Private sector financing on economic growth in Malaysia: results from the bounds test.

ABSTRACT

In this paper the impact of private sector financing by commercial banks on economic growth in Malaysia is investigated over the period from 1976 to 1999. The short-run and long-run relationship between commercial banks private sector financing and other conditioning variables on economic growth are estimated by using the autoregressive distributed lag (ARDL) model proposed by Pesaran and Shin (1995a). Based on the ARDL model and given the sample size for this study is small, a robust estimation method known as the 'bounds test' (Pesaran et al., 2001) is used to examine if economic growth is cointegrated with the explanatory variables included in this study. The empirical analysis showed that the economic growth is cointegrated with the regressors and that there is a significant positive relationship between commercial banks private sector financing and economic growth.

Keywords: financing, economic growth, cointegration, commercial banks

1. INTRODUCTION

The financial systems in countries all over the world have undergone tremendous changes over time. In the early days the commercial banks represented the major players in most financial systems and their activities particularly with respect to mobilization of funds was thought to have a significant contributory effect on economic growth. However, today the financial systems have become more complex with new developments of various financial products, services and institutions. Firms today can turn to a wider array of institutions for financing purposes. This changing environment raises questions with regards to the impact of commercial banks private sector financing on economic growth. In this context, Alawode (1996) asserts that the current reliance of many developing countries on the banking sector for finance is likely to persist for some time although the development of capital markets and the entry of other players in the financial system may in the future reduce the role of commercial banks in economic growth.

To this extent it is worth noting that commercial banks make up a substantial component of the Malaysian financial system. As of the 2003, the banking system accounted for 66.9% of the total assets of the Malaysian financial system and the commercial banks account for 60.2% of the banking system. Thus, the primary aim of this study is to analyze the impact of commercial banks' private sector financing on the country's economic growth. This paper is organized as follows. In Section 2 a brief review of the literature is given. Section 3 presents the data and the econometric methodology. In Section 4, empirical findings are discussed. Results and interpretation are discussed in Section 5. Concluding remarks and policy implications are given in Section 6.

2. LITERATURE REVIEW

As early as 1776, Adam Smith recognized the importance of the financial intermediary role of commercial banks and the substantial contribution that it could make to the economic growth of a nation (Porter, 1966). Subsequently numerous studies [Schumpeter (1934), Gurley and Shaw (1955), Goldsmith (1969), McKinnon (1973), Shaw (1973), Lucas (1988), Gelb (1989), Roubini and Sala-i-Martin (1992) and King and Levine (1993a), Levine et al., (2000) among others] have also examined the role of banks on economic growth by recognizing the important role financial intermediaries play in bridging the gap between savers and borrowers within an economy. According to these authors, economic growth can be sustained only if scarce resources are mobilized efficiently among the most desirable and productive investments and subsequently transformed effectively into productive economic activities. One of the scarce resources essential for long term economic growth is capital formation and since funds are needed to finance capital formation, the efficiency of the financial sector in allocating savings to investments is potentially important, for the long-term performance of an economy.

In this context, the role of commercial banks in facilitating domestic, regional and international flow of funds for productive investments and trade cannot be overstated. The foregoing discussion presents evidence of some consensus among economists that a sound financial system is essential for the growth of an economy. However, others like Robinson (1952) and pioneers of development economics such as Lewis (1955), Rostow (1960), Rozental (1970), and Sinclair (1976) had questioned the validity of a relationship between financial intermediaries and economic growth. They argued that the growth of the financial sector was an incidental reflection of a growing economy.

The conflicting views on the contribution of the financial sector to the growth of the real sector in an economy can be basically described as opposing views on the direction of causality between financial sector development and economic growth. Patrick (1966) developed the 'demand-following' and 'supply-leading' approaches to describe whether financial intermediation was really a reflection of the growing economy or vice versa. In the 'demand--following' approach, the development of the financial sector was a result of the development of the production sectors of the economy. On the other hand, in the 'supply-leading'approach, Patrick explained that the existence of financial institutions and their related services had induced the economy to take advantage of these services for growth purposes. The financial intermediaries thus playing a dominant role, were therefore, considered as a prime determinant for promoting economic growth. To this extent, McKinnon (1988) argued that although a higher rate of financial growth is positively correlated with successful real growth, the issue regarding the direction of causality between finance and economic growth as proposed by Patrick (1966) is still uncertain.

