What Do Analysts’ Provision Forecasts Tell Us About Expected Credit Loss Recognition?

52 Pages Posted: 5 May 2020

See all articles by Anne Beatty

Anne Beatty

Ohio State University (OSU) - Department of Accounting & Management Information Systems

Scott Liao

University of Toronto - Rotman School of Management

Date Written: June 28, 2019

Abstract

We examine the incremental predictive ability and information content of analysts’ provision forecasts to explore the potential effects of the FASB’s new current expected credit loss (CECL) accounting method. Controlling for the recognized loan loss provision, consensus analyst provision forecasts are incrementally associated with future non-performing loans (NPLs) and equity markets returns. This incremental association is increasing in banks’ unconstrained ability to estimate future losses as evidenced by their loan fair value disclosures and the extent of the constraints imposed by the incurred loss model measured by the fraction of heterogeneous loans individually reviewed for impairment. For a sample of analysts that also forecast NPLs, the association between the individual analyst’s provision forecast and future NPLs is increasing in the analyst’s comparative NPL forecasting accuracy. Finally, we find that the association between the analyst’s provision forecast and future NPLs is increasing in EPS forecast errors but decreasing in target price forecast errors. Together these results suggest that the reported provisions under the current incurred loss model do not fully incorporate banks’ information about future losses and that the extent to which this occurs depends both on banks’ ability to forecast expected losses and the constraints imposed by the incurred loss model. These results suggest that analysts forecast future losses beyond those reflected in the recognized provision and that the incremental information in these forecasts and potential CECL provision timeliness effect is greater when the constraints imposed by the incurred loss model are greater.

Keywords: CECL, Expected Loss Model, Bank Accounting

JEL Classification: M41, G21

Suggested Citation

Beatty, Anne L. and Liao, Wei-Yi (Scott), What Do Analysts’ Provision Forecasts Tell Us About Expected Credit Loss Recognition? (June 28, 2019). Available at SSRN: https://ssrn.com/abstract=3572406 or http://dx.doi.org/10.2139/ssrn.3572406

Anne L. Beatty

Ohio State University (OSU) - Department of Accounting & Management Information Systems ( email )

2100 Neil Avenue
Columbus, OH 43210
United States

Wei-Yi (Scott) Liao (Contact Author)

University of Toronto - Rotman School of Management ( email )

105 St. George Street
Toronto, Ontario M5S 3E6 M5S1S4
Canada

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
239
Abstract Views
848
rank
159,925
PlumX Metrics