Impact of Macroeconomic Factors and Political Events on the Market Index Returns at Palestine and Amman Stock Markets (2011–2017)

This study aims to investigate the effect of macroeconomic factors on Palestine and Amman Stock Exchange returns. Also, the study handles the political events in the area and their impact on Palestine and Amman stock markets returns. This study appliedthe macro-econometric model based on Arbitrage Pricing Theory. In addition, themost important political events are selected, and their effect was tested using the event study methodology. The results show that the consumer price index, gross domesticproduct, and exchange rate have a significant impact on stock index returns, but industrial production index and balance of trade have no significant effect. In addition, theresults reveal that the political events have a significant effect on Palestine and Amman stock markets returns. For instance, at Palestine Stock Exchange, seven out of eleven events had a significant impact on the Palestinian general index returns. Regarding the Amman Stock Exchange, there were nine out of eleven events, which had a significant impact on the Jordanian general index returns. The main results show that themacroeconomic factors and political events have a significant impact on the Palestine and Amman stock market returns. Both Palestine and Amman Stock Markets are inefficient and the markets do not absorb uncertain information and noisy events.


INTRODUCTION
Financial market is the link between individuals and organizations, whereas stock prices reflect the hopes and fears for thousands of buyers and sellers and formulate their decisions based on their evaluations to many considerations. The results of many studies showed that the macroeconomic factors have an obvious effect on the financial markets (Hunjra, Chani, Ijaz, Farooq, & Khan, 2014). But it is important to mention that the effect on the advanced markets differs from that on the developing markets, as well as the political events play the same role in the financial markets (Momani & Alsharari, 2012). However, there are limited up-to-dated researches on how these factors affect the developing stock markets (Hussainey & Khanh Ngoc, 2009).
Based on that, the aim of the research is to study the impact of macroeconomic factors, as well as to study the influence of political events on Palestine and Amman stock markets index returns, in order to make relevant and valuable recommendations for investors, organizations, and financial analysts, and to give a new insight about the Zhu (2012) studied the effect of macroeconomic factors on energy sector returns in Shanghai stock market by applying Augmented Dickey-Fuller test and Arbitrage Pricing Theory model. The results mainly showed that macroeconomic factors have an effect on the energy sector stock returns, but regarding the industrial production index, it does not have an impact on the energy sector stock returns. Altinbas and Biskin (2015) also examined which macroeconomic factors have an influence on predicting the market returns on Turkish stock market by using sequential forward selection algorithm by taking BIST100 index returns as the dependent variable and industry production index as the independent variable. The results revealed that one month lagged stock the market index values are enough to predict market index's future values.