In this context, King and Levine (1993a) highlighted that the causality between finance and economic growth which is, very much a controversial issue could only be resolved using empirical evidence. They found a positive relationship between long-run growth and financial intermediation except in the case of a panel of Latin American countries. De Gregorio and Giudotti (1995) explained that the latter finding in relation to the Latin American countries was the result of financial liberalization with poor regulatory framework. However, the findings of King and Levine (1993a) and De Gregorio and Giudotti (1995) are subject to limitations due to the cross-sectional nature of their study. In the cross-sectional approaches adopted by these authors, it is difficult to identify the country specific patterns of causality. In addition, the underlying assumption of homogeneity of all economies in relation to growth path is unrealistic in cross-country studies (Quah, 1991) due to the varying financial policies and implementation patterns across countries. To this extent, it is worth noting that Demetriades and Hussein (1996) also argued that country specific factors are likely to influence the causal nature of the relationship between financial development and economic growth. In a subsequent study (Levine and Zervos, 1998) made some progress in the empirical links between various measures of stock market development and banking development on economic growth and suggested that financial factors are an integral part of the growth process.

At the outset, it can be noted that finance does play an important role in the growth of an economy, however, Loayza et al's. (2002) argument has shifted the focus to emphasize that the financial fragility in terms of banking crises has an effect on economic growth. Their results highlighted a positive long run relationship between financial intermediation and growth but in the short run it was vice versa. Although numerous studies have been undertaken in this area on the importance of private sector financing on economic growth, this paper contributes to this debate and improves on previous studies by using a new econometric technique called the ARDL framework proposed by Pesaran et al. (2001). The new framework reduces the shortcomings of small sample biases.

Mah (2000) argued that the traditional cointegration methods (Engle-Granger, Johansen-Juselius) are not reliable for small samples studies. In past studies, researchers have used quarterly or monthly data, and in the event these data are not available, the researchers have disaggregated annual time-series data to quarterly data to avoid small sample biases (Vaithilingam et. al., 2003). However, others [Shiller and Perron, 1985; Campbell and Perron, 1991 and Hakkio and Rush, 1991] found that the power of time series analysis with particular emphasis on cointegration lies in the span of data rather than the number of observations. It is for these reasons that in this study annual data is used.

The growth of the economy was estimated as a function of private sector financing, inflation and government consumption. Based on the ARDL model, a new cointegration test known as the 'bounds test' (Pesaran et al., 2001) was used to assess if the growth of the economy is cointegrated with the explanatory variables. The framework applied in this study has an advantage over the traditional cointegration methods as the 'bounds test' is applicable regardless of whether the explanatory variables are integrated of order zero I(0) or one I(1) (Pesaran et al., 2001). Another advantage of bounds test is that the estimation is possible regardless of whether the variables are exogenous or endogenous. In addition, this estimation method is not plagued by the small sample bias problem faced by other traditional cointegration methods (Mah, 2000).

3. DATA AND METHODOLOGY

This paper utilizes annual data for Malaysia from 1976 to 1999, obtained from the IMF International Financial Statistics (CD ROM, July 2004). The period of study considered is from 1976 to 1999 owing to anomaly immediately after the 1997-1998 Asian financial crisis. To this extent, the 2000-2003 periods has been excluded from the analysis as it had a significant impact on Malaysia and other countries in this region. After the 1997-1998 financial crisis, a significant number of Malaysian companies defaulted their debts and the government spent more that US$14bil taking over bad loans and providing lenders with capital. It can be noted that the outcome can be different if the time period is extended to 2003 thus indicating that the outcome can be susceptible to time period chosen.

