Financial Integration of Stock Markets in the Gulf: A Multivariate Cointegration Analysis

ABSTRACT This paper uses multivariate cointegration techniques developed by Johansen (1988, 1991, 1992b) and Johansen and Juselius (1990) to test for the existence of long-term relationships between share prices in the gulf region. Using a vector-error-correction model the paper also investigates the short-term dynamics of prices by testing for the existence and direction of intertemporal Granger-causality. The analysis of weekly price indices in Kuwait, Bahrain, and Oman stock markets shows that: (i) share prices are cointegrated with one cointegrating vector and two common stochastic trends driving the series, which indicates the existence of a stable, long-term equilibrium relationship between them; (ii) prices in Kuwait and Bahrain are adjusting to the long-term equilibrium state whereas prices in Oman are exogenous; and (iii) prices are not affected by short-term changes but are moving along the trend values of each other. Therefore, information on the price levels is helpful for predicting their changes. JEL: G15 Keywords: Gulf Cooperation Council; Share prices; Cointegration; Causality; Prediction. I. INTRODUCTION Stock markets in the Gulf Cooperation Council (1) (GCC, thereafter) countries have changed drastically over the last five years; privatization programs and new issues of shares have come to surface. The development in computer-based trading and the interlisting of shares on their stock markets is the concern of authorities in these countries. In the past few years there was cooperation between stock markets in the region. This cooperation has come in the form of cross listing. In March 1995, the Bahrain stock market established full linkage with Omani stock market. This gave opportunity for investors in both countries to deal in 110 listed shares. Bahrain Stock market is also considering listing other GCC companies. Kuwait stock market is not linked to any of the stock markets in the region but it has allowed GCC nationals to own shares of the Kuwaiti companies listed on the exchange, with a restriction that they cannot hold more than 49% of the total shares outstanding of banks and insurance companies. Empirical evidence by Swanson (1987) suggests that world stock markets are becoming more integrated. This might be true for stock markets in developed countries as in United States, Japan, United Kingdom and other European countries, but to what extend this might be the case for stock markets in developing nations, especially in the gulf states where share dealings is a new phenomena. Hence, the purpose of this paper is to test the efficient market theory (2), i.e., to investigate empirically the existence of long-term relationships between share prices in the GCC stock markets. For this we use recent multivariate cointegration techniques developed by Johansen (1988, 1991, 1992b) and Johansen and Juselius (1990). In addition, short-term relationships are investigated using the Granger-causality test. As Granger (1986, 1988) points out, if two variables are cointegrated then granger-causality must exist in at least one direction. It has been well established in the literature that international stock markets are becoming more integrated, and equity prices on these markets are exhibiting long-run relationships. This is due to the increasing liberalization and globalization of capital markets. Numerous numbers of studies has tested the relationships among international stock markets, however, most of these studies focused on highly developed stock markets in USA, London, Japan, Germany, Australia, Singapore and Hong Kong. For example, Chou et al (1984) base their work on weekly data of the stock market indices of the United States, the United Kingdom, Japan, France, Germany and Canada. Moreover, Kwan et al (1995), base their study on the monthly data drawn from nine major stock market indices of Australia, Hong Kong, Japan, Singapore, South Korea, Taiwan, the United Kingdom, Germany and the United States. …

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.