LITERATURE REVIEW
H01: There is a statistically significant relationship between industrial production index and the stock market returns.
Ilahi, Ali, and Jamil (2015) used multiple linear regression model to test the impact of interest rate, inflation rate, and exchange rate on stock market returns; consumer price index was used as an indicator for inflation and Pakistan Karachi Stock Exchange 100 index as an indicator for stock market returns. The results showed that exchange rate, interest rate, and inflation rate have insignificant relationship with Pakistan Karachi stock index returns. By contrast, El-Nader and Alraimony's (2012) research aimed to examine the relationship between Amman stock market index returns and macroeconomic variables by applying the normality test, unit root tests, OLS, and ARCH/GARCH models. The findings revealed that nominal interest rates, real exchange rate, inflation, and real money supply have a negative effect, whereas the increase in the real gross domestic product has a positive effect.
H02: There is a statistically significant relationship between consumer price index and the stock market index returns.
Kyereboah-Coleman and Agyire-Tettey (2008) found that exchange rate losses do not affect the equities on the market, and exchange rate losses benefit the investors as a result of domestic currency depreciation. Against Ouma and Muriu's (2014) study that examines the macroeconomic factors of the stock market returns in Kenya, indicated by the NSE-20 share index returns, the researchers found that exchange rates have a negative effect on stock returns, by using the Capital Asset Pricing Model and Arbitrage Pricing Theory, then applying the ordinary least squares technique.
H03: There is a statistically significant relationship between exchange rate and the stock market index returns.
Karunanayake, Valadkhani, and O'Brien (2012) examined the impact of GDP growth rates on stock market returns in four Anglo-Saxon economies by using the multivariate GARCH model.
The main results revealed that GDP growth has an impact on stock market returns, which exist in the US growth towards its stock market. In addition, Taulbee (1997) noted that the greatest economic determinant of stock prices is the real GDP, also real GDP had a positive impact on the US stock market indices.
H04: There is a statistically significant relationship between gross domestic product and the stock market index returns. H05: There is a statistically significant relationship between balance of trade and the stock market index returns.
Furthermore, Nguthi (2013), using the event study methodology to measure the behavior of Nairobi Securities Exchange 20 share index returns around the Kenya general election in March 2013, found that political news affect stock returns, continued rise in stock prices after the event date and the average abnormal returns positive before and after the event date. In addition, Dangol (2008), using the event study methodology to find that new political information has the impact on the Nepalese market prices in average 2 or 3 days from the event date, illustrated that the Nepalese stock market in inefficient but there is a strong relationship between political uncertainty and stock returns generation. Mahmood, Irfan, Iqbal, Kamran, and Ijaz (2014) examined the impact of political events on the KSE-100 index returns, fifty major political events are selected to calculate the abnormal returns, event study methodology is applied. The results of each window verify that negative abnormal returns are observed a few days before, and few days after an event happens. The study revealed that there is a significant relationship between political events and stock market returns.
H06: There is a statistically significant relationship between political events and the stock market index returns.
Based on that, the study considered the following questions: Index value Real market index at month .  Dangol (2008), the events are chosen subjectively during the study period but tried to test the events that have caught the media's attention and could be important from the investors' point of view. In this study, many political events are selected, as identified in Table 1.

Research model for macroeconomic factors
Following the literature reviewed, this study postulates the relationship between the performance of the stock market index returns and selected macro-economic indicators, based on Asset Pricing Theory, as a macro-econometric model modified from the version of the models by Chen, Roll, and Ross (1986), Rjoub, Türsoy, and Günsel (2009), Ouma and Muriu (2014). Thus, we specify the following model to be estimated:

No.
Political events Events' date Market It is important to mention that the data are used in differences to meet the theoretical model of Asset Pricing Theory, and economic time series data were assumed to be stationary (Ouma & Muriu, 2014  after AR The insignificant results of t-test reveal that the market is efficient because it absorbs uncertain information. Thus efficient market hypothesis is true. But if we have significant results of t-test, we find that the market is inefficient and it does not absorb noisy information (Sajid Nazir, Younus, Kaleem, & Anwar, 2014).

Empirical results of macroeconomic factors at Palestine stock market
First, we apply the stationary analysis (unit root tests) using the Augmented Dickey-Fuller test. Table 2 illustrates the results for five macroeconomic variables. Table 2 shows a rejection of the null hypothesis of non-stationarity at the 1% level for all macroeconomic variables except GDP that has a rejection of the null hypothesis of non-stationarity at the 5% level because Prob. value is just about 1.2%. So we can conclude that all of the macroeconomic series are stationary and can continue to estimate the ARCH/GARCH models. Then we apply the correlation analysis in order to make sure that there is no evidence of autocorrelation.
As seen in Table 3, all the correlation coefficient values for the macroeconomic variables are less than 0.24, which revealed that there is no problem of multicollinearity in this model. Now we can proceed with regression model in order to investigate the macroeconomic variable effect on index returns. By applying the autoregressive conditional heteroskedasticity (GARCH) model, the results are described in detail in Appendix (1-6) and summarized in Table 4.

