The Reaction of Stock Markets in the Gulf Cooperation Council Countries to Economic Policy Uncertainty in the United States

Abstract This study investigates if the changes in economic policy uncertainty in the U.S. can explain the returns on stock markets of Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the United Arab Emirates. The study also examines how the stock market returns of the six GCC countries respond to the changes in economic policy uncertainty in the U.S. The results demonstrate that changes in economic policy uncertainty in the U.S. are not significantly linked with the returns on all the stock markets except Oman stock market, which shows a statistical significant negative relationship with the changes in economic policy uncertainty in the U.S. Controlling for the effects of the U.S. stock market and oil price, returns on all the six GCC markets including Oman show insignificant coefficients. The returns on all the stock markets do not respond to the changes in economic policy uncertainty. The results of Granger causality tests show that the changes in economic policy uncertainty in the U.S. do not cause the returns of all the six GCC stock markets.


INTRODUCTION
By the beginning of the 1980s, six countries in Arabian Peninsula decided to form strong alliance in order to overcome economic and social difficulties they were facing. At that time, Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the United Arab Emirates formed regional intergovernmental political and economic union consisting of all Arab States of the Persian Gulf. This institution is called the Gulf Cooperation Council (GCC). In 1998, total intra-trade by the GCC countries was about $8 billion; while in 2017 it was $175 billion (World Bank, 2010;European Commission, 2018). Right from the beginning the USA has recognized the GCC as an important economic partner. By 1998, total trade between the GCC and the USA was $30 billion, while in 2017 it reached $72 billion (Anthony, 1999;European Commission, 2018). Since trade between the US and the GCC increased significantly, they were all open not just for new business opportunities, but also for spillovers. The financial crisis of 2007-2008 and the 2010s oil gluts produced ______________________________________________________________________ 2019 / 33 23 numerous effects on the economies of the USA and GCC, which in the end slowed down the increase in trade.
When it comes to spillover effects, they usually start with existence of uncertainty among analysed subjects. Bernanke (1983) sees uncertainty as a huge problem for companies that are forced to delay investments when the costs of planned projects face unpredicted volatility. This volatility eventually goes from micro to macro level, which can cause spillover effects around the world. Some authors point out that other reasons for the existence of economic uncertainty are precautionary spending cutbacks by households, upward pressure on the cost of finance, managerial riskaversions and interactions between nominal rigidities and search frictions (Panousi & Papanikolaou, 2012;Pastor & Veronesi, 2013;Gilchrist, Sim & Zakrajšek, 2014;Basu & Bundick, 2014;Leduc & Liu, 2015). On the other hand, there are studies that emphasise detrimental economic effects of monetary, fiscal and regulatory policy uncertainty (Friedman, 1968;Rodrik, 1991;Higgs, 1997;Hassett & Metcalf, 1999). Pastor and Veronesi (2012) developed a model that helped in understanding the connection between market fluctuations, policy uncertainty and stock market volatility in order for policymakers to better prepare for potential shocks. Born and Pfeifer (2014), also Fernandez-Villaverde et al. (2015) studied multi-factored policy uncertainty by using the DSGE models. They found the existence of moderately negative effects.
Due to economic shocks that hit numerous countries at the same time, many authors have started to pay more attention to international spillovers. With the reduced form of GARCH models, a group of authors detected some spillovers from the U.S. to Japan and the UK equity markets, both for returns and in particular for conditional volatility (Hamao, Masulis & Ng, 1990;King, Sentana & Wadhwani, 1994;Lin, Engle & Ito, 1994). Becker, Finnerty and Friedman (1995) found that US news and information were responsible for the existence of spillovers between the U.S. and UK stock markets. Ehrmann and Fratzscher (2004) confirmed that news and information could have strong impact on changing conditions in some markets, but these spillovers were much stronger from the U.S. to Euro market than in the opposite direction. Since the U.S. has strong and well-established financial markets, it is not unusual that many well-known global spillovers have originated in the U.S. (Goldberg & Leonard, 2003;Bayoumi & Swiston, 2007;Ehrmann & Fratzscher, 2009;Kim, Salem & Wu, 2015;Fratzscher, Lo Duca & Straub, 2018). At the regional level, customs, monetary and economic unions are very exposed to spillovers between member countries. Fadejeva, Feldkircher and Reininger (2017) indicated that the emerging European economies are the most vulnerable economies to all shocks either from the U.S. or euro zone. Spillover effects are sometimes larger than domestic response which is why developing economies are struggling to overcome external economic shocks. Sum (2012) empirically proved that changes in the European economic policy uncertainty have negative effects on stock markets in the European Union, Norway, Russia, Switzerland, Turkey and Ukraine. Beirne et al. (2010) examined global and regional spillovers for 41 emerging economies in Asia, Europe, Latin America and the Middle East. They found that global spillovers had a stronger impact on the economies in Asia and Europe, while regional spillovers produced a more significant effect than global ______________________________________________________________________ 2019 / 33 24 spillovers in Latin America and the Middle East. Sugimoto, Matsuki and Yoshida (2014) analysed spillovers in Africa during the U.S. financial crisis and the European sovereign debt crisis. The study found the existence of strong and negative relationship between global spillovers and African stock markets. As for regional spillovers, their impact in Africa does not produce significant effects. Another study showed that the South African regional market was the most influential in propagating shocks to other African markets (Boako & Alagidede, 2018). Several studies found the existence of strong and statistically significant impact of spillover effects among the regions in China and Spain (Johansson & Ljungwall, 2009;Zhou, Zhang & Zhang, 2012;Ouyang & Fu, 2012;Marquez, Ramajo & Hewings, 2015;Marquez-Ramos, 2016).
This study investigates the effect of changes in economic policy uncertainty in the U.S. on the returns of stock markets in Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the United Arab Emirates. The study also examines how stock market returns of the six GCC countries respond to changes in economic policy uncertainty in the U.S. The Granger causality test is carried out to know if changes in economic policy uncertainty in the U.S. cause returns of the GCC stock markets. This study will help in providing a better understanding of the GCC country financial markets and their relationship with the U.S.

