تنظیمات سهام نازل شده سرمایه گذاران فردی: شواهدی از بازارهای سهام چین

We examine the stock preferences of Chinese individual investors as revealed by their executed trades. Results show that their stock preferences vary with wealth levels. Wealthier individuals prefer stocks with high liquidity and volatility, greater state-ownership, high growth potential, and strong past performance. Less wealthy individuals, however, prefer stocks with high beta, high liquidity, poor past performance, low price, and small capitalization. Overall, the investment choices of Chinese individual investors do not necessarily suggest only behavioral biases as existing studies might have implied, but instead, they reveal to some extent the rational investing behavior of Chinese individual investors.

Empirical research has shown that equity holdings of financial institutions reveal the institutions' preferences for certain stock characteristics. Studies on the U.S. and other developed countries show that financial institutions generally prefer stocks with large market capitalization, highly liquid stocks, and stocks with international visibility and investor recognition.2 However, similar research on the stock preferences of individual investors is lacking. The goal of this paper is to examine whether trading activities of individual investors reveal any of the investors' preferences for stock-specific characteristics. In particular, we examine whether individual investors with varying levels of wealth exhibit similar or different stock preferences. In addition, our study also includes analyzing the revealed stock preferences of a small group of institutions. Studying the preferences of both individual and institutional investors from the same market allows us not only to compare the preferences of the two types of investors on the same basis, but also to determine the robustness of prior results on financial institutions. This paper makes several contributions to the existing literature. First, this study represents the first to examine for evidence of stock preferences of individual investors, as revealed by their trading activities. In particular, we examine whether the trading decisions of individual investors are influenced by the fundamental financial information of a firm. To the best of our knowledge, none of the existing research has investigated whether individual investors exhibit preferences for basic stock characteristics, such as a stock's riskiness, liquidity, growth potential, or past performance. Instead, the current finance literature has largely looked at the behavioral biases associated with the trading behavior of individual investors. Investors, particularly individual investors, tend to exhibit various types of behavioral biases that may lead them to make cognitive errors (Hirshleifer, 2001). Also, individual investors tend to use “heuristics” in making financial decisions (Kahneman and Tversky, 1973 and Tversky and Kahneman, 1974). There is a voluminous amount of research that examines effects of cognitive biases on investor decision-making processes. For example, Odean (1999) finds that individual investors tend to trade excessively, are more risk taking, and make poor investment decisions. Barber and Odean (2003) argue that because individuals face great difficulty in searching the stocks they can potentially buy, they tend to engage in attention-based buying. Sirri and Tufano (1998) provide evidence that fund investors' buying decisions are consistent with reducing search costs, and that media coverage of mutual funds is an important determinant of consumer decisions. Zhu (2002) and Ivkovic` and Weisbenner (2005) find that individual investors exhibit local bias by investing a disproportionate share of the equity portfolios in geographically proximate stocks. Zhu argues that accounting numbers and information asymmetry matter less to individual investors' local bias, whereas Ivkovic and Weisbenner contend that individual investors' preferences for local stocks are induced primarily by their superior information about local stocks. Dhar and Zhu (2002) find that sophisticated individuals, in particular those who are wealthy and work in professional occupations, show significantly smaller local bias. Chen et al. (2004) extend existing U.S. studies by using Chinese brokerage accounts data and find that Chinese individual investors tend to be overconfident, inclined toward a disposition effect, and exhibit representativeness bias. The implication of all these prior studies is that individual investors are less likely to make rational decisions. As such, their decisions are not influenced by stock fundamentals that are related to a firm's financial information or future prospects. Instead, individuals' decisions are more likely driven by their emotions, e.g. acting as noise traders, and hence a stock's fundamental financial information plays no role in their trading decisions. Our study therefore allows us to explore this implication for the investment behavior of individual investors. Second, we examine a wide heterogeneity of active individual investors in Chinese equity markets. Our analysis employs a new unique data set that contains 64.2 million trades of A-Shares initiated by about 6.8 million active local investors.3 The data are compiled by the Shanghai Stock Exchange (SHSE) and are available to us for the period April 2001 to April 2002. SHSE is one of the two domestic stock exchanges in Mainland China; the other is Shenzhen Stock Exchange (SZSE). SHSE and SZSE were established in December 1990 and July 1991, respectively. The development of the equity markets over the past decade has been rapid and robust. For example, the ratio of market capitalization to gross domestic product grew from 4% in 1992 to 53% in 2002. The