Empirical results of macroeconomic factors at Amman stock market
First, we apply the stationary analysis (unit root tests) using the Augmented Dickey-Fuller test. Table 5 illustrates the results of five macroeconomic variables. Table 5 shows a rejection of the null hypothesis of non-stationarity at the 1% level for all the macroeconomic variables, so we can conclude that all the macroeconomic series are stationary and can continue to estimate the ARCH/GARCH models. Then, we apply the correlation analysis in order to make sure that there is no evidence of autocorrelation.
As seen in Table 6, all the correlation coefficient values for the macroeconomic variables are less than 0.23, which revealed that there is no problem of multicollinearity in this model. Now we can proceed with regression model in order to investigate the macroeconomic variable effect on index returns. By applying the autoregressive conditional heteroskedasticity (GARCH) model, the results are described in detail in Appendix (7-11) and summarized in Table 7.

Empirical results of political events
In this subsection, we showed the effect of political events on general stock index returns, tested by applying the paired t-test on the abnormal returns before and after the event windows of 5, 10 and 30 days for Palestinian and Jordanian general index. The results were illustrated in Table 8.

Palestine stock market
The statistical results showed that both variables consumer price index and gross domestic product have an impact with statistical significance less than 0.05 on the Palestinian general index returns. The value of CPI coefficient is negative. The increase in the cost of living makes the income devoted to consumption purposes rather than saving or investing in market instruments this led to the decrease on demand on shares and stock index returns due to difficulties in the Palestinian economy (Palestine Monetary Authority, 2016). This result is consistent with El-Nader and Alraimony (2012). We also found that value of coefficient for the GDP is negative. The reason for this finding is that despite the increase in GDP and growth in the economy, investment is moving towards other sectors or even saving instead of attracting investment in the stock market. There is difference in the results of this study from the previous studies that have been reviewed, such as Karunanayake et al. 's (2012) study that showed a statistically significant relationship between stock market returns and GDP in Anglo-Saxon economies, and Taulbee's (1997) study that found a statistically significant relationship between US stock market and GDP. The reason of this difference is that these countries have advanced markets, but Palestinian market is less developed stock market and has weak form efficiency. In addition, we noted that there is no significant relationship between the balance of trade and Palestinian general index returns. Mainly this is due to the continuing deficit in the Palestinian balance of trade (Palestinian Central Bureau of Statistics, 2018). Therefore, whether there is the improvement in this factor or notes do not have an effect on the investors' decisions or the demand on shares. This result is consistent with Bhattacharya and Mookherjee's (2001) study that has been applied in India, noting that there is a deficit in the Indian balance of trade. But, by reviewing previous studies that have been applied on advanced countries, we found different results, such as Antonakakis, Gupta, and Tiwari's (2015) study that showed a positive relationship between US stock return and balance of trade. Regarding the industrial production index, the results showed that there is no significant relationship with Palestine general index returns. The main reason for this is the weakness of liquidity in the Palestine stock market and the weakness of investors' confidence in it (Palestine Monetary Authority, 2016), which led to the fact that even the improvement in the macro economy or the decline does not reflect the impact on the Palestinian Stock Exchange. Also, the fact that Palestine Stock Exchange consider as service market rather than industrial market, the number of industrial companies is just 12 out of 48 companies listed on Palestine Stock Exchange (Palestine Exchange, 2018). It is worth mentioning, the political risk factor, which has a big role. This result is consistent with Zhu (2012). The result also showed that the exchange rate has an impact with statistical significance less than 0.10 on the Palestine general index returns. We can see that value of coefficient for the exchange rate is negative. We can justify this result by saying that when the exchange rate of the dollar -shekel increase, the value of the money that will be converted from shekels to dollars will be decreased, as the prevailing currency is the shekel, which is reflected negatively on the demand for shares and the stock index returns. This result matches with Ouma and Muriu (2014).
The results showed that the consumer price index has a negative significant effect on Palestine insurance index return, industry index return, service index return and bank index return, also the exchange rate has a negative significant effect on Palestine investment index return and bank index return, in addition to gross domestic product that has a negative significant effect on Palestine service index return, moreover balance of trade has a positive significant effect on Palestine industrial production index return, but industrial production index has no significant effect on Palestinian sectors' indices returns.

