The effect of ability, independence, and investor sentiment on analysts’ propensity to use their own earnings forecasts in stock recommendations

This paper examines the factors that affect analysts’ propensity to use their own earnings forecasts in stock recommendations. We find that ability, independence, and investor sentiment all contribute to the inefficiency with which analysts incorporate their own earnings forecasts into stock recommendations. The transformational inefficiency is higher for analysts who have a shorter history of following a stock, work for brokers that have investment banking business or a larger brokerage operation, and follow firms that have experienced larger equity sales by either the firm itself or its officers and directors, have higher institutional ownership, and during years of extreme investor sentiments. There is evidence that analysts’ transformational inefficiency reduces stock price efficiency with regard to earnings forecasts. We wish to thank Ashiq Ali, Qi Chen, Mike Crawley, Hemang Desai, Mei Feng, Jeff Hales, Lisa Koonce, Henock Louis, Michele Meoli, Karl Muller, Robert Freeman, Neale O’Connors, Greg Sommers, Hun Tong Tan, Ramgopal Venkataraman, Yanfeng Xue and workshop participants at the American Accounting Association annual meeting, Beijing University, the HKUST Summer Symposium, the Lone Star Accounting Research Conference, the Pennsylvania State University brownbag seminar, City University of HK, Nanyang Business School, the University of Texas at Austin, the Asian Finance Association annual meeting, and the Seventh Accounting and Finance Conference at Xiamen University for helpful comments. We thank Alexander Ljungqvist for sharing his data on the number of sales representatives employed by brokerage firms. Corresponding author Bin Ke Pennsylvania State University 316 Business Building University Park, PA 16802 USA 814-865-0572 (phone) 814-863-8393 (fax) [email protected] (email)