Mohan R. Lavi on a study about earnings forecast and stock price
Mohan R. Lavi
OVER the last decade or so, it has become customary for companies to issue "earnings forecast" which is primarily intended to inform investors, analysts and the bourses as to what the company expects would happen in the future. Though not statutorily bound, companies resorted to this as a matter of good governance though the language of the forecast statement is rife with assumptions.
What commenced as a trend abroad has caught on in India too with every company affirming, altering and, in a few instances, negating the first earnings forecasts. Naturally, questions began to surface as to the credibility of such forecasts in an unpredictable world. The initial reaction was that management forecasts were credible because one could compare it with the audited results as a reality check to judge the credibility.
Jonathan L. Rogers of the University of Chicago and Philip C. Stocken of Dartmouth College did some research on this, the results of which were carried in a recent issue of The Accounting Review. They proceeded on the assumption that managers have incentives to bias their earnings forecast. They are, however, constrained in this because the subsequent earnings report can affirm or deny their forecast. Their research focused on whether managers bias their forecasts based on the market's ability to assess the truthfulness of the management's earnings forecast. The paper also evaluated whether the market's response to management forecasts is consistent with investors identifying predictable forecast bias.
A regression model was developed that examined how forecast error is affected by the managers' incentives coupled with a proxy for the market's ability to detect manager misrepresentation. Incentives induced by the litigation environment, insider transactions, firm financial condition and industry concentration were built into the model. The primary finding was that the willingness of managers to misinterpret information in response to their incentives varied with the market's ability to detect their misinterpretation. Specific results were:
The research also looked into the stock market's response to the predicted errors in forecasts. For good news forecasts, the market responded less positively to forecasts with greater predicted optimism and more positively to forecasts with greater predicted pessimism.
For bad news forecasts, weak evidence was found that the market varies its response to the predicted bias in forecasts. The conclusion was that the market predicts and filters bias from management forecasts but the efficiency of the market's response varies with the forecast news.
Legally, the safe-harbour provision that all forecasts contain was assumed to be protection enough. However, there was a widespread feeling that this provision gave the firm a " licence to lie" as far as forecasts are concerned. The Private Securities Litigation Act was enacted in the US in 1995 to offer protection from litigation in forecasts.
This research proves that there is a correlation between the earnings forecast and the stock prices at the time the quarterly earnings are announced. It is also clear that the market would take a drop in performance due to extraneous reasons into account when they react to earnings news.
Managements can ill-afford bias in their forecasts or attempt to adjust erroneous forecasts in subsequent quarters. " Under-promise, over-deliver" could be a good policy for them to follow.
(The author is a Hyderabad-based chartered accountant.)