Amman stock market
The statistical results illustrated that exchange rate has an impact with statistical significance less than 0.05 and gross domestic product has an impact with statistical significance less than 0.01 on the Jordanian general index return. We note that the value of coefficient for the exchange rate is negative. The explanation of this result is mainly as follows. Since Jordan is an importing country and not an exporter (Jordanian Department of Statistics, 2018), the increase in the US dollar index is negatively reflected in terms of increasing the cost of sales, leading to a decline in the demand on shares and decrease on the stock index return. This result is consistent with Ouma and Muriu (2014). We can also note that the value of coefficient for the gross domestic product is negative. The reason behind this is that despite the increase in GDP and growth in the economy, investment is moving towards other sectors or even saving instead of attracting investment in the stock market. The difference in the result of this study from the previous studies that have been reviewed relates to the fact that the Jordanian market also is considered as a weak form of efficiency. We can just note this negative relation that the The main results revealed that the consumer price index has a negative significant effect on Jordanian service index return and insurance index return.
The exchange rate has a negative significant effect on Jordanian bank index return, insurance index return, service index return, and industry index return. In addition, gross domestic product has a negative significant effect on Jordanian bank index return, and service index return. On the one hand, the balance of trade and industrial production index have no significant effect on Jordanian sectors' indices returns.

Political events
In reference to the political events and through the statistical results, we found that there is a significant effect of political events on both Palestine and Amman stock markets. . This event showed a statistically significant response at 5% for Palestine market in event window of 30 days. On March 2, 2016, the end of the security operation on the attacks in the city of Irbid, the death of a security officer in the rank of captain, while killing seven of the armed group. The Jordanian market did not show any significant response to this event.
On December 18, 2016, attacks and shootings in the historic Karak Fort in Karak, which resulted in the killing of 10 people, including seven security men and a Canadian tourist. This event showed a statistically significant negative response at 10% for Amman Stock Exchange in event window of 30 days. The inauguration of the 45th President of the United States took place on January 20, 2017. The Palestine market did not show any significant response to this event. However, the Jordanian market showed a statistically significant positive response at 5% in event window 30 days. In June 2017, Saudi Arabia, the United Arab Emirates and Bahrain, as well as Egypt, cut off diplomatic relations, transportation and trade links with Qatar.
The Palestine market showed a statistically significant response at 1% in event window of 10 days, and at 5% in event window of 30 days. Regarding the Jordanian market, this event had a significant effect at 10% on Amman general index return in event window of 10 days. The last event tested was, the Palestine popular revolution, which took place in the Palestine, following the recognition by US President that Jerusalem is the capital of Israel. This event had a significant effect at 5% on Palestine market in the event window of 30 days. The main results concluded that the political events have a significant impact on the stock market returns. For events that are insignificant, we can assume that they are less important from other events.

CONCLUSION
The main results conclude that the macroeconomic factors and political events have a significant impact on the Palestine and Amman stock market returns. Both Palestine and Amman Stock Markets are inefficient and the markets do not absorb uncertain information and noisy events. Based on the study results, it is necessary to note the importance of implementing the prudent economic policies, due to their impact on the stock markets and investors' decisions, such as working on reducing the inflation rate, since we found that it has a significant effect on stock returns. Moreover, we recommend developing a strong financial strategy that can face challenges to protect investors from the political events, to reduce their impact on stock price returns. Also, encourage studying the impact of macroeconomic factors on the stock returns on a wider scale as a study of the impact of other economic factors, such as unemployment rate and interest rates. And the effect of additional political events on stock market return on a large scale, in order to have a more comprehensive view and to generalize the results. Furthermore, we recommend extending the study to include other markets, in both advanced and developing countries, to get different results and compare them.