Academic journal article Financial Management

New Evidence on Serial Correlation in Analyst Forecast Errors

Academic journal article Financial Management

New Evidence on Serial Correlation in Analyst Forecast Errors

Article excerpt

We reexamine the serial correlation of forecast errors using a method that allows analysts to react differently to good and bad news. Our method also controls for the influence of a normal non-zero, firm-specific component of forecast error. Our results indicate that forecast errors exhibit positive serial correlation when there is bad news in the prior forecast error, negative serial correlation when there is good news in the prior forecast error, and no serial correlation when there is no news in the prior forecast error. These findings are consistent with analysts having optimistic reactions to new information.

A traditional analysis based on the rational behavior of experts in the market for information predicts that financial analysts' forecasts of earnings incorporate new information immediately, accurately, and without bias. In the traditional view, financial analysts are rational experts who forecast earnings, evaluate risk, and identify mispriced securities, and provide statistically optimal forecasts.

Several recent studies show that analysts' earnings forecasts are biased, [1] that analysts either underreact or overreact to new information, [2] and that forecast errors exhibit positive serial correlation. In particular, Mendenhall (1991), Abarbanell and Bernard (1992), Ali, Klein and Rosenfeld (1992), and Brown, Han, Keon, and Quinn (1996) find positive serial correlation in analysts' errors. [3] The findings of these studies have been interpreted as evidence that analysts fail to incorporate new information into their earnings forecasts immediately and accurately, i.e., they systematically underreact to new information. These studies suggest that serial correlation is inconsistent with rational forecasts and, perhaps, with an efficient market for expert information. Such forecast inefficiency could have important implications for the efficiency of securities markets if these markets believe that analysts' forecasts are both rational and statistically optimal.

The evidence of serial correlation consists of pooled cross-sectional, time-series regressions of the forecast error from one period on the forecast error from a prior period. This approach has two shortcomings. First, it does not distinguish between informative and normal components of prior forecast error. This is important, because prior research suggests that analysts' errors may contain a normal nonzero component that differs across firms and across time. To the extent that the firm-specific component of the forecast error persists across time, it provides no new information to analysts. Furthermore, in the presence of cross-sectional differences in the level of this component, regressing the current forecast error on a prior error across all firms will result in positive serial correlation. The intercept in such a regression would indicate the level of forecast error common to all firms, and the slope would include any adjustment to this normal error that was applicable for individual firms or groups o f firms. Consequently, serial correlation in such errors does not necessarily reflect analysts' effectiveness in incorporating new information into their forecasts.

The second shortcoming of the approach used in the prior studies is that it does not allow for the possibility that analysts' reactions are dependent on the nature of the information they receive (i.e., good or bad news). This is important because recent studies suggest that analysts have incentives to be optimistic. [4 ] The sell-side analysts analyzed in forecast error studies are employed by brokerage and investment banking firms. Therefore, they face economic incentives to promote the purchase of stocks, rather than to produce statistically efficient forecasts. In addition, analysts derive part of their expertise from their access to top executives of the firms they follow. This access might be diminished if they did not present the firm in a favorable light. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.