“Analyzing the effect of financial development on economic growth – the Jordanian experience”

This study came to inspect the impact of the development of both financial market and banking system on the economic growth of Jordan based on the annual data covering the period 1993–2017. Through the use of many methodologies: Johansen co- integration test, (VECM), and Granger causality test, where real GDP was used as an indicator of economic growth, the real market value of stocks (Market Capitalization) (LCAP) and Share Turnover (LTURN) are indicators for the financial market, Money supply in the broad concept (LM2), and Local domestic credit (LCR) are indicators for the banking sector. The results of this study reported that the study variables are stationary, and in the level of order 2, they are integrated, and a long-run relationship between the study variables existed according to the Johansen co-integration test. VECM model result and the target model result confirm a short-run causality running from the all variables toward GDP. Granger causality test underline a single directional causality running from variables of our study to GDP and denote the short-run impact between LCAP, LTURN, LM2, LCR, and LGDP. The analysis of the variance decomposition shows that the development of the banking system affects economic growth almost equally with the impact of the development of the financial market. The results go to the same line of supply-leading hypothesis.


INTRODUCTION
The financial systems of developing countries have witnessed many obstacles and restrictions due to the measures taken by the state to impose control over financial and banking activity. Economists therefore called for the adoption the policy of financial liberalization to cancel or effectiveness.
The importance of economic growth comes from people belief that their real freedom lies in achieving a decent standard of living. The growth of any economy is a procedure in which real national income of prevailing economic system grows for a long period of time. If the economic growth rate at constant prices is higher than population growth, this means that the average per capita income is increasing.
To achieve economic growth, it is a necessary to provide a set of elements, the most important one is the existence of a developed and efficient financial system. The financial system of any country consists of a group of financial institutions, most notably financial market and the banking system, which play a pivotal role in achieving high growth rates in any countries of the world. So the efficient financial system is one of the basic requirements for economic growth in any country.
The financial sector is the main channel through which the surpluses are collected from different sectors and distributed to various investment fields by providing financing resources. This is done directly through the financial market or indirectly through the banking system and other financial institutions, which is known as financial intermediation services. The positive role of financial intermediation services is strongly correlated with the efficiency of the financial system and its ability to transfer funds from surplus units to deficit units, and the extent to which the financial system is integrated into real economic units. The development of the financial sector affects growth rates through its impact on capital productivity or its ability to convert financial assets into real investment. In light of the challenges of globalization and technological development in the telecommunications and information sector, especially in a small and developing economy, and open as the Jordanian economy, which is characterized by its political influences and the scarcity of its economic resources.
The Jordanian government has been alerted to the financial development role in achieving better economic growth by job creation, improving the competitiveness of the economy and maximizing wealth. The financial sector represented by the banking system and the financial market has witnessed remarkable development in the past years through organizing the electronic transfer process and increasing the number of banks and commercial banks. On the other hand, the Amman Financial Market witnessed a remarkable development as a result of Jordanians' tendency to invest in financial assets and also increasing volume of foreign direct investment.
In theory, both the banking system and financial markets have a clear effect on the economic growth, but due to outstanding argument among economists about this effect, and the fact that many of those who supported the existence of this relationship differed in the direction of the causal relationship and the results of different researches in different countries, this study seeks to answer the following question: Inquire about the type and nature of the relationship that connects between development of financial markets and the development of the banking system to economic growth in Jordan?
The study importance came from financial system importance to achieving economic growth in our country. The management of the financial system is one of the most important challenges faced by people who put the economic policies of the Jordanian state. The overall economic performance depends to a large extent on the financial system, and economic stability in the short term. The distortions in the financial sector are severe obstacles to long-term economic growth by damaging capital accumulation. This study can lead to important results that help decision-makers in our country to make sound decisions that may serve the economy. This paper will be completed in the following order: section 1 debates or reviews the literature, section 2 describes the empirical results, section 3 presents the econometric methodology used in the paper. Finally, conclusions and recommendations are presented in the last section.

