We develop a measure of market integration score by extending the methodology of Akdogan [J. Portf. Manage. (1996) 33; J. Portf. Manage. (1997) 82] to reflect the degree of integration of domestic equity markets with other markets in and beyond the region. We empirically estimate the integration scores for a sample of six Latin American markets between January 1988 and December 2001. We find a trend towards increased regional integration relative to global integration until the mid-1990s. A distinct change in trend is noted during the second half of the 1990s, with global integration proceeding faster than regional integration.
The emerging equity markets in a number of developing countries in Asia, Eastern Europe, Latin America, and the Middle East grew rapidly during the second half of the 1980s and throughout most of the 1990s. This growth was made possible to a significant degree by market-oriented, financial liberalization policies. Despite temporary fluctuations during the Mexican peso crisis of 1994–1995 and the Asian crisis of 1997–1998, portfolio equity flows into developing economies increased from US$140 millions to US$51 billions1 over the 15 years from 1985 to 2000. This rapid growth of emerging equity markets raises the question as to whether these markets are becoming increasingly integrated, both with respect to each other and with respect to the developed equity markets. This is an important question to the extent that international portfolio investment strategies depend on the degree of integration of these markets.
There exists an extensive literature on the topic of international equity market integration. Only a handful of studies, however, have compared the degree of financial integration of a group of countries on a global level with that on a regional level. Yet the growing tendency among nations in recent years has been to form regional economic blocs. The objective of this study is to develop a theoretical measure of market integration by extending Akdogan, 1996 and Akdogan, 1997 to reflect the degree of integration of domestic equity markets with other markets within their region as well as outside their region, and to empirically estimate the integration scores for a sample of emerging markets in Latin America. Following Barari and Sengupta (2002), we highlight the time-varying nature of integration by calculating integration scores over different time windows. Given that the process of financial liberalization has been gradual, this time-varying approach provides additional insights into the issue by depicting the historical evolution of a country's financial integration, regionally as well as globally. We use a sample of six Latin American countries, namely, Argentina, Brazil, Chile, Colombia, Mexico, and Venezuela. We calculate their respective integration scores against a regional Latin America index as well as a global index (comprising both developed and emerging equity markets). Our findings suggest increased regional relative to global integration for most of the Latin markets during the late 1980s and the first half of the 1990s. However, the pace of global integration accelerated around the mid-1990s and has remained strong in the aftermath of the Asian crisis.
The paper is structured as follows. In Section 2, we provide a brief literature review and discuss the contribution of the present study to the existing body of research on the topic. In Section 3, the regional and global integration scores are theoretically derived. In Section 4, we provide empirical estimation of integration scores using a time-varying approach for the six Latin American markets, and we analyze the results. Concluding remarks are provided in Section 5.
In this paper, we have empirically investigated the time-varying nature of equity market integration using a group of six Latin American markets between January 1988 and December 2001. Following Akdogan (1996), we used a quantifiable measure of market integration that makes use of country β values vis-a-vis a global benchmark while also allowing each market to contribute to global market risk. We extended the basic Akdogan model to measure a market's integration not only with a global benchmark, but also with a regional benchmark. We calculated and plotted the global and regional integration scores and their ratios over different time windows in order to capture the time varying nature of integration.
Our method enables us to comment on the status of regional relative to global integration over time (time series) as well as across markets (cross-section). Our figures based on historical and moving average plots suggest that during late 1980s and first half of the 1990s, there was a move toward regional integration and away from global integration for most of the Latin markets in our sample. Our findings in this respect lend support to previous findings of Heaney et al. (2002) for the same set of Latin markets for a similar time period, using different methodology. Our figures also demonstrate that the pace of global relative to regional integration accelerated around mid-1990s and has remained strong for the most part. Although the timing suggests possible cross-region contagion effect resulting from the Asian crisis, more detailed study is needed along Bracker et al. (1999), Johnson and Soenen (2003), and Pretorius (2002) to identify the causes of increased global integration in the post-Asian crisis period.
