-run equilibrium level.
-.024
-.020
-.016
-.012
-.008
-.004
.000
.004
1 2 3 4 5 6 7 8 9 10 11 12
(1) Response of D(BY) to Cholesky One S.D. GDP growth Innovation
-.3
-.2
-.1
.0
.1
.2
1 2 3 4 5 6 7 8 9 10 11 12
(2) Response of D(MCY) to Cholesky One S.D. GDP growth Innovation
-.25
-.20
-.15
-.10
-.05
.00
.05
.10
1 2 3 4 5 6 7 8 9 10 11 12
(3) Response of D(VTY) to Cholesky One S.D. GDP growth Innovation
-.25
-.20
-.15
-.10
-.05
.00
.05
.10
.15
1 2 3 4 5 6 7 8 9 10 11 12
(4) Response of SR to Cholesky One S.D. GDP growth Innovation
(Regression Results Based on Johansen Error Correction Procedure)
Figure 3: Responses of: (1) Bank Credit Ratio, (2) Market Capitalization Ratio, (3) Value Traded Ratio, and (4) Stock Market Returns to Cholesky One S.D. GDP Growth Innovation
Conclusions and Policy Implications
The main contribution of this study was to provide empirical evidence for the relationship
between the economic development and financial sector developments. We examined the causal relationships between the GDP growth, banking sector development, and stock market development in a multivariate vector error correction model. The findings provided strong evidence of a stable long-run relationship between the banking sector and economic growth, and between the banking sector and the stock market. The study reported a bi-directional causality between banking sector development and economic growth in the long run, and a bi-directional causality between the banking sector and stock market. The causality runs from GDP growth to the stock market and not in the opposite direction, implied that the health of the stock market was not reflective of an improvement in the health of the economy in Jordan. This finding had many
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
33
implications for the kind of transactions in the Jordanian stock market in the recent years. The findings of this study suggested that the policies relating to the stock market should be directed toward the creation of transparent and guiding investors to take a long-term view rather than serving as a caterer to satisfy the needs of speculators.
References
Adjasi, C. K. D., & Biekpe, N. (2009). Do stock markets matter in investment growth in Africa?
Journal of Developing Areas, 43(1), 109-120. Agrawalla, R. K., & Tuteja, S. K. (2007). Causality between stock market development and
economic growth. Journal of Management Research, 7(3), 158-168. Agrawalla, R. K., & Tuteja, S. K. (2008). Share prices and macroeconomic variables in India.
Journal of Management Research, 8(3), 136-146. Ahmad, J. (2001). Causality between exports and economic growth: What do the econometric
studies tell us? Pacific Economic Review, 6(1), 147-167. Alkhathlan, K. A. (2009). The role of the financial system in the economic growth: Some
evidence from Saudi Arabian economy. Journal of International Finance and Economics, 9(4), 1-13.
Amman Stock Exchange. [on line] http://www.ase.com.jo Arestis, P., Demetriades, P. O., & Luintel, K. B. (2001), Financial development and economic
growth: The role of stock markets. Journal of Money, Credit and Banking, 33(1), 16-41. Barro, R. J. (1990). The stock market and investment. Review of Financial Studies, 3(1), 115-
131. Caporale, G. M., Howells, P. & Soliman, A. M. (2005). Endogenous growth models and stock
market development: Evidence from four countries. Review of Development Economics, 9(2), 166-176.
Central bank of Jordan [on line] http://www.cbj.gov.jo Choudry, T. (1995). Long-run money demand function in Argentina during 1935-1962: Evidence
from cointegration and error correction models. Applied Economics, 27,66-167. Dep, S. G., & Mukherjee, J. (2008). Does stock market development cause economic growth? A
time series analysis for Indian economy. International Research Journal of Finance and Economics, 21, 142-149.
Dickey, D. A., & Fuller, W. A. (1981). Likelihood ratio statistics for autoregressive time series with a unit root. Econometrica, 49, 1057-1072.