The measure of economic growth used in this study is the real gross domestic product (RGDP). This is consistent with earlier studies (Gelb, 1989; Roubini and Sala-I-Martin, 1992; King and Levine, 1993a, 1993b; Demetriades and Hussein, 1996; Allen and Ndikumana, 1998; Rousseau and Wachtel, 1998 and Beck, Levine and Loayza, 2000) on financial intermediation and economic growth.

The value of loans made by the commercial banks to the private sector divided by GDP denoted by BP is used to measure the private sector financing role of commercial banks and is in line with De Gregario and Guidotti (1995). This proxy does not include credit to the private sector by non-deposit money banks, the central bank, credit issued to governments or public enterprises and hence is an exclusive measure of the intermediary role of commercial banks. This proxy represents the financial intermediary in one of its main function namely that of credit intermediation by channeling savings to investors.

In addition, from previous studies of financial development and economic growth [De Gregorio and Guidotti (1992, 1995), Levine and Zervos (1998), Bekaert and Harvey, (1998),] two other explanatory variables were selected to be included in the model namely the rate of inflation (INF) and the ratio of government consumption to GDP (GC). These variables are used to assess the strength of an independent link between the banking sector development and economic growth. Although variables such as black market premium (Levine, Loayza and Beck, 2000) and number of revolutions and assassinations (DeGregorio and Guidotti, 1995) were included in previous studies, it has been omitted in this study due to unavailability of data for the study period.

The econometric methodology applied is dictated by the stationarity properties of variables concerned. In this context, the Philips-Peron (PP) unit root test is applied as it not only controls for higher serial correlation in the series but assumptions made on the distribution of errors are milder. The PP test is also more robust against heteroskedasticity. The results of the unit root test, indicates that the least squares method could give rise to 'spurious' regression and as such the Unrestricted Vector Autoregression (UVAR) model is applied. The following is the UVAR model relating the growth of the economy to the private sector financing with the other conditioning variables included:

[Z.sub.t] [mu] + [delta]t + [p.summation over (J=1)] [[phi].sub.j] [Z.sub.t-j] + [[epsilon].sub.t] for t = 1 ,2,.....,. T (1)

Where [Z.sub.t] = [[y.sub.t], [x.sub.t]]', [mu] is the vector of intercepts where [mu] = [[mu].sub.y], [[mu].sub.x]]', [delta] denoting a vector of trend

coefficient, t is the linear trend and [[phi].sub.j] is a matrix of VAR parameters for lag j. The vector [y.sub.t] is the dependent variable, which is the log of the economic growth measured in terms of real GDP. The independent variables are characterized by a matrix [x.sub.t] = [lB[P.sub.t], lG[C.sub.t], IN[F.sub.t]], where lBP is log of the value of loans made by commercial banks to the private sector divided by GDP, lGC is the log of the ratio of government consumption to GDP and INF is the rate of inflation.

Since [Z.sub.t] = [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], the error term [[epsilon].sub.t] is partitioned as [[epsilon].sub.t] = [[epsilon].sub.y,t], [[epsilon].sub.x,t]' ~ N(0,[OMEGA])where [OMEGA] is positive

definite where [OMEGA] = [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII].

Equation (1) can be rewritten as a vector error correction model (VECM), as follows:

[[DELTA][Z.sub.t] = [mu] + [delta]t + [lambda][Z.sub.t-1] + [p-1.summation over (j=1)] [[gamma].sub.j [DELTA][Z.sub.t-j] + [[epsilon].sub.t] for t = 1,2....., T (2)

Where [DELTA] = 1-L and [[gamma].sub.j] = [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] = -[p.summation over (k=j+1)] [[phi].sub.k], and [lambda] = [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] = [I - [p.summation over (j=1]) [[phi].sub.j]]. The matrix I is the

identity matrix. The diagonal elements of matrix [lambda] are unrestricted, hence allowing for the possibility of the series having different order of integration, either I (0) or I (1). If [[lambda].sub.yy] is zero, then y is integrated of order one and when [[lambda].sub.yy] <0, y is integrated of order zero.

