Losers, winners and hype: Consumer choice of stocks under uncertainty

Marketers have not looked at how consumers purchase stocks under uncertainty, especially when they use past information as a heuristic for purchases. To judge a stock's investment value individual investors are likely to use past performance of a stock, such as a string of earnings increases or decreases. For example, mutual fund managers regularly advertise the merits of their funds based on past performance.; However, the efficient market hypothesis, which is the dominant paradigm in financial economics, assures us that market prices reflect all past information and future expectations about a stock. Any new information about a stock is reflected almost instantaneously in market prices. In effect therefore, past information, such as a string of earnings increases or decreases, should not affect the price of a stock.; Though traditionally considered a problem in finance, marketing's focus on an understanding of consumers can bring a unique perspective to this problem.; This study proposes a theory of hype to explain how consumers may not make “optimal” stock purchases when faced with past information and uncertain futures. In particular, consumers may not see current prices as fully incorporating information in past successes or failures. Instead they see trends in past information and form biased expectations about future earnings potential. These expectations are biased because they differ from what financial economics predict. The reason may be that consumers misjudge event probabilities because they do not understand the underlying probability distributions. I use the term hype to describe the mental process by which consumers misjudge probabilities and overvalue assets that have a streak of earnings increases or decreases. In the context of the stock market such behavior can result in inflated prices of hot stocks that have enjoyed a string of earnings increases.; I develop a theory and specific hypotheses about the behavior of consumers in such a scenario. I test the hypotheses in four separate studies.