Dickey, D. A., & Fuller, W. A. (1979), Distribution of the estimation for autoregressive time series with a unit root. Journal of American Statistical Association, 79, 355-367.
Engle, R .F. & Granger, C.W. J. (1987). Co integration and error correction: Representation, estimation and testing. Econometrica, 55, 251-276.
Fama, E. F. (1981). Stock returns, real activity, inflation and money. American Economic Review, 71, 545-565.
Fry, M. J. (1997). In favour of financial liberalization. The Economic Journal, 107, 754-770. Granger, C. (1988). Some recent development in a concept of causality. Journal of
Econometrics, 39, 199-211. Guru-Gharan, K. K., Rahman, M., & Parayitam, S. (2009). Influences of selected
macroeconomic variables on U.S. stock market returns and their predictability over varying time horizons. Academy of Accounting and Financial Studies Journal, 13(1), 13- 31.
Journal of Business & Economic Studies, Vol. 17, No. 2, Fall 2011
34
Johansen, S. (1988). Statistical analysis of cointegration vectors. Journal of Economic Dynamics and Control, 12, 231-254.
Kassimatis, K., & Spyrou, S. I. (2001). Stock and credit market expansion and economic development in emerging markets: Further evidence utilizing cointegration analysis. Applied Economics, 33, 1057-1064.
Khedhiri, S., & Muhammad, N. (2008). Empirical analysis of the UAE stock market volatility. International Journal of Finance and Economics, 15, 249-260.
King, R. G., & Levine, R. (1993a). Finance and growth: Schumpeter might be right, Quarterly Journal of Monetary Economics, 108, 717-738.
King, R. G., & Levine, R. (1993b). Finance, entrepreneurship and growth, Journal of Monetary Economics, 32, 513-542.
Levine, R. (1997). Financial development and economic growth: Views and agenda. Journal of Economic Literature, 35, 688-726.
Levine, R. (2002). Bank-based or market-based financial systems: Which is better? Journal of Intermediation, 11, 1-30.
Levine, R., & Zervos, S. (1996). Stock market development and long-run growth. World Bank Economic Review, 10, 323-339.
Levine, R., & Zervos, S. (1998). Stock markets, banks and economic growth. American Economic Review, 88, 537-558.
Liang, H-Y., & Reichert, A. (2007). Economic growth and financial sector development. The international Journal of Business and Finance Research, 1(1), 68-78.
Mayer, C. (1988). New issues in corporate finance. European Economic Review, 32, 1167-1188. Mazur, E. A., & Alexander, R. J. (2001). Financial sector development and economic growth in
New Zealand. Applied Economics Letters, 8, 545-549. Oyama, T. (1997). Determinants of stock prices: The case of Zimbabwe. International Monetary
Fund (WP/97/117), 1-44. Phillips, P., & Perron, P. (1988). Testing for a unit root in time series regression. Biometrica, 75,
333-346. Sedik, T. S., & Petri, M. (2006). The Jordanian stock market: Should you invest in it for risk
diversification or performance? International Monetary Fund (WP/06/187), 1-36. Shahbaz, M., Ahmed, N., & Ali, L. (2008). Stock market development and economic growth:
Ardl causality in Pakistan. International Research Journal of Finance and Economics, 14, 182-195.
Singh, A. (1997). Financial liberalization stock markets and economic development. The Economic Journal, 107, 771-782.
Somoye, R. O. C., Akintoye, I. R., & Oseni, J. E. (2009). Determinants of equity prices in the stock markets. International Research Journal of Finance and Economics, 30, 177-189.
Toda, H. Y., & Phillips, P. C. B. (1993), Vector Autoregressions and Causality. Econometrica, 61, 1367-1393.
Woertz, E. (2006). GCC stock markets at risk. Gulf Research Center Working Papers, Dubai.
Copyright of Journal of Business & Economic Studies is the property of Dowling College and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.