The model in (2) can be written as an autoregressive distributed lag (ARDL (p,q)) model as follows:

[DELTA][y.sub.t] = [[alpha].sub.o] + [[alpha].sub.1t] + [phi][y.sub.t-1] + [psi][x.sub.t-1] + [summation] [[beta].sub.y,j] [DELTA][y.sub.t-j] + [p-1.summation over (j=1]) [[beta].sub.x,j][DELTA][x.sub.t-j] + [omega][DELTA][x.sub.t] + [[mu].sub.t] (3)

Model (3) above is used to determine whether y is co-moving with its determinants. The test is called the 'bounds test' procedure proposed by Pesaran et al. (2001). The procedure entails two stages. In the first stage, model (3) is estimated using ordinary least squares (OLS) method. In the second stage, the test of a long run relationship is examined between [y.sub.t] and [x.sub.t] by restricting the coefficients of [Y.sub.t-1] and [x.sub.t-1] to be zero. The bounds test is similar to the Wald type test (F statistics), where the null hypothesis and the alternative hypothesis are as follows: [H.sub.o] : [phi] = 0 and [PSI] = [0.bar];[H.sub.1] : [phi] [not equal to] [0.bar] and [PSI] [not equal to] O. Note that [0.bar] is a (3x1) vector of zeros. Pesaran et al. (2001) showed that the asymptotic distribution of the bounds test statistic is non-standard under the null hypothesis of no cointegration relationship between [y.sub.t] and [x.sub.t].

If the computed F statistics is above the upper critical value given in Pesaran et al. (2001), then it can be concluded that the y is cointegrated with its determinants. If the computed F statistics falls below the lower critical value, then it can be concluded that the y is not cointegrated with its determinants. However, if the computed F-statistics falls within the critical bounds, than inconclusive results are obtained and it is necessary to determine the orders of integration before any conclusions can be drawn. If y is cointegrated with its determinants, then there exist a stable long-run relationship between

[y.sub.t] and [x.sub.t]. The long run elasticity of y with respect to the determinants is measured as -[psi]/[phi] (Bardsen(1989)).

4. EMPIRICAL FINDINGS

In order to estimate equation (1), a test for stationarity is carried out using Phillip-Perron (PP) test. However, a pre-requisite to the stationarity test is the determination of the lag length. The number of lags for all the variables namely RGDP, GC and INF, included in the ARDL model is based on the Akaike Information Criterion (AIC). The model ARDL (3, 3) was chosen as the optimal specification for this study. Moreover, due to sample size constraints and the problems of lack of degree if freedom, it should be noted that a lag length exceeding 3 was not considered in this study.

The results of the unit root test using Phillip-Perron (PP) test with constant and trend were reported in Table 1 for both the levels data and the first difference. The PP tests involve the testing of the null hypothesis that the series is non-stationary, against the alternative that the series is stationary, that is no unit root exist. The critical value for the rejection of the null hypothesis (H0) with linear trend and constant is -3.594 at the 5% level of significance for a sample size of 24. Based on the findings of the unit root test, the null hypothesis of non-stationarity for all of the variables could be rejected in levels. With regards to the order of integration, it can be noted that all the variables exhibited an order of one.

Table 2 present the results on different diagnostic analysis on the optimal specification to assess the robustness of the model and it was found to satisfy all the test namely test of normality, Lagrange multiplier test of serial correlation, the Autoregressive Conditional Heteroskedastic (ARCH) test and lastly Ramsey's RESET test for functional form with an [R.sup.2] = 0.836.

The Regression Specification Error Test (RESET) proposed by Ramsey (1969)is used to test if the model is correctly specified. Based on the RESET test, the model ARDL (3, 3) is specified correctly. In addition to the above test, the CUSUM square test and (CST) and the Recursive Residual Test (RRT) was applied to assess the parameter stability of the dependent variable. It is evident that both the CST and RRT statistics are within the 5% critical line, indicating stability in the estimated model.

