Abstract:
This dissertation titled “Informational Content of Financial Experts’ Recommendations and their Impact on Capital Markets” analyzes the informational content of stock recommendations issued by financial experts. The first part of this work (namely Chapter 2 to Chapter 4) concentrates on recommendations issued by journalists who work for Personal Finance Magazines. The second part (namely Chapter 5 and Chapter 6) focuses on research issued by financial analysts that work for investment banks and brokerage houses. Due to increased importance of investments in stocks within the last decade, all kinds of stock recommendations were in great demand from private and institutional investors. Around 2000, it seemed as if financial experts always forecasted the future development of stocks correctly. However, financial experts’ role came under scrunity when markets crashed in 2000 but recommendations still remained positive.
In Chapter 1, we analyze the informational content of buy recommendations issued by Personal Finance Magazines in the years from 1995 to 2003. Based on an event study, we observe significant abnormal returns before and around the publication of the stock recommendations. This is especially true for small stocks and value stocks. Additionally, we find that the short-term price reaction of small stocks and glamour stocks is mainly based on naïve buying pressure.
In Chapter 2, we analyze the long-run performance of buy and sell recommendations based on buy-and-hold returns. As a benchmark, we use the CDAX and, additionally, a portfolio of stocks that are comparable to the sample in terms of size and price-to-book value. Generally, sell recommendations prove to be more valuable for private investors.
In Chapter 3, we focus on how journalists decide on which stocks to recommend. Based on logistic regressions, we find that journalists are prone to primarily focussing on stocks which attract their attention by unusually frequent news disclosure, extraordinary past performance, and excessive trading volumes. Additionally, they tend to recommend the same stocks over and over again.
In Chapter 4, we analyze stock recommendations issued by financial analysts within their reports. For this purpose, we take a random sample of 1,000 analysts’ reports from the Investext database from the years 2002 to 2004. Apart from searching for the general recommendation level, forecasted earnings and target prices, we read every report in its entirety to analyze, along 15 information categories, whether the analyst paints a positive or negative picture of the company. The results show that this information measure is highly significant in explaining the short-run abnormal returns around the publication of the report. At the same time, we show that potential conflicts of interests between the bank and the covered company do not influence the abnormal returns.
Within the last chapter, we analyze target price accuracy and try to distinguish potential factors that are relevant for the accuracy. For this purpose, we define a new measure that evaluates target price accuracy. Results show that accuracy is higher for buy recommendations compared to sell recommendations. Furthermore, we reveal that factors such as overconfidence, detail of analysis (analyst-specific factors), size, and volatility (firm-specific factors) are relevant for explaining the accuracy of target prices.