Biases and information in analysts' recommendations: The European experience

Abstract

Financial analysts are viewed as playing an important intermediary role in gathering and interpreting data and thus converting it into information that is more useful for, and accessible to, the investment community. In recent years, however, it has become more apparent that the analysts come under much internal and external pressure when making their forecasts and recommendations. Jegadeesh et al. (2004, ‘Analyzing the Analysts: When Do Recommendations Add Value?’ Journal of Finance) have highlighted that this results in the recommendations by the US equity analysts being biased towards large, high-momentum growth stocks which presumably contributed to their finding that the recommendations provided little or no value in their own right. They did find, however, that the analysts' recommendation changes provided useful incremental investment insights. Azzi and Bird (2005, ‘Prophets during Gloom and Doom Downunder’, Global Finance Journal, 15(3)) when evaluating Australian analysts similarly found that it was only the recommendation changes that provided useful information to investors. They also found evidence to suggest that the analysts attempt to adjust the biases in their recommendations over the market cycle. The implication being that biases identified in the Jegadeesh et al. study may have been a reflection of the analysts pursuing the types of stocks that were performing well during the period they examined rather than representing a long-term bias that one might expect to find in these recommendations.

This paper extends the analysis to consider the recommendations made by the European analysts. It is found that, as a group, their recommendations have a similar strong bias towards large, high-momentum stocks and a weaker bias towards growth stocks. Over the sample period of ten years commencing April 1994, no evidence was found to suggest that either their recommendations or changes in these recommendations provided any useful information to investors. When the sample period was divided up into the boom years of the 1990s and the gloom years of the early 2000s, however, some weak evidence was found to suggest that the European analysts adjusted their recommendations particularly towards value stocks that performed best during the gloom years. On initial evaluation, it was found that the recommendations made during the boom years did provide useful insights to investors, but this proved to be only a consequence of the biases in the recommendations and thus suggested little contribution from the analysts. It was also found that the recommendation changes provided useful information to investors during the gloom years but, in this instance, further examination confirmed that this may be reflective of some special skills attributable to the analysts. At the country/region level, some variation was found from the typical biases found in analyst recommendations, with the UK analysts displaying a strong preference for small cap stocks, and the English, German and Italian analysts apparently neutral with respect to the growth and value stocks. The analysis suggested that the recommendations made by both the German and Italian analysts provided useful insights as to future stock performance over and above those that could be explained by the biases in their recommendations. The news was not so good for the UK and French analysts, whose recommendations seemed to provide the market with disinformation as to future stock returns. Overall, the European findings suggest that one would have to take care in extending the original Jegadeesh et al. findings to other markets and other time periods.