LITERATURE REVIEW
The theoretical interest in the nature of the linkage between developments of financial sector and economic growth is attributed to writings of Bagehot (1873). After that, Schumpeter (1912) stressed the important role of commercial banks in providing the necessary funding to stimulate economic growth. The studies of Shaw (1973), McKinnon (1973) are considered the first leading studies that highlighted the importance of financial developments on economic growth. Further, some economists such as Robinson (1952) believe that economic growth is leading to development of financial sectors.
Given the important nature of these relationship, the applied studies become more important to re-solve the controversy in the direction of the relationship between them. Patrick (1966) is the first researcher who pointed out the possibility of bidirectional causal relationship between economic growth and financial development. But Goldsmith (1969) clarified the difficulty of determining the causal directional relationship between these variables. Gupta (1984) used Granger methodology to determine the causal linkage between the development of financial sector and economic growth in 14 developing countries. Gupta (1984) results confirmed the result of Patrick (1966) which the former called supply-leading, which means that the development of financial sector had led to the economic growth in 8 countries and had not been confirmed by Patrick (1966) which he called demand-following. Levine and Zervos (1996) tested the relationship between economic growth and development of financial sector between 1976 and 1993 using the Two Stage Least Square methodology (2SLS), and the outcomes confirm a positive impact of development of financial sector on the economic growth.
Since Darrat (1999) focused on the causal relationship between the financial deepening and economic growth in Saudi Arabia, United Arab Emirates and Turkey during the period 1980-1995 using the Granger causality test depending on error correction model, the results of this study reinforce the viewpoint that financial deepening is a causal factor necessary for the economic growth in spite of the evidence strength varies across countries. Koivu (2002) focused on effect of the banking sector on economic growth for 25 transition countries using the fixed-effects panel model, and the results show that interest rate is negatively correlated to economic growth, while the rise in the bank credit ratio didn't accelerate economic growth. The researcher pointed out that this result applies to countries living in a transitional state that was characterized by banking crises rocked its financial sectors during the first decade of the transition, and banking sector has also been characterized by easy credit restrictions, which has led to this result. Beck and Levine (2004) examined the effect of both financial markets and the banking sectors on the economic growth of forty countries for the period from 1976 to 1998 by using OLS methodology. The study results showed a positive impact of the development in both the financial market and banking system on economic growth.
The study of Perera (2009) that described Sri Lanka for the period from 1955 to 2005 using the adopted Johansson methodology and VECM, reported a causal relationship between the financial development and the economic growth. Adusei (2013) inspect the linkage between the financial development and the economic growth in Ghana depending on annual data for the period between 2010 and 1971, through the joint integration model, error correction model, and FMOLS model. The researcher found in this study that economic growth was gradually damaged through financial development in Ghana. And last but not least, Greenwood et al. (2013) found that the development of financial sector is very important to the economic development of any country.
In this context, many studies have been conducted and have reached conflicting results and depending on the results of the previous empirical studies, there is no obvious consensus on the direction of the causal relationship between financial development and the economic growth. Also, the empirical results represent the specific countries. Depending on the previous literature, our study came to inspect the impact of the development of both financial market and banking system on the economic growth of Jordan based on the annual data covering the period 1993-2017.

METHODOLOGY
Depending on the previous studies and in order to achieve the objective of this study, we have chosen a set of variables to represent the main variables of this study to inspect the type of the relationship between development of financial sector and economic growth in our country Jordan. According to the economic theory, we describe this relationship as follows:

Economic growth variable
Gross Domestic Product (LGDP): real GDP is represented by the logarithmic formula as a variable representing economic growth. All variables in the study were represented by the logarithmic formula.

Financial market variables
This study used two variables as indicators of development the financial market, one of which reflects the size of the market and the other reflects the liquidity of the market and its efficiency.
1. The real market value of stocks (Market Capitalization) (LCAP): the market value of the shares shows the size of the market and represents the value of shares listed on the Amman Stock Exchange, according to economic theory, the expectation refers to a positive impact on GDP. The reason for using market value as a variable in this study is because size of market is positively or negatively correlated with ability to mobilize capital, and diversification of macroeconomic risks.

Share Turnover (LTURN)
: is an important indicator that reflects stock market liquidity. It is equal the total value of the domestic shares exchanged over a given period of time as a percentage of market value, from which we can measure the volume of transactions relative to the size of the market, and according to economic theory, the expectation refers to a positive impact on GDP.

