Financial Crisis, Stock Market and Economic Growth: Evidence from ASEAN-5.

AuthorSamsi, Siti Muliana

This paper examines the effects of the Asian Financial Crisis (AFC) and Global Financial Crisis (GFC) on economic growth in the ASEAN-5 countries. Indonesia, Malaysia and Thailand were found to be most affected by the AFC but not the GFC. Additionally, for Malaysia and Thailand, real output responded positively and significantly to shocks in the stock market. These findings are consistent with the results that show that the shock in the Kuala Lumpur Stock Exchange has a larger effect on real output than other variables for twenty periods. However, a shock in the Bangkok stock exchange has a higher effect on real output and the effect remained strong until the tenth period. This suggests that the stock market is an important economic growth driver in Malaysia and Thailand.

Keywords: Stock market, economic growth, financial crisis, ASEAN.

  1. Introduction

    The stock markets of Indonesia, Malaysia, the Philippines, Singapore and Thailand have undergone substantial liberalization since the late 1980s and early 1990s (World Bank 1993, 1997). These five stock markets have had a long history--with the Malaysian stock exchange dating from 1960--but remained small in capitalization until the late 1980s. However, they have grown rapidly since 1992, following the increase in investment from foreign investors after the end of the Cold War in 1991.

    Dominated by the Malaysian and Singapore stock markets, they experienced significant declines during the Asian Financial Crisis (AFC) and Global Financial Crisis (GFC) through the "contagion effect". However, lessons learned during the AFC helped ASEAN's financial sector to control delinquencies and limit nonperforming loans during and after the GFC.

    The AFC and, to a lesser extent, GFC disrupted a period of healthy economic growth in the ASEAN-5. An interesting fact is that growth rates of the ASEAN-5 have become increasingly intercorrelated since the AFC because of the growth in intraindustry trade arising from product fragmentation (Rana 2006). The sources of this growth are: outward orientation, such as trade openess and foreign direct investment (FDI); and human capital investment. Foreign trade has also promoted the dissemination of new products and new technologies, while international investment brought technological improvements (Lim and McAleer 2004).

    These developments set the context for this study's analysis of the relationship between the stock market and economic growth in the ASEAN-5 in light of disruptions caused by the two major crises. In the next section, a brief literature review of the stock market-economic growth relationship is presented. The third section describes the estimation techniques and data used. The estimation and test results are elaborated in the subsequent section. The final section concludes.

  2. Literature Review

    The relationship between the stock market and economic growth has been the focus of theoretical and empirical research since Goldsmith (1969) and Bosworth (1975) found a positive relationship between stock returns and economic growth. Subsequent work by Barro (1990) Fama (1981; 1990), and Schwert (1990) also confirmed that real stock returns are highly correlated with future real activity. These stuthes reveal that stock returns are a leading indicator for future economic activity, and that stock returns are never led by any real variable (Fama 1981).

    The importance of a healthy and vibrant national stock market is underlined by stuthes showing that a robust stock market not only promotes economic growth, but also predicts it (Rousseau and Wachtel 2000; Hassapis and Kalyvitis 2002; Caporale, Howells and Soliman 2004; Liu and Sinclair 2008; Cooray 2010; Kim and Lin 2013; Pradhan et al. 2013). They contend that a positive economic outlook will lead to long-run stock market optimism because robust economic growth not only directly finances entrepreneurial activity, but also fosters the development of the stock market.

    But there is disagreement on the dynamics of the stock market-economic growth nexus. Cooray (2010), Zivengwa et al. (2011), Kolapo and Adaramola (2012), Kim and Lin (2013), Pradhan et al. (2013) and Teng, Yen and Chua (2013) have all demonstrated me validity of a "supply-leading" view, with unidirectional causality from stock market to economic growth. In contrast, Dritsaki and Dritsaki-Bargiota (2005), Nieuwerburgh, Buelens and Cuyvers (2006) and Liu and Sinclair (2008) support the "demand-following" view, where the causality runs from economic growth to stock markets.

    Other stuthes such as Enisan and Olufisayo (2009) claim to have uncovered "feedback" whereby the causality runs in both directions. Based on this literature, an economy with a well-developed stock market promotes economic expansion through technological changes, products and services innovation. This will, in turn, create a high demand for stock market products. As the stock market effectively responds to this demand, economic growth is stimulated. Both financial and economic developments are therefore interdependent and their relationship could be characterized as having bidirectional causality (Naceur and Ghazouani 2007; Wu, Hou and Cheng 2010; Pradhan et al. 2014).