The long run relationship between economic growth and explanatory variables was investigated using the 'bounds test'. Based on the 'bounds test', the computed F-statistics is 11.377 is greater that the critical value at 1% level of significant, as such the economic growth is cointegrated with the explanatory variables namely BP, GC and INF. This implies that there exist a long-run relationship between economic growth and these variables. The short-run and long-run elasticities of economic growth with respect to BP, GC and INF is reported in Table 3. In the short-run the growth of the economy is inelastic with respect it BP, GC and INF with elasticities of -0.525, 0.085 and 0.350 respectively. From the short-run coefficients, it can be noted that a 1% increase in private sector financing will lead to a decrease of 0.525% in the growth performance. In the long-run, economic growth was found to be elastic to the BP, GC and INF, with values of 7.085,-8.236 and 18.444 respectively.

5. RESULTS AND INTERPRETATION

The interrelationship between the four factors considered in this study in the Malaysian context can be described as follows. The private sector financing by bank (BP) has a direct positive impact on real gross domestic product (RGDP) in the long-run. However, in the short-run, a direct negative impact on real gross domestic product (RGDP) was observed. To this extent, the empirical results do reveal that there is an important distinction between the short-run and long-run effect of private sector financing and economic growth.

This is logical because private sector financing alone cannot augment economic growth as it has to be combined effectively with the other factors in the economy. In the short run if the mix of factors is such that there is too much private sector financing relative to the other economic factors, it may yield negative marginal returns and hence seem to have a negative impact on RGDP. However, in the long run, all other factors are able to vary in tandem with the increased private sector financing by commercial banks and this is reflected in the observed positive impact on RGDP. The increase in RGDP in turn may augment investors' confidence resulting in an increase in investments and hence the demand for more bank loans. This would culminate in higher interest rates. However, given the positive economic climate and the investors' confidence, the higher interest rates would only promote banks to offer more loans to the private sector in line with demand, resulting in the positive relationship between private sector financing and economic growth in the long-run.

With regards to government spending and economic growth, a negative relationship was observed in the long run. This is also logical since governments usually tend to increase spending during poor economic conditions to boost the economy. In the Malaysian context, the government had even taken drastic measures such as reducing mandatory Employees Provident Fund deductions during recent times to encourage consumer spending to boost domestic demand. However, these efforts may not be necessary in an environment where banks are flushed with liquidity and willing to increase private sector financing coupled with a healthy economic environment, which has seen uninterrupted growth rates of around 8% consistently for close to a decade during the sample period. Thus, the negative relationship between government spending and economic growth during the sample time period may be the result of a conscious effort to reduce expenditure.

A significant positive long run effect was observed between inflation and economic growth. When inflation rates are low, marginal increases in the price level can have a positive impact on the supply side of the economy. To this extent, Malaysia's inflation rates during the study period was around 3-4 % and it can be noted that Malaysia has for several decades experienced low inflation rates. The empirical results in this paper provide strong support that when inflation is kept at low levels, moderate increases in the rate of inflation can have a positive effect of economic growth. This is inline with the structuralist view that inflation is essential for economic growth. More recent work on this issue by Khan and Senhadji (2001) and Malik and Chowdhury(2001), found that there is a threshold rate of inflation that is, when inflation is fairly low, say around 7-11 percent (Khan and Senhadji, 2002) for developing countries then moderate increases in inflation do have a positive impact on the growth of the economy. When prices increase, producers are encouraged to produce more, thus resulting in increasing the supply of goods and services thus contributing positively towards economic growth.

6. POLICY IMPLICATIONS AND CONCLUSIONS

This paper reexamines the role of private sector financing by commercial banks on economic growth using a more robust econometric technique, the ARDL framework. Without having to disaggregate annual data to quarterly data to overcome sample size biasness, the ARDL provides a suitable approach for statistical inference on the dynamics between economic growth and private sector financing.

For the past two decades, Malaysia has experienced rapid economic growth compared to other developing countries in the ASEAN region. The period of study considered is from 1976 to 1999 in order to avoid the anomaly immediately after the 1997-1998 Asian financial crisis. The empirical results strongly suggest that there exists a strong positive linear relationship between private sector financing by commercial banks and the real sector in the long run. However, the same relationship does not hold in the short run. This implies that any shocks to the determinants will have a long-run impact on real GDP. The empirical results indicate that the level of responsiveness of the RGDP to the determinants differ in the short-run and long-run. The RGDP is inelastic to changes in the private sector financing in the short run and elastic in the long run. A possible explanation for this is that in the long-run, banks can facilitate the reallocation of resources to entrepreneurs with high growth potential.