METHODOLOGY AND DATA
In order to conduct research, appropriate data should be collected. The study used monthly data on economic policy uncertainty (EPU) in the U.S. and GCC stock market indices from 2004:7 to 2018:8. The monthly data on economic policy uncertainty in the U.S. were collected from the Economic Policy Uncertainty Index website at http://www.policyuncertainty.com. The monthly data on stock markets of the GCC countries, stock market of the United States (S$P500) and oil price (Brent) were obtained from Bloomberg.
Hypothesis that will be tested in this paper is: H1: Changes in economic policy in the U.S. have negative impacts on the stock markets of the GCC countries. First, descriptive statistics and correlation analysis are carried out. After that, time-series regression analysis is implemented. Time-series regression will help us understand the impacts of economic policy uncertainty in the U.S. on each of the GCC countries. In this case, economic policy uncertainty in the U.S. is an independent variable, while stock market prices of each individual GCC country is a dependent variable as shown in (equation 1). To control for the effects of U.S. stock market and oil price, returns on S&P500 and Brent are included in time-series regression analysis. Vector autoregressive (VAR) model is employed to know how returns on each GCC country respond to a shock in economic policy uncertainty in the U.S. Implementing all these steps will help us understand the ways in which GCC countries are exposed to economic shocks in the U.S. and how they behave under these circumstances.
= + ∆ + , ______________________________________________________________________ 2019 / 33 where Rt is return on a country's stock market index in month t; Rt−i is return on a country's stock market index in month t−i; SPt is return on the S&P 500 index in month t; SPt−i is return on the S&P 500 index in month t−i; BREt is return on Brent oil price index in month t; BREt−1 is return on Brent oil price index in month t−i; ΔEPUt is change in the U.S. economic policy uncertainty index in month t; ΔEPUt−i is change in the U.S. economic policy uncertainty index in month t−i. Table 1 shows the descriptive statistics of all the variables under study; the correlation between the variables are presented in Table 2.   Table 2 indicate a negative relationship between economic policy uncertainty in the U.S. and all the six GCC countries. As reported in Table 3, monthly returns on each GCC stock market are regressed on economic policy uncertainty in the U.S. using time-series OLS regression; the results in Table 3 present a statistical insignificant negative correlation for all returns on the GCC markets except Oman, which is statistically significant with a negative coefficient. Thus, economic policy uncertainty in the U.S. affects return on the Oman stock market negatively.  In order to control for the effect of U.S. stock market and oil price, monthly returns on each GCC stock market is regressed on economic policy uncertainty in the U.S. (EPU), return on the U.S. stock markets (S&P500) and oil price (Brent). The results of full sample data from 2004:7 to 2018:8 show that all the coefficients of the GCC markets are statistically insignificant, i.e., economic policy uncertainty in the U.S. has no impact on all the six GCC countries.      The results of VAR analysis indicate that the returns on all the six GCC markets do not respond to a shock in economic policy uncertainty in the U.S. (Figs. 1-5). The results of Granger Causality tests in Tables 5-9 reveal that economic policy uncertainty in the U.S. does not cause returns of all the six GCC markets.