The Joint Effect of Management's Prior Forecast Accuracy and the Form of its Financial Forecasts on Investor Judgment

Abstract

We examine how investor reaction to management earnings forecasts is a joint function of the form of the forecast and management's perceived credibility. In a laboratory experiment involving 126 individual investors, we compare investors' earnings predictions and their confidence therein after receiving point and closed range forecasts issued by managements whose previous forecasting accuracy is known to be either high or low. We used point and range forecasts, because they differ in the degree to which they communicate management's uncertainty about the future. We use management's prior forecasting accuracy as a measure of management's credibility, because prior research has documented the importance of this factor when considering the usefulness of management's voluntary forecasts.

Our results show that, as expected, investors' earnings predictions are responsive to management's forecasts. However, as we hypothesized, forecast form did not influence investors' earnings estimates. In contrast, investors' confidence in their earnings predictions was influenced by the form of management's forecasts, but this effect emerged only when management was previously accurate in their forecasting. A similar interactive pattern was found in the dispersion of investors' predictions about the company's future earnings. Finally, consistent with the hypothesis that confidence is an important determinant of investor behavior, we find that investors' judgments of future stock price appreciation are a positive function of both unexpected earnings and the change in their confidence.

Our study extends the literature on management forecasts by empirically testing the joint influence of management's credibility (i.e., forecasting accuracy) and forecast form. The prior literature has argued that both factors should be important, but has not delineated whether or how these two factors might interact. We present a theoretical framework that indicates when both factors should influence investor judgment.

JEL Classification: M41, G12, C91, D82