Investor behavior, differential information, and asset pricing

The dissertation investigates how individual investors respond to different information and its implication on capital market asset pricing. The first essay studies individual investors' bias towards nearby companies. Using data from a large U.S. discount brokerage, we find that individual investors tend to invest in companies closer to them relative to the market portfolio. Unlike findings on institutional investors, we find that advantageous information cannot explain individual investors' local bias. Instead, we find that individuals' non-fundamentally based familiarity with local companies and ready reaction to local information are more plausible explanations. Consistent with this hypothesis, we find that individual investors are more likely to invest in remote companies that spend heavily on advertising. Evidence from investors' reactions to earnings announcements also confirms the hypothesis.; The second essay shows that individual investors move together in their domestic equity investments. Using methods designed in Lakonishok et al (1992), we find that individual investors from a large U.S. discount brokerage firm exhibit stronger tendency to move together than institutional investors reported in previous studies. Individuals herd more on large stocks, value stocks and are contrarians in their herding. Investors from large metropolitan areas herd significantly more on local companies than remote ones. Stocks that individuals herd most on buying under-perform those that individuals herd least, suggesting that information is not a major force behind individuals' herding.; The third essay analyzes the trading records of a major discount brokerage house to investigate the disposition effect, the tendency to sell winners too quickly than losers. In contrast to previous research that has demonstrated the disposition effect by aggregating across investors (Odean, 1998), our main objective is to identify individual differences in the disposition bias and explain this in terms of underlying investor characteristics. Using demographic and socio-economic data as proxies for investors' sophistication, we find empirical evidence that wealthier and individual investors in professional occupations exhibit less disposition effect. Consistent with experimental economics, trading experience also tends to reduce the disposition effect.