Variables of the banking sector
This study used two variables to express the development of the banking system.
1. Money supply in the broad concept (LM2), which is considered the most important indicator that indicates the financial development of any state, and positively affects the economic activity. M2 is a broader money concept than M1 that refers to the money supply in the narrow concept, it also includes savings deposits, mutual funds, market securities, and other time deposits.
2. Local domestic credit (LCR): the liabilities from public institutions, private sector, financial institutions, and the net liabilities from the government. The most important role of credit lies in the financing of investment, since according to the multiplier theory, the changes in investment have multiple effects on income, as investment is one of the basic components of aggregate demand. The change in investment leads to a greater change in income. Accordingly, the expectation refer to a positive impact on GDP.
For all variables we used annual time series at macro level for the time period from 1993 to 2017. Our data sources came from Central Bank of Jordan (CBJ).
To investigate the dynamic relationship between the variables of our study, we use Vector Error Correction Model (VECM) as econometric methodology, where it was adopted as a method of analysis, because it is one of the modern methods that we can use to investigate the relationship between different economic variables.
The following expression was used: We specify the Vector Auto Regression (VAR) model as follows: For all the endogenous variables after the estimation is made in the model, we can utilize it to perform many tests such as Granger causality test.

VECM equation estimation is as follows:
( ) They are vectors of exogenous variables from equation (3).
One of the characteristics of this system is that it does not require distinguishing between endogenous and exogenous variables where all variables are treated as internal variables in the system.

EMPIRICAL RESULTS
Depending on the preceding literature, our research postulates an overlapping relationship between the variables under study, so this research is an attempt to display measurement trends of this relationship, and to determine the nature of this relationship, either short-or long-term.
One of the most statistical problems that the researcher may face is non-stationary problem for the time series. In other words, the mean and variance change over time. Because most the economic data suffer from this problem. The regression we get between the series variables at their levels is mostly a spurious regression, and in spite of the possibility in obtaining a high value for coefficient of determination 2 . R As the stationarity test is one of the most important tests in order to avoid the spurious regressions, we implemented stationarity test for our time series under study, firstly, Augmented Dickey-Fuller (ADF) and, secondly, the Phillip-Perron (PP). In order that we implemented the two tests, the results are reported in Table 1.
Depending on the ADF and PP results reported in Table 1 for the all-time series, we reached the result of rejection of the null hypothesis for unit root. Therefore, we can make our conclusion that all variables under study are stationary, and integrated in the level two I(2).
In order to determine the lag length, which is one of the most important requirements for the test of Johansen co-integration, we should adopt the appropriate lag length before conducting or estimating the VAR model, and depending on the results of five criteria presented in Table 2, we choose the lag 2, which is the appropriate lag length, which was selected on the basis of the minimum value of each criterion.
The use of the ordinary least squares method in estimating the regression between the non-stationary time series variables leads to spurious relationship between these variables. Therefore, the co-integration test is utilized to inspect the relationship between non-stationary and integrated variables for the same level in order to estimate correct relationship between them and to eliminate the problem of spurious regression between it. The Johansen co-integration test is one of the most common tests, especially in short-time series. To explore whether the variables of our study are co-integrated (long-term relationship) before using the VECM model, we perform the multivariate Johansen co-integration test, and the Table 3 shows the results of this test.
Johansen co-integration test depends on two types of tests: firstly, trace test, and, secondly, maximum Eigenvalue test. The outcomes of these tests are reported in Table 3A and Table 3B. The outcomes refer to the existence of 3 co-integrating equations. This means that there are three linear combination between our variables.
Co-integration equation results for all our variables under study, show in Table 4, have a significant positive impact on GDP.    LGDP → LCR 00.88033 .4318 Know we will go forward to Variance Decomposition Analysis (VDA), which is utilized to assist in the interpretation of VAR model. The Variance Decomposition refers to the information amount in the autoregression that each variable shares it to the other variables. It is pointed by Pesaran and Shin (1999). The results of VD are reported in Table 7. The analysis of the Variance Decomposition shows that the development of the banking system affects economic growth almost equally with the impact of the development of the financial market.