Estimating the Dynamic Impact of Financial Liberalization on Equity Returns in Southeast Asian Markets

Multinational Business Review, Winter 2008 by Ramcharran, Harridutt, Kim, Doseong

Abstract:

Recent studies of the impact of financial liberalization in emerging markets have not examined the dynamic impact of the liberalization process on equity returns despite the important implications on ongoing reform policies. We analyze six Asian equity markets using a dynamic adjustment model with three independent variables: market capitalization value, price-book value ratio, and price-earnings ratio. We use panel data for the period 1991-2000 and the LSDVR (least square dummy variable regression) approach to identify the timing effects of liberalization. The stability of the model is also tested. The results indicate, in most cases, the significance of all three variables and the timing effects. Evidence of significant structural changes is also supported.

1. INTRODUCTION

The rapid inflow of portfolio investments in emerging equity markets (EEM) in recent years, resulting from financial liberalization policies (the passage of laws that enable foreigners' free access to local capital markets and local investors to foreign markets), has stimulated a plethora of studies investigating the impact of these policies. These studies include Henry (2000), Bekaert and Harvey (2000), Kawakatsu and Morey (1999), Kim and Singal (1997, 2000), and Basu, Kawakatsu and Morey (2000). Two of the major issues investigated are: the pricing efficiency of EEM and the benefits to investors /borrowers after liberalization. Some of these studies have produced mixed results; for example, Kim and Singal (1997) report that EEM have become more efficient based on testing the random walk hypothesis, and Kawakatsu and Morey (1999) find that liberalization does not seem to have improved the efficiency of EEM; i.e., the behavior of prices has not changed because, apparently, EEM were efficient before the actual liberalization period. Two studies find significant benefits from liberalization: first, Henry (2000) finds that the equity price index experiences abnormal returns of 3.3 percent per month during an eight-month window leading up to the implementation of the initial liberalization, and second, Bekaert and Harvey (2000) find a significant decrease in the cost of capital after liberalization, they contend that it results from excess volatility and speculation which enhance informational and allocation efficiency.

The estimation techniques employed in previous studies have not focused on the examination of the dynamics of changes in EEM due to liberalization, and the potential impact on stock prices /returns. This is crucial since the initial liberalization in EEM took place in the late 1980s and continued in the 1990s. During this period EEM experienced considerable return volatility attributed to internal factors (regulatory and economic) and external factors, for example, the greater openness of emerging markets to external shocks (Saunders and Walter, 2002). Bekaert and Harvey (2000) identify and list a spectrum of financial liberalization developments that have affected capital flows in EEM.1 Beim and Calomiris (2001) also state that in the long run investors' perceptions of the values of EEM shares are highly dependent on the success of the countries' liberalization experiment.

This paper, using a dynamic adjustment model, estimates the impact of financial liberalization on equity returns of the six largest Southeast Asian markets (Korea, Philippines, Taiwan, Indonesia, Malaysia, and Thailand) using panel data for the period 1991-2000. The liberalization dates (clustered around the late 1980s and early 1990s) of these markets, as well as some of the reforms, are listed in Appendix A. We focus on Southeast Asian markets because of the region-specific characteristics (geographic proximity for investors, and possible contagion effects) exhibited by the financial crisis in the late 1990s.

The model includes the following: (1) an independent variable indicated by the capitalization value of each market; this variable indicates the changing size of the market due to financial liberalization; (2) two independent variables, the price book value (P/BV) and the price earnings (P/E) ratios of each market; these variables are commonly used in international asset valuation models (Fama and French 1992; Claessens, Dasgupta and Glen 1998); Solnik (1999) also notes that international investors usually show their greatest interest in the P/E and the P/ BV ratios of EEM; (3) the temporal dimension of the liberalization process is examined by the "fixed effects" approach estimated by the least square dummy variable regression (LSDVR) and the significance is tested by applying the Chow test suggested by Baltagi (1995) and the restricted F-test suggested by Gujarati (2003); and (4) the significance of structural changes introduced by the liberalization process (the structural stability of the model) is tested using the Chow (1960) test (see Maddala 2001). The panel data approach is used; two of the advantages, listed in Baltagi (1995) and Gujarati (2003), are: (1) the accommodation of a better study of the dynamics of adjustment, and (2) it takes heterogeneity explicitly into account by allowing for individual specific variables.2 The estimation techniques correct for groupwise (cross-country) heteroskedasticity and contemporaneous cross-country correlation.