Investors’ opinion divergence and the stock market

Traditional finance is based on the market efficiency hypotesis (EMH), according to which security prices always full reflect the available information (Fama, 1970). The efficient market theory is founded on three theoretical assumptions. The first premise leans on investor rationality, implying stock assets to be valued at fundamental values. The second assumption is relative to arbitrage. In particular, the theory poses the offsetting of noise traders’ demands by arbitrageurs quickly, with no significant impact on prices. The third assumption is relative to independent errors accross investors. In particular, noise investors are assumed to trade randomly and thus their mistakes cancel out in equilibrium. The behavioral finance literature challenges the notions of efficient markets. In fact, pysicological evidence shows that many investors’ conduct is not rational in relation to the normative model. For example, noise traders are likely to evaluate gains and losses in relation to some reference point and to show loss adversion (Kahneman and Tversky, 1979). In addition, in forecasting future stock variables, they systematically violate Bayes rule (Kahneman and Tversky, 1973). In relation to the second theoretical argument of traditional finance, numerous studies have questioned the assumption of arbitrage’s power to eliminate mispricing. For example, Shleifer and Vishny (1997), Gromb and Vayanos (2002) and Chen et al. (2002) provide compelling theoretical explanations of why arbitrageurs may fail to close the arbitrage opportunity. Indeed, arbitrage could be risky and thus limited. Close substitutes for securities affected by noise investors are not often available, and even if they are available, rational arbitrageurs face fundamental risk. In fact, substitutes may be imperfect; even if they are perfect, mispricing could become worse before desappearing, implying the so-called noise trader risk (De Long et al. 1990). In relation to the the third assumption, according to psycological evidence, most noise investors deviate from rationality in the same way. For example, people share similar heuristics and systematic biases have been confirmed in numerous experimental psycological studies (Hirshleifer, 2001). Considerable research addresses the first two theoretical arguments, while the independent mistakes assumption has been little investigated. Mistakes are independent if investors form their demands according to their own beliefs. For all the above mentioned reasons, we believe that is worth conducting further investigations on the divergence of opinion among investors. Consistently, the aim of the thesis is to assess the role of investors’ opinion divergence in the stock market, especially in stock market volatility and stock market efficiency. We use divergence of sentiment captured from mass media content as proxy for investors’ opinion divergence. This objective can be disentangled in the following research questions: 1. To what extent divergence of sentiment affects stock market volatility? 2. Does divergence of sentiment on day t relate to stock market volatility on day t+1? 3. Is the cited relation stronger when individuals are more likely to trade? 4. To what extent sentiment affects stock market volatility? 5. Does sentiment on day t relate to stock market volatility on day t+1? 6. To what extent divergence of sentiment affects stock market efficiency? The remainder of the thesis is organised as follows. Chapter 2 provides a theoretical introduction to the investigated phenomenon, defining the concept of investors’ opinion divergence. Then we present a literature review of studies dealing with the investors’ opinion divergence phenomenon. We pay attention to theoretical contributions and empirical analyses about the relation between divergence of opinion and stock market variables. Chapter 3 presents the empirical analysis conducted in order to answer the research questions from 1 to 5. We follow the behavioural approach to finance. We first define our concept of investors’ opinion divergence, then we construct several proxies in order to relate them to stock market volatility, as measured by implied and realized volatility. We focus on the U.S. market for the period ranging from 2008 to 2013. In addition we conduct an analysis about the relation between investor sentiment and stock market volatility. Chapter 4 presents the empirical analysis in order to answer the sixth research questions. In particular, we explore the relation between investors’ opinion divergence and stock market efficiency. This analysis can be considered the first exploratory study providing the first insights about this relation. Main findings, considerations about the limits of the thesis and possible further developements are finally presented in the Concluding Remarks.