RDP 2004-05: Big Fish in Small Ponds: The Trading Behaviour and Price Impact of Foreign Investors in Asian Emerging Equity Markets 1. Introduction

The rapid growth of cross-border equity investment in recent years has generated much interest in the behaviour and impact of foreign investors, especially in emerging markets. Foreigners are frequently viewed as influencing prices in these countries and their trading is closely watched. Although there is a growing body of research in this area, there are still many open questions. For example, many observers have labelled foreign investors as momentum investors, but there is limited evidence as to whether the trading of foreigners occurs subsequent to price changes, or is concurrent with them. Furthermore, although there is a literature that attempts to ascertain the effect of greater participation of foreign investors on the volatility of returns, there is actually very limited evidence about the magnitude of the impact of the trades of foreign investors on the level of returns. Finally, although it has been argued that investment in emerging markets is substantially affected by conditions in mature markets (‘push’ factors), it remains unclear whether external or internal (‘pull’) factors are dominant in explaining flows to emerging markets.

The most comprehensive study to date of the relationship between flows and returns is the work of Froot, O'Connell and Seasholes (2001) who use proprietary data for flows from State Street Bank and Trust. One of the strengths of the State Street data is that they are available for a very large number of countries, and Froot et al include 44 countries in their study. However, the data are only a partial measure of the flows of foreign investors, since they relate only to the trades of one particular custodian. Further, the data from State Street are not for the actual trades of foreign investors but are based on contractual settlement dates. Froot et al use data on settlement conventions in each country to infer the dates that trades actually occurred.^[1]

The possible shortcomings of the State Street data suggest that it may be worthwhile to instead analyse precise daily data for the actual trades of all foreign investors. Accordingly, this paper provides new evidence on the determinants of foreign investment flows and the impact of foreign trading on domestic asset prices, using daily data over 1999–2002 on total foreign net inflows into six Asian equity markets. The markets comprise the Jakarta Stock Exchange, Korea Stock Exchange, Philippine Stock Exchange, Stock Exchange of Thailand, Taiwan Stock Exchange, and Kosdaq Stock Market. The sample size of six markets is large enough to provide results that are potentially fairly general, yet is small enough to allow more attention to market-specific data and modelling issues than might be possible in datasets with a larger number of markets. In addition, the use of daily data allows closer analysis of high-frequency relationships between flows and returns than is possible in previous studies using weekly, monthly or quarterly data (e.g., Bohn and Tesar 1996; Brennan and Cao 1997; Clark and Berko 1997; Karolyi 2002; Dahlquist and Robertsson 2004). And because the dataset includes the purchases of all foreign investors, it has broader coverage than datasets that cover only one group of investors – for example US investors in the numerous studies using the US Treasury data (e.g., Bekaert, Harvey and Lumsdaine 2002), or mutual funds (e.g., Borensztein and Gelos 2003) or customers of a particular custodian (Froot et al 2001).^[2]

The use of precise data for the daily trading of all foreign investors allows the discovery of two new ‘stylised facts’ that have not been apparent in earlier work. First, foreigners' flows into several markets show strong positive-feedback trading with respect to foreign (especially US) returns, particularly in the two largest markets. Indeed, analysis using vector autoregressions (VARs) suggests that foreign returns on average explain a greater proportion of the variance of net inflows than domestic returns, suggesting that ‘push’ factors may be as important as, or more important than, ‘pull’ factors in explaining the dynamics of inflows into emerging markets. The paper considers possible explanations for these effects, and suggests it is unlikely to be due to portfolio-rebalancing effects or microstructure effects from differences in the use of limit orders by domestic and foreign investors. Instead, the importance of foreign and domestic returns in driving inflows is more likely because shocks to returns lead some foreign investors to revise their expectations about prospects for emerging markets. This would be consistent with the very strong finding that returns on US technology stocks are most important in explaining inflows into the technology-oriented Korean and Taiwanese markets, where these returns can be viewed as news about fundamentals. Alternatively, the explanation may be more behavioural, and based on the sentiment of foreign investors being affected by returns in both emerging markets and their home markets.

The second major result of the paper is the identification of a strong positive correlation between the net purchases of foreigners in a market and same-day returns in that market. VAR analysis provides some evidence that surprises in inflows have ongoing effects on prices beyond the day of the inflow, though most of this impact is complete within a few days. These results stand in marked contrast with the results of Froot et al (2001) that the contemporaneous price impact of the trades of foreigners is small, but that there is a substantial impact seen in the weeks and months following their trades. The paper discusses the question of what type of model can best explain why net purchases by foreigners tend to be associated with increases in stock prices. The most likely explanation would seem to be that the net purchases of foreigners represent substantial shocks to net investor demand in these markets. However, such a ‘price pressure’ explanation need not be entirely independent of an information explanation – foreign inflows are presumably not completely uninformed, but may be based on perceptions that valuations are cheap or that increased allocations to emerging markets offer other portfolio benefits. Although the estimated price impact of foreign inflows is substantially larger than previous estimates for emerging markets, it is in line with estimates of price impacts of order flow in the US market. Overall, the results of the paper suggest a much larger role for foreign investors and conditions in mature markets than has been suggested by previous work.

Footnotes

However, it is quite possible that this introduces errors, because the data used in the study appear to show trading on all weekdays, including on public holidays in each market, but appear not to show Saturday trading in those markets where trading occurred on Saturdays for some or all of the sample period. [1]

The study that is closest in design to this paper is a concurrent paper by Griffin, Nardari and Stulz (2003) that looks at the five main boards studied here (but not the Kosdaq), as well as the markets of India, Sri Lanka, South Africa and Slovenia. [2]