    It is obvious that the literature on the causal relationship between the stock market development and economic growth remains inconclusive.

  3. Data and Methodology

    The quarterly data used in this study is derived from a number of sources. Selected macroeconomic variables (GDP, broad money M3, interest rate, CPI and exchange rate) are from official websites of the respective governments' departments of statistics. The stock price index has been collected from the International Financial Statistics of the International Monetary Fund (IMF) and Datastream International. The period covered is from 1990Q1 to 2016Q4. The data are analysed using two statistical softwares: Microfit 5.0 and Eviews 9.0.

    Various techniques including unit root tests, Johansen cointegration test, vector error correction model (VECM) and dynamic analysis (impulse response function and variance decomposition) were used in this study. Unit root tests and cointegration test prove the existence of a stable long-run linear relationship between variables. The vector error correction model (VECM) analyses the causal relationship between the stock market and economic activity. The relationship amongst variables is estimated through various lag regressors. The relative strength of the channels--both in the short and long run--is examined explicitly. Impulse response functions (IRF) and variance decomposition (VD) investigate the impact of a shock from economic growth on stock markets.

    Understanding the properties of the forecast errors is helpful to determine interrelationships between variables in the system. The estimated model of this study is based on the VAR model.

    3.1 Model of Crisis Impact

    The effect of financial crisis on economic growth in ASEAN-5 is examined using the following model:

    [mathematical expression not reproducible] (1)

    where sp is stock market of ASEAN-5, while y, m, r, p, e, crisis, and ECT represent: real GDP, money supply, interest rate, inflation, exchange rate, dummy crisis, and error-correction term, respectively. It should be noted that ECT is obtained from the cointegration equation using the Johansen maximum likelihood procedure.

    3.2 Unit Root Test

    Three tests using unit root analysis have been used: Augmented Dickey Fuller (ADF) (Dickey and Fuller 1979; 1981); Phillips-Perron (PP) (Phillips and Perron 1988); and Kwiatkowski-Phillips-Schmidt-Shin (KPSS) (Kwiatkowski et al. 1992).

    While for both the ADF and PP tests the null hypothesis suggests the presence of a unit root, the KPSS null hypothesis claims stationarity. If the ADF and PP both reject the null hypothesis while the KPSS fails to reject the null hypothesis, it signifies the existence of non-stationarity or a unit root in levels.

    The critical problem of the ADF test lies in the difficulty of selecting the appropriate lag length p. If p is too small, the test can give a biased result because of the remaining serial correlation in the errors. If p is too large, the power of the test will be affected. To mitigate this issue, the statistical software Eviews 9.0 allows lag length to be selected automatically using the Akaike Information Criteria (AIC) and Schwarz Information Criteria (SIC), with a maximum lag length set at nine.

    The PP test uses Newey-West (1987) heteroscedasticity and an autocorrelation-consistent covariance matrix estimator to account for serial correlation. It also

    checks for heteroscedasticity in the error term and does not require a lag length specification in the regression.

    The PP test, while superior to the ADF test regarding lag length specification, encounters severe issues of "bandwidth" parameter selection as part of the Newey-West estimator. However, the Eviews software allows the bandwidth to be selected automatically using the Bartlett Kernel function.

    To strengthen the conclusion from unit roots, the KPSS test is additionally used. If a series is found to be non-stationary, it must be differenced to become stationary in order to solve the spurious equation issue. The times of differencing needed for the series to become stationary is referred to as the order of integration, or the number of unit roots. An integration of order d can be denoted as I(d) or I~(d).

    3.3 Cointegration Test

    This study adopts the Johansen and Juselius (1990) maximum likelihood method to test multivariate regression. Generally, the approach is applied to 7(1) variables. The method is an extension of Johansen (1988) and provides a likelihood-ratio statistic to test for the maximum number of independent equilibrium vectors in the cointegrating matrix.

    The Johansen-Juselius procedure has the advantage of taking into account the error structure of the underlying process. It can incorporate both the short-run and long-run dynamics of...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT