Financial Systems and Economic Growth: Empirical Evidence from Australia

This paper examines the dynamic impact of both bank- and market-based financial development on economic growth in Australia during the period from 1980 to 2012. The study uses the autoregressive distributed lag (ARDL) bounds testing approach to examine this linkage. Unlike certain previous studies, this study uses both bank- and market-based financial development indices to measure the level of financial sector development in Australia. These indices were computed using the means-removed average method. The empirical results of this study show that while bankbased financial development has a short-run positive impact on economic growth in Australia, market-based financial development has no significant impact on economic growth, both in the short run and in the long run. These results imply that, in Australia, it is of paramount importance to concentrate on pro-banking sector policies, at least in the short run, to stimulate growth.


Introduction
Although there is rich literature on the finance-growth nexus, the bulk of such literature is on the relationship between bank-based financial development and economic growth. Only a handful of studies provide coverage on the relationship between market-based financial development and economic growth. However, even in studies that have explored the economic growth impact of market-based financial development, the conclusions are far from conclusive.

Financial Systems and Economic Growth: Empirical Evidence from Australia
In this context, the current study aims to examine the impact of bank-based and market-based financial development on economic growth using data for Australia over the period from 1980 to 2012. The study period was dictated by the availability of the stock market data. This study differs fundamentally from most of the previous studies on the finance-growth nexus in a number of ways. Firstly, it splits financial development into bank-and market-based components, and it focuses on the impact of each component on economic growth. Secondly, the study uses the indices of bank-and market-based financial development, which are created from a wide range of bank-and marketbased financial development indicators. The use of these indices ensures that the financial landscape of the studied country is captured as accurately as possible, unlike in most other studies in which one or two bank-based financial development indicators are used to capture a whole financial system. Thirdly, this study uses the recently developed autoregressive distributed lag (ARDL) bounds approach to cointegration, which is appropriate even when a sample size is too small (see also Odhiambo, 2008). Finally, contrary to the bulk of the previous studies that have over-relied on cross-sectional data, which may not have adequately addressed country-specific issues, this study uses time-series data analysis methods to address country-specific issues (see also Ghirmay, 2004;Odhiambo, 2009).
The study focuses on Australia because the country has not received much individual coverage in terms of the finance-growth nexus research in recent years. Australia also makes an interesting case study because of its recent visibility as one of the leading economies and its distinguished resilience in the context of the recent global financial crises. Australia has one of the best-developed financial systems in the world. Both the bank-and the market-based financial segments of the financial sector are equally well developed.
At the top of the Australian financial system is the Reserve Bank of Australia, which is the country's central bank. The Reserve Bank of Australia is responsible for monetary policy and related matters, and it ensures that the Australian financial fundamentals are in or-der (Reserve Bank of Australia, 2013). The Australian banking sector is stable, and its banks are well capitalized in the context of a sound and effective supervisory environment (Bologna (2010). From the market-based financial side, the Australian stock market is made up of three stock exchanges, namely, the Australian Securities Exchange Group, the National Stock Exchange of Australia, and the Asian Pacific Stock Exchange. These stock exchanges were born out of a string of stock exchanges that merged over time. Of the three, the Australian Securities Exchange Group is the largest.
As with any other financial sector, over the years, the Australian financial sector has undergone a wide range of reforms. According to Perkins (1989), the financial reform period can be divided into three phases: (i) A fully regulated era, which lasted into the late 1960s; (ii) a phase of attempted reform during the 1970s; and (iii) a reformed era, which began during the 1980s and continues into the present. In the banking sector, these reforms concentrated on improving legal, judiciary, regulatory and supervisory environments, promoting financial liberalization, rehabilitating financial infrastructure, restoring bank soundness and improving financial services for consumer protection. From the stock market perspective, the reforms focused on addressing the legal, regulatory, judiciary and supervisory aspects of the market, as well as the transformation of the trading environment. These wide-ranging reforms resulted in a well-developed financial sector, which is competitive and globally recognized.
The remainder of the article is set forth as follows.
The next section provides a review of the related literature. The data, variable descriptions and model specifications are covered in section three. The results are set forth and discussed in section four, and some concluding remarks are drawn in section five.

Review of the Related Literature
Although the relationship between financial development and economic growth has received widespread attention in the modern history of economics, the conclusions have been far from conclusive. The financegrowth nexus debate can be traced to the work of Schumpeter (1911) during the early 20 th Century. The thrust of the debate has been whether financial development has any impact on economic growth, and, if it has, whether the impact is positive or negative. To date, the overwhelming empirical evidence has been in favor of Schumpeter's (1911) notion that financial development has a positive impact on economic growth. From the bank-based financial development perspective, Odedokun (1996), Ahmed and Ansari (1998) (2010) and Bernard and Austin (2011) provide evidence that market-based financial development has a negative impact on economic growth in certain countries.
In addition to the strong view that there is a relationship between financial development (both bankand market-based) and economic growth, irrespective of whether this relationship is positive or negative, there have been a few studies that suggest that financial development, whether bank-or market-based, has no impact on economic growth. These studies provide evidence in support of the notion that financial development and economic growth are not related, and they are two different phenomena that are independent of one another. Such studies include Ram (1999) and Andersen and Tarp (2003). Table 1

Data
The annual time series data that are utilized in this study cover the period from 1980 to 2012; and were

Variable Description
y Growth rate of real gross domestic product. It is a proxy for economic growth.

BD
An index of bank-based financial development, calculated as a means-removed average of M2, M3 and credit provided to the private sector by financial intermediaries. It is a proxy for bank-based financial development (see also Demirguc-Kunt and Levine, 1996) MD An index of market-based financial development, which is a means-removed average of stock market capitalization, stock market traded value and stock market turnover. It is a proxy for market-based financial development (see also Demirguc-Kunt and Levine, 1996) IN Investment, calculated as gross fixed capital formation as a percentage of GDP.

SA
Gross savings as a percentage of GDP TO Trade openness, which is the sum of the share of total imports in GDP and the share of total exports in GDP

Variable Description
The description of the variables that are used in this study is given in Table 2.
The annual growth rate of real GDP is used as a proxy for economic growth (y). This proxy has been used extensively in the literature (Majid, 2008;Odedokun, 1996;Shan & Jianhong, 2006;Wood, 1993).

The Model
The empirical model that is used in this study to test the impact of bank-based and market-based financial development on economic growth is specified as follows: Where α 0 is a constant, α 1 -α 5 are respective regression coefficients and ε t is the error term.
The ARDL model based on the specified empirical model in equation (i) is expressed as follows: Where:α 0 is a constant, α 1 -α 6 and Φ 1 -Φ 6 are respective regression coefficients; ∆ is the difference operator; n is the lag length; and μ t is the white noise error term.
The associated ARDL-based error correction model is specified as follows:

Unit Root Tests
The variables were first subjected to unit root tests using the Phillips-Perron (PP) unit root test. To allow for possible structural breaks in data, the Perron (1997) unit root test (PPURoot) was also utilized. The detailed results of the unit root tests for all of the variables are presented in Table 3.
After being differenced once, the results that are reported in Table 3 show that all of the variables became conclusively stationary. Although the ARDL technique does not require that variables be pre-tested for unit roots, the stationarity test provides guidance as to whether the ARDL analysis is suitable because it is only applicable for the analysis of variables that are integrated of order zero or one. In this case, all of the variables are integrated of either order zero or one. As a result, the ARDL bounds testing method can be used in the estimation of the model.

ARDL Bounds-Testing Approach
The cointegration analysis in this study is based on the fairly newly developed ARDL bounds testing approach because of the numerous advantages that it offers over other alternative empirical analysis methods. First, the ARDL test has superior small sample properties when compared to the other conventional methods of testing cointegration (Pesaran and Shin, 1999). Thus, the ARDL test is suitable even when the sample size is small. Second, the ARDL method employs only a sin-

Variable
Stationarity

169
Financial Systems and Economic Growth: Empirical Evidence from Australia gle reduced-form equation, unlike the conventional cointegration methods that estimate the long-run relationships within a context of a system of equations (see also Duasa, 2007). Third, the technique provides unbiased estimates of the long-run model and valid t statistics even when some of the regressors are endogenous (see also Odhiambo, 2008). Finally, this technique can be employed regardless of whether the regressors are integrated of the same order or not, as long as they are integrated of an order of not more than one. Therefore, the ARDL approach is considered to be well-suited for the analysis of the impact of bank-and market-based financial development on economic growth in this paper. The method has also been increasingly used in recent empirical research.

Bounds F-Test for Cointegration
This section examines the long-run relationship between the variables in the specified model using the ARDL bounds testing approach. First, the order of lags on the first differenced variables in equation (ii) was determined. Finally, a bounds F-test was applied to equation (ii) to establish the existence of a long-run relationship between the variables under study. The results of the bounds F-test are displayed in Table 4.
The results of the ARDL bounds test for cointegration, which are displayed in Table 4, show that the calculated F-statistic of 5.760 is higher than the critical values that were reported by Pesaran, Shin, & Smith (2001) in

Impact Analysis
Because y, BD, MD, IN, SA and TO are cointegrated, the ARDL procedure is used in the estimation of the model. The optimal lag-length for the specified model was determined using the Akaike information criterion (AIC) or the Bayesian information criterion (BIC). The optimal lag-length that was selected based on BIC was ARDL(1,1,0,1,0,0). The BIC-based model was chosen because it was more parsimonious than the AIC-based model. The long-run and short-run results of the selected model are reported in Table 5 Panel 1 and Panel 2, respectively.
The empirical results that are reported in Table 5 reveal that, in Australia, the impact of bank-based financial development on economic growth is time variant; while it is positive in the short run, it is negative in the long run. The positive impact is confirmed by the bank-based financial development coefficient in Panel 2, which is positive and statistically significant, as expected, while the negative impact is supported by the bank-based financial development coefficient in Panel 1, which is statistically significant but negative.
Although the long-run bank-based financial development coefficient for Australia has an unexpected sign, it is not unique to this study. Several other studies have shown evidence of a negative association between the two (Adu et al., 2013;De Gregorio & Guidotti, 1995).
Further, the results that are displayed in Table 5 show that market-based financial development has no significant impact on economic growth in Australia, irrespective of whether the model is estimated in the long run or in the short run. This finding is confirmed by the coef-   Figures 1 and 2, respectively, shows that there is stability, and there is no systematic change identified in the coefficients at a 5% significance level over the study period. The CUSUM and CUSUMSQ graphs, therefore, confirm that the parameters in this model are stable over the sample period.

Concluding Remarks
This paper examined the impact of bank-and marketbased financial development on economic growth in Australia during the period from 1980 to 2012 using the In the long run, its impact on economic growth is largely negative. These results imply that, in order to stimulate growth in Australia, it is of paramount importance to concentrate more on the pro-banking sector policies, at least in the short run. These results also show that the relationship between financial development (whether bank-based or market-based) and economic growth is not clear-cut; as it is proxy-dependent and time-variant.
Hence, the notion that both bank-and market-based financial development have a positive impact on economic growth calls for further scrutiny.