The role of private sector financing by commercial banks on economic growth is crucial because it has implications for development policies. Thus, policy makers should take the positive interaction between commercial banks and economic growth into account when making policy decisions. In this context it is worth noting that during the last two decades, the Malaysian financial sector had witnessed gradual deregulation and this coupled with the trends in globalization and financial liberalization was the basis for the Central Bank to push towards efforts to consolidate the Malaysian financial sector with an aim to create one-stop financial supermarkets. This resulted in a wave of mergers and acquisitions, which had reduced the number of banks from 21 local banks to 10 local anchor banks in view of strengthening the banks and preparing for increased competition with foreign banks. To this extent, it can be noted that Malaysia's move in consolidating the banking system as a mechanism for mobilizing and allocating funds efficiently among competing resources is prove of support that the growth of the economy is strengthen by the financial sector reforms.

In addition, it is worth noting that the Small and Medium Enterprises (SMEs) is a major generator of employment and income in Malaysia. Based on the finding of the positive impact of private sector financing by commercial banks on long run economic growth, it would be beneficial if government policies could be directed towards encouraging commercial bank lending to this sector. This is especially so since banks are often unwilling to extend credit to the SMEs due to lack of proper documentation and collateral.

In conclusion, in order to maintain stability in the long run rate of economic growth, strategies and reforms in the financial sector play a crucial role. The results from this paper provides strong empirical evidence that policymakers in Malaysia have recognized this view and have been and are still supporting this view by ensuring that appropriate policies are undertaken.

ACKNOWLEDGEMENT

The authors would like to thank Monash University Malaysia for funding this project (Project Number: MUM Research Grant 2004-B-5-04). The authors would also like to acknowledge the expert research assistance by Mr. V.K. Mudiarasan for this project. The standard caveat applies.

REFERENCES:

Allen, D. S. and Ndikumana, L., Working Paper 98-004B. Federal Reserve Bank of St. Louis USA, 1998.

Alawode, A.A., "Some Reflections on the Role of Stock Markets in Economic Development", Indian Journal of Economics, 77, 305, 1996, 249-260.

Bardsen, G, . Estimation of Long Run Coefficients in Error Correction Models Oxford Bulletin of Economics and Statistics, Blackwell Publishing, Vol. 51(3), 1989, 345-50.

Beck, T., Levine, R. and Loayza, N., Finance and the Sources of Growth. Journal of Financial Economics, Vol. 58, Issue 1-2, 2000, 261-300.

Bekaert, G. B. and Harvey, C.R., Capital Markets: An Engine for Economic Growth, The Brown Journal of World Affairs 5:1, 1998, 33-53.

Campbell, J.Y. and Perron, P., Pitfalls and Opportunities: What Macroeconomists Should Know about Unit Roots. NBER Macroeconomics Annual, 6, 1991, 141-201.

DeGregorio, J. and Guidotti, P.E., Financial Development and Economic Growth. IMF Working Paper, WP/2/101, 1992, December, IMF, Washington, D.C.

DeGregorio, J. and Guidotti, P.E., Financial Development and Economic Growth. World Development, 23, 3, March, 1995, 433-48.

Demetriades, P.O. and Hussein, K.A., Does Financial Development Cause Economic Growth? Time Series Evidence from 16 Countries. Journal of Development Economics, 51, 1996, 387-411.

Gelb, A., Financial Policies, Growth, and Efficiency. Working Paper WPS 202 June, 1989, The World Bank, Washington D.C.

Goldsmith, R.W., Financial Structure and Development. Yale University Press, 1969, New Haven.

Gurley, J G. and Shaw, E. S., Financial Aspects of Economic Development. American Economic Review, 45, 1955, 515-538.

Hakkio, C. S. and Rush, M., Cointegration: How Short is Long-Run? Journal of International Monetary and Finance, December, 1991, 571-581.

