Are Analysts’ Cash Flow Forecasts Naive Extensions of Their Own Earnings Forecasts?
Andrew C. Call
Arizona State University (ASU) - School of Accountancy
University of Texas at Austin - Red McCombs School of Business
Yen H. Tong
Nanyang Technological University (NTU) - Nanyang Business School
March 1, 2012
We examine the sophistication of analysts’ cash flow forecasts to better understand what accrual adjustments, if any, analysts make when forecasting cash flows. As a preliminary step, we first demonstrate that prior empirical tests used to evaluate the sophistication of analysts’ cash flow forecasts are not diagnostic. We then present three sets of evidence to triangulate our conclusion that analysts’ cash flow forecasts incorporate meaningful accrual adjustments. First, we review a stratified random sample of 90 analyst reports and find that the majority of these analysts include explicit adjustments for working capital and other accruals in their cash flow forecasts. Second, using a large sample of analysts’ cash flow forecasts from 1993-2008, we find that these forecasts outperform time-series cash flow forecasts in correctly predicting the sign and magnitude of accruals. Finally, we find a significant market reaction to analysts’ cash flow forecast revisions, suggesting that investors find these revisions informative. Collectively, our findings demonstrate that analysts’ cash flow forecasts are not simply naïve extensions of their own earnings forecasts, but that they reflect meaningful and useful accrual adjustments. These findings are relevant to researchers who examine analysts’ cash flow forecasts in a variety of settings, and to investors and practitioners who employ these forecasts for valuation purposes.
Number of Pages in PDF File: 47
Keywords: analysts, cash flow forecasts, accrual adjustments, forecast revisions
JEL Classification: M40, M41, M49working papers series
Date posted: June 1, 2012
© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo3 in 0.562 seconds