Browse by author
Lookup NU author(s): Professor Simon Hussain
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
This study provides one of the first insights into how UK brokers' institutional characteristics may impact on the forecasting performance of their financial analysts. The study focuses on brokerage house size and finds it to be a significant factor explaining cross sectional variation in forecasting performance. This is consistent with evidence from several recent US studies (Jacob et al. 1997; Clement, 1999). It is likely that this broker-size effect reflects the resources (human, IT) available to brokers' analysts to support them in their activities. It may also reflect larger brokers' superior access to company managers and information. However, this broker-size effect appears to be significant only for forecasts made at horizons of one year or less. The sign of the earnings change being predicted also has a significant impact: for observations where earnings changes are negative, the broker-size effect is larger than for positive changes, though the effect is significant for both cases. In addition, the form of the model employed here suggests diminishing marginal returns to broker size. More generally, this study reiterates the importance of controlling for the most commonly cited explanatory variables for forecast accuracy, and there is evidence that the heavy industry sectors may be more difficult to forecast, echoing the conclusions of UK studies from the 1980s.
Author(s): Hussain S
Publication type: Article
Publication status: Published
Journal: Accounting and Business Research
Year: 2002
Volume: 32
Issue: 3
Pages: 153-170
Print publication date: 01/01/2002
ISSN (print): 0001-4788
ISSN (electronic):
Publisher: Institute of Chartered Accountants in England and Wales