Signaling Long-Term Information Using Short-Term Forecasts

A common concern with disclosure is revealing proprietary information. Many studies examine this proprietary cost hypothesis––that concerns about revealing proprietary information can prevent full disclosure––within the context of short-term earnings forecasts. However, these forecasts primarily accelerate the release of financial information by only a few months. Since their informational value is relatively short-lived, researchers also contend that the decision to provide such a forecast does not provide actionable information to competitors and, as a consequence, is unlikely to entail proprietary costs.

In this study, we shed light on this debate by showing that the decision to issue a short-term earnings forecast can, in fact, signal managers’ private information about long-term firm performance. Using a dynamic voluntary disclosure model, we show that the decision to disclose a short-term earnings forecast is informative of long-term earnings even if earnings are intertemporally independent. Consistent with this prediction, we find that these disclosures predict firm performance for up to three years, even after controlling for current financial performance.

This predictive power does not arise from earnings persistence but rather from the strategic nature of disclosure decisions. Simply put, managers with favorable long-term expectations are more inclined to issue short-term forecasts, while those with weaker long-term prospects are more likely to withhold guidance. The key mechanism behind this separation is that investors infer from issuing a short-term forecast that the manager has relevant private information. Given this investor expectation, failing to issue a forecast in the future may raise investor suspicion, leading to a decline in the firm’s stock price. Therefore, disclosing today incurs a greater cost of nondisclosure in the future. This cost is particularly high for firms with poor long-term prospects, as revealing weak future performance will further depress the future stock price. Anticipating this effect, firms with poor long-term prospects are less likely to disclose than those with better long-term prospects. Therefore, the decision to disclose short-term performance signal superior future firm performance even when the content of the disclosure is uninformative about the future.

Next, we examine economic conditions that impact the extent to which earnings-forecast decisions signal long-term performance information. First, we consider the role of current period performance (i.e., the forthcoming earnings number). Our model predicts that the relation between earnings-forecast decisions and long-term performance strengthens when current performance is worse. Disclosing bad current performance is costly in that it hurts firm value, other things equal, and increases the manager’s cost of nondisclosure in the future, as discussed above. The manager, aware of this, only discloses bad current performance when he expects positive long-term performance will offset these costs. Consistent with this prediction, we find that the relation between earnings-forecast decisions and long-term performance strengthens when the firm reports a loss or low earnings surprise (our proxies for poor current period performance).

Second, we consider managers’ horizons. Consistent with intuition, our model predicts that an earnings-forecast decision is less informative about long-term performance when the manager has a shorter horizon. In the extreme, if the manager only cares about the current period price (i.e., is myopic as defined in our model), the decision to disclose an earnings forecast will be driven exclusively by concerns about the current stock price and will not contain information about long-term performance. To test this prediction, we use the fraction of outstanding option awards at the end of the year and the amount of equity issuance of the next year to measure the manager’s horizon. We expect managers to care more about the future stock price (which corresponds to a longer horizon in our model) when they have more stock options at the end of the year and will issue more equity in the following year. Consistent with our prediction, we find that the relation between earnings-forecast decisions and long-term performance strengthens when managers have longer horizons.

Third, we consider the cost of revealing long-term performance due to competitive entry. We model the cost of disclosure as a function of the information it reveals about long-term performance. Specifically, in our model, an entrant decides about the investment in the industry, and the entrant’s investment reduces the incumbent’s cash flows. Our model predicts that, when the cost of revealing information about long-term performance is low, disclosing an earnings forecast provides more information about long-term performance. To test this prediction, we use the percentage change in Chinese exports in a given industry year to non-US developed countries to measure the cost (to US firms) of revealing long-term information. We also use a measure of competitive entry developed by prior research as a second proxy for entry threat. Consistent with our prediction, we find that the relation between earnings-forecast decisions and long-term performance weakens in industry years with higher threat of entry.

These findings challenge the notion that short-term earnings forecasts convey only short-lived information. Instead, they demonstrate that the decision to issue a forecast can, in many situations, carry valuable signals beyond its numerical content, shaping investor expectations and market perceptions. The revelation of long-term performance likely gives competitors sufficient lead time to erode the competitive advantage of the disclosing firm. Thus, we caution against the notion that a short-term earnings forecast does not entail proprietary costs.

Disallowing short-term earnings forecasts, as often discussed, comes at a cost, as it removes an important mechanism through which firms can signal their long-term prospects. This restriction is likely to have a disproportionate impact on smaller firms, which have fewer alternative channels to communicate with investors. Unlike large, well-established firms that benefit from extensive analyst coverage and other forms of investor engagement, smaller firms rely more heavily on voluntary disclosures like earnings forecasts to establish credibility and attract capital.

From a governance perspective, our study underscores the need for investors, regulators, and corporate boards to interpret short-term earnings forecasts not only as reflections of immediate earnings expectations but also as signals of long-term corporate health. Our findings suggest that the act of issuing a forecast itself—independent of the earnings number being forecasted—conveys meaningful information about managers’ expectations of future performance. Investors and analysts should therefore pay closer attention to firms’ disclosure decisions, particularly when managers deviate from established forecasting patterns, as such changes may signal shifts in long-term prospects.