Chinese equity markets therefore offer an excellent laboratory for the purpose of our study for several reasons. One, Chinese equity markets have one of the largest individual investor populations in the world. By the end of 2002, the number of individual investor accounts opened at the Chinese stock exchanges reached 68.5 millions (Chinese Securities Depository and Clearing Co. Ltd, Report 2002). Individual investor accounts make up of 99.5% of the total number of investor accounts in the markets, whereas institutional accounts form merely 0.5%. Our sample of active investors contains by far the largest number of individual investors ever studied, compared to the fewer than 100,000 household accounts examined by all existing U.S. studies (Odean, 1999 and Barber and Odean, 2000, among others) and the fewer than 50,000 individual investors by Chinese studies (Chen et al., 2004 and Feng and Seasholes, 2004).4 Two, the turnover of our sample accounts for approximately 34% of the total market turnover, indicating that our sample of investors have been actively trading in the markets.5 Given such an enormous heterogeneity of investors in the markets, our sample allows us to perform a comprehensive and thorough analysis of the investment patterns of individual investors. Three, our panel of data increases our power to detect any systematic patterns of stock preferences of individual investors. Even if preferences for certain characteristics affect individual investors' trading activity, we know there are many other important influences on any given trading activity. Some variation in trading activity might be driven by individual investors' behavioral biases as well as economic events and news. Given the high variability of trading activities, it is useful to maximize power by using a large number of trades initiated by a large investing group of individual investors. Fourth, in contrast to earlier studies that examine equity holdings of financial institutions, we examine preferences of investors as revealed by their flows of money to stocks. Our approach of determining stock preferences of investors therefore incorporates the dynamic information that investors employ in their stock trading decisions. Last but not least, our analysis also includes the remaining 0.5% of institutional accounts in the sample. Analyzing this relatively smaller group of investors facilitates comparisons of results not only between Chinese institutions and Chinese individual investors, but also between Chinese institutions and those from developed markets that have been previously studied. While Chinese equity markets have experienced a phenomenal growth in the past decade, they are still considered to be in their nascent stage of development when compared to those developed markets. Thus, our findings of preferences of Chinese institutions for stock characteristics would allow us to determine whether such preferences are peculiar to financial institutions from an emerging market, or whether existing results are robust even to emerging markets. Our results find strong systematic stock preferences of Chinese individual investors, as revealed by their trading activities. Less wealthy individual investors tend to favor stocks with high betas, low market prices, high turnover, and small market capitalization, and stocks that have performed poorly in the past year. In contrast, wealthier individual investors prefer highly liquid and volatile stocks, stocks with high state-ownership, and stocks that have performed well over the past year. Less wealthy individual investors trading more in small and low-priced stocks reflects in part the regulatory restrictions imposed by Chinese securities authorities. In China, short selling and margin trading are prohibited. Hence investors can sell only stocks they own and can buy stocks with immediate cash balances in their brokerage accounts, or with funds from personal borrowing. Such regulatory restrictions have a stronger impact on less wealthy than wealthier individuals, causing the trading preference of the former to tilt toward small-cap and low-priced stocks. Purchases of such stocks typically require smaller cash outlays. In general, our finding that the set of predetermined stock characteristics can explain a large cross-sectional variation of money flows from Chinese individual investors, especially those who are less wealthy, suggests that the investment choices of individual investors cannot be merely driven by emotions or behavioral biases. Instead, we are inclined to interpret that Chinese individual investors in general act more like rational investors than “noise” traders. Finally, similar to institutions in developed markets, Chinese institutions also prefer large firms and firms with high earnings per share and large volatility. In contrast to Chinese individual investors, Chinese institutions are not influenced by stock-specific characteristics such as the past performance of the stock, betas, book-to-market equity ratio, and state-ownership in a firm, but are affected by the listing period of a stock. Chinese institutions invest more in firms whose stocks having shorter listing periods, as these firms have met the more recently stringent disclosure requirements and monitoring procedures required by Chinese securities authorities. Therefore, institutions would view the newer-listed firms to have better accounting standards and better management. The remainder of the paper is organized as follows. The next section provides a brief review of the related studies in the existing literature. Section 3 contains a detailed description of our unique data set and the construction of all variables employed throughout the study. Section 4 reports the empirical results, and the final section summarizes the paper.