Herding in real estate security analysis

In this study, we examine how the transparency of securities affects herding among security analysts, measured in terms of the prevailing consensus and recent revisions by other analysts. Prior literature suggests a relationship between market information and analysts herding behavior, with the likelihood of herding increasing with the strength of prior public information (Graham, 1999). When aggregate public information is strongly held and reinforced by the actions of the market leader, an individual analyst is less likely to take an opposing view based on private information. Welch (2000) finds that herding toward consensus among analysts is more prevalent in up-markets than in down-markets. Welch argues that there is less independent information in the market (poor information aggregation) when conditions are bullish. Bikhchandani et al. (1992), describe such a consequence of poorer information aggregation as ìfragilityî, associated with more ìfickleî markets. An examination of this literature suggest a question: if greater transparency in a security makes the analystís task easier, is herding toward the prevailing consensus less likely? To investigate this question, we develop the research hypotheses that analyst herding towards consensus will occur less with REIT stocks than non-REIT stocks, and that the positive influence of analysts on the recommendations of other analysts is less when analyzing REIT stocks as compared to non-REIT stocks. Following Emery and Li (2009), we will use data primarily from the Institutional Brokers Estimate System (IBES). IBES provides each analyst's name, brokerage affiliation, earnings forecasts, investment recommendations, and a unique code that allows tracking of analysts even if they change affiliations. Following Welch (2000), we consider a parsimonious parametric specification which examines whether herding does or does not occur and how the transition from one category to another is affected by the prevailing consensus and recent revisions by analysts.