International Monetary Fund, International Financial Statistics (CDROM, July 2004), Washington D.C. Khan, Mohsin S. and Senhadji, Abdelhak S., Threshold Effects in the Relationship between Inflation and Growth. IMF Working Paper, WP/00/110; IMF Staff papers, 48, 1,2001, 1-21.

Khan, Mohsin S. and Senhadji, Abdelhak S., Threshold Effects in the Relationship between Inflation and Growth. IMF staff Papers, 45, 1, 2002.

King, R. G. and Levine, R., Financial Intermediation and Economic Development, in Financial Intermediation in the Construction of Europe. Ed. Mayer, C. and Vives, X. London:Centre for Economic Policy Research, 1993a, 156-89.

King, R. G. and Levine, R., Finance and Growth: Schumpeter Might be Right. Quarterly Journal of Economics 108, 3, August, 1993b, 717-38.

Levine, R. and Zervos, S. Stock Markets, Banks, and Economic Growth. American Economic Review 88, 2, June, 1998.

Levine, R., & Loayza, N., & Beck, T., Financial Intermediation and Growth: Causality and Causes. Journal of Monetary Economics, 46, 1, 2000, 31-77.

Lewis, W.A., 1955, Theory of Economic Growth. Unwin University Books, London.

Loayza, N and Ranciere, R., Financial Development, Financial Fragility and Growth. CESifo Working Paper No. 684, 2002, 5.

Lucas, R. E. Jr., On the Mechanics of Economic Development. Journal of Monetary Economics, 22, 1, July, 3-42, 1988.

Mah, J. S., An empirical Examination of the Disaggregated Import Demand of Korea-the case of information technology products. Journal of Asian Economics, 11, 2000, 237-244.

Malik, G. and Chowdhury, A., Inflation and Economic Growth: Evidence from Four South Asian Countries. Asia-Pacific Development Journal, 8, 1, June, 2001.

McKinnon, R. I.,, Money and Capital in Economic Development. Brookings Institution, Washington D.C, 1973.

McKinnon, R. I., Financial Liberalisation in Retrospect: Interest Rate Policies in LDC's. In: Rains. G. and Shultz, T. P., ed., The State of Development Economics: Proqress and Perspectives, 1988, Basil Blackwell, New York.

Patrick, H., Financial Development and Economic Growth in Underdeveloped Countries. Economic Development and Cultural Change, 14, 1966, 174-189.

Pesaran, H., and Shin, Y., An Autoregressive Distributed Lag Modelling Approach to Cointegration Analysis, DAE Working Paper No. 9514, Dept. of Applied Economics, University of Cambridge, 1995a,.

Pesaran, H., Shin, Y., and Smith, R.J., Bounds Testing Approaches to the Analysis of Level Relationships. Journal of Applied Econometrics, 16, 2001, 289-326.

Porter, R.C., The Promotion of the 'Banking Habit' and Economic Development. Journal of Development Studies, July, 1966.

Quah, D., Empirical Cross-Section Dynamics in Economic Growth. European Economic Review, 37, 1991, 426-434.

Ramsey, J.B., Test for specification errors in classical linear least squares regression analysis. Journal of the Royal Statistical Society B, 1969, 350-71.

Robinson, J., The Rate of Interest and Other Essays. MacMillan, London, 1952,.

Rostow, W.W., The Stages of Economic Growth. Cambridge University Press, Cambridge, 1960.

Roubini, N. and Sala-I-Martin, X., Financial Repression and Economic Growth. Journal of Development Economics, 39, 1, July, 1992, 5-30.

Rousseau, P.L. and Wachtel, P., Financial Intermediation and Economic Performance: Historical Evidence from Five Industrialized Countries. Journal of Money, Credit and Banking, 30, 4, 1998, 657-678.

Rozental, A.A., Finance and Development in Thailand. Praeger, New York,,, 1970.

Schumpeter, J.A., The Theory of Economic Development, 1912. translated by Redvers Opie, Harvard University Press Cambridge, MA, 1934,.

