خوشه بندی قیمت و ثبات قیمت سهام  Price clustering and the stability of stock prices

1. Introduction Much of economic theory revolves around the formation of equilibrium prices. However, in practice, frictions might adversely affect the ability of prices to find their equilibrium. For instance, empirical research shows that prices tend to cluster on round increments in both equity and commodity markets (Wyckoff, 1963; Niederhoffer and Osborne, 1966; Ball, Torous, and Tschoegl, 1985; Harris, 1991; Alexander and Peterson, 2007; Ikenberry and Weston, 2008). Explanations for this type of anomalous price behavior generally fall into two camps. The first explanation suggests that investors’ prefer round numbers in attempt to mitigate cognitive processing costs (Wyckoff, 1963; Niederhoffer and Osborne, 1966; Ikenberry and Weston, 2008). The second explanation, which is not mutually exclusive from the first explanation, is predicated on the idea that investors prefer to deal in round prices in attempt to minimize negotiation costs (Ball et al., 1985; Harris, 1991).1 While prior research has documented the presence of clustering in financial markets, few, if any studies, have examined the effect of clustering on the quality of financial markets. The main objective of this paper is to take a step in this direction. In particular, we test the hypothesis that the degree of clustering on round pricing increments leads to less stable stock prices. The theory underlying this hypothesis is based on the notion that the price system transmits information to market participants (Hayek, 1945; Friedman, 1977). When clustering on round increments exists, the lack of granularity in stock prices may reduce the informativeness of prices. Therefore, stocks with more clustering may exhibit higher levels of volatility. The implications of our tests are broad, as they suggest that investors’ preferences for round prices – whether because of cognitive biases or an aversion to negotiation costs – can adversely affect the informativeness of prices and subsequently increase the volatility of stock prices. Besides extending the literature that discusses both clustering and volatility, our tests have important practical implications. Analysts and other investment professionals use models that rely on volatility forecasts (Hamid and Iqbal, 2004). Furthermore, managers attempting to maximize the value of shareholders must also be concerned with the level of volatility in the firms’ stock price, given that volatility can affect the firms’ cost of capital projections. While prior research has found that financial markets exhibit excess volatility (Shiller, 1981), we argue that frictions in how prices are formed can, in part, explain this excess volatility.