Research On Analysts' Earnings Forecast Bias Pattern In China

Overall optimistic bias exists in analysts'earnings forecasts. The article chooses domestic analysts'forecasts to listed companies in Shenzhen main board and Shenzhen small and medium-sized enterprises board from 2005 to 2009 to be research samples to find out whether the degree of optimism of analysts'earnings forecast varies for different group of firms, characterized by their differing information disclosure quality and financing need. We find that low-information firms tend to have larger optimistic forecast bias than those for the high-information firms, and earnings forecasts are more optimistic for firms with higher financing needs. These two factors form the systematic forecast bias pattern and the forecast bias pattern changes slightly when earnings report date approaches. We investigate the difference between star analysts and non-star analysts'earnings forecast behavior and find that there is some difference in relative percentage of each type of firms covered by star analysts and non-star analysts. Whether star analysts provide more accurate or more optimistic forecasts than non-star analysts depend on different firm characteristics. Star analysts tend to issue more accurate forecasts to firms with high information disclosure quality or low financing need, and equally optimistic forecasts to firms with low information or high financing need, in comparison with non-star analysts. Our results suggest that conflicts of interests affects analysts'forecast behavior, and the systematic bias pattern is the consequence of balance between maintaining relationships with institutional investors, listed companies'management, and investment banking department.