Shaw, E.S., Financial Deepening in Economic Development. Oxford University Press, New York, 1973.

Shiller, R.J. and Perron, P., Testing the random Walk Hypothesis: Power versus Frequency in Sub-Saharan Africa. Journal of Development Economics, 61, 1985, 237-253.

Sinclair, W.A., The Process of Economic Development in Australia. Cheshire, Melbourne, 1976.

Smith, A., The Wealth of Nations, Introduction by Seliqman, E.R.A. (1960). Dent, London, 1776.

Vaithilingam, S., Guru B.K. and Shanmugam, B., Bank lending and Economic Growth in Malaysia. Journal of Asia-Pacific Business, 5, 1, 2003, 51-69.

Santha Vaithilingam, Monash University, MALAYSIA

Mahendhiran Nair, Monash University, MALAYSIA

Balachandher Krishnan Guru, Multimedia University, MALAYSIA

AUTHORS PROFILES:

Dr. Santha Vaithilingam joined the Econometrics and Statistics Department, Monash University Malaysia in the Year 2003. She was previously attached to the Faculty of Management, Multimedia University. Dr. Santha completed her PhD in Banking and Finance in the Year 2002.

Dr. Balachandher Krishnan Guru is a Senior Lecturer at the Faculty of Management, Multimedia University. He obtained his Ph.D. from University of New England, Australia in 1998. He the Co-Chair for the Centre of Excellence for Banking, Investment and Accounting and also a member of the Faculty Research Committee.

Dr. Mahendhiran S. Nair is a Senior Lecturer in the Department of Econometrics & Business Statistics and Director of Research & Development in the School of Business, Monash University Malaysia. He holds a Ph.D. in economics, majoring in econometrics, an MA in Statistics, and an MA in Economics from York University, Canada. His Ph.D. thesis was nominated the best doctoral thesis at York University in 1996.

TABLE 1: UNIT ROOT TEST RESULTS FOR THE VARIABLES USING
PHILLIP-PERRON TEST

Factor   Level     1st Difference   Order of Integration

RGDP     -2.157    -3.9307 ***      l(1)
BP       -1.2002   -4.0465 **       l(1)
GC       -2.0756   -5.2184 *        l(1)
INF      -2.7571   -6.6252 *        l(1)

Note: The PP test statistics are generated by a model with constant
and trend. The *, ** and *** indicates the rejection of null
hypothesis at the 1%, 5% and 10% level of significance based on the
MacKinnon Critical values.

TABLE 2: THE ESTIMATED ARDL (3,3) FOR THE ECONOMIC GROWTH IN MALAYSIA

Dependent variable: [DELTA] RGDP
Sample period: 1976-1999

Variables:          Coefficients            t-statistics

C                   4.1507                  2.4319
[DELTA] RGDP (-1)   -0.10550                -.29976
[DELTA] BP (-1)     .60139                   3.0577 *
[DELTA] GC (-1)    -.81107                  -3.4437 **
[DELTA] INF (-1)    1.0814                  1.1957

* and ** are significant at 1 and 5 percent significance level
respectively

Diagnostic Analysis:
R-Squared:                              0.836
Residual sum of squares:                0.013277
Akaike Info. Criterion:                 40.3235
Durbin-Watson:                          2.3467
F-statistic:                            6.5266 (0.02)
Jarque bera:                            1.2424 (0.537)
LM test:                                1.9639 (0.161)
Autoregressive Conditional
  Heteroskedasticity:                   1.6676 (0.197)
Ramsey RESET:                           0.6982 (0.403)

Note: Probabilities are given in ()

TABLE 3: LONG RUN AND SHORT RUN ELASTICITIES OF THE ECONOMIC
GROWTH IN MALAYSIA

Determinants        Short-run coefficient   Long-run coefficient
                    (p-value)               (p-value)

BP                  -0.5247 *** (0.072)     7.0852 * (0.012) **
GC                  0.0852 (0.717)          -8.2359 * (0.001) *
INF                 0.3499 (0.689)          18.4443 * (0.013) **

* , ** and *** are significant at 1, 5 and 10 percent significance
level respectively