Earl K. Stice

What Do Stock Splits Really Signal

We observe significant post-split excess returns of 7.93 percent in the first year and 12.15 percent in the first three years for a sample of 1,275 two-for-one stock splits. These excess returns follow an announcement return of 3.38 percent, indicating that the market underreacts to split announcements. The evidence suggests that splits realign prices to a lower trading range, but managers self-select by conditioning the decision to split on expected future performance. Presplit runup and post-split excess returns are inversely related, indicating that our results are not caused by momentum.

Analyst forecasting ability and the stock price reaction to forecast revisions

We empirically identifysuperior analysts using their past forecasting track record fora specific firms earnings and demonstrate that subsequent forecastannouncements by these superior analysts have a greater impacton security prices than do the forecasts of other analysts. Wefind that, in our sample, the price effects of this firm-specificforecasting ability do not spill over to other firms followedby the same analyst. We also demonstrate that an analysts forecastingability with respect to the earnings of a certain firm is relativelymore important in the period immediately preceding an earningsannouncement by that firm.

The Market Reaction to the Choice of Accounting Method for Stock Splits and Large Stock Dividends

Prior research has used inaccurate classification rules to distinguish between stock splits and stock dividends. The CRSP classification of two-for-one stock distributions agrees with the actual accounting treatment only 23% of the time. In addition, the accounting treatment impacts the announcement period reaction—two-for-one distributions accounted for as stock dividends are associated with five-day announcement period returns of 2.70%, significantly greater that the 0.93% announcement returns for distributions accounted for as stock splits. Announcement returns are positively related to earnings growth in the two years following the distribution for stock dividend firms but not for stock split firms. The accounting choice appears to be used to confirm managements private information about future earnings revealed at the time of the distribution announcement.

Creating a Bigger Bath Using the Deferred Tax Valuation Allowance

The provisions of SFAS No. 109 allow US companies to make an earnings big bath even bigger through the establishment of a deferred tax valuation allowance. At the time a firm recognizes a non-cash charge, it also recognizes a deferred tax asset to represent the future tax benefits of the charge. Recognition of the deferred tax asset partially mitigates the negative earnings impact of the special charge. However, if the firm does not expect to have sufficient future taxable income to utilize the future tax benefits of the charge, SFAS No. 109 requires the firm to establish a deferred tax valuation allowance, effectively eliminating the recognized deferred tax asset. Thus, the establishment of the valuation allowance amplifies the negative earnings impact of the non-cash charge. We use a valuation allowance prediction model to identify firms that create a larger-than-expected valuation allowance; these firms may be creating a large valuation allowance as a reserve to be used to manage earnings in a subsequent period. We find that the vast majority of these larger-than-expected valuation allowances apparently reflect informed management pessimism about the future in that these firms actually do have poorer operating performance in subsequent periods. We do not find any evidence that subsequent reversals of valuation allowances are used to turn a loss into a profit. However, we do find a very small number of firms that appear to have used a valuation allowance reversal to meet or beat the mean analyst forecast. Copyright (c) 2008 The Authors Journal compilation (c) 2008 Blackwell Publishing Ltd.

The Power of Numbers: Base-Ten Threshold Effects in Reported Revenue

We provide evidence that managers have a revealed preference for reporting total revenue numbers just above base-ten thresholds (i.e., “round” numbers) of the form N × 10K. Examples are

Accounting rules and the signaling properties of 20 percent stock dividends

10 million (1 × 107) and

Motivation on day one: The use of Enron to capture student interest

4 billion (4 × 109). Our finding is consistent with a literature in psychology demonstrating that humans are susceptible to a cognitive bias associated with base-ten reference points. However, we also document several rational explanations for this revenue management behavior on the part of managers. First, analyst revenue forecasts also exhibit this regularity, especially in early forecasts when greater uncertainty can potentially induce analysts to rely to a greater extent on heuristics, suggesting that managers may be managing reported revenue numbers to meet externally-determined base-ten-influenced benchmarks. In addition, the effect that we document is stronger for firms that face greater pressure to report high revenue growth, while firms that exceed base-ten revenue thresholds for the first time benefit from increased press coverage. Finally, we show that the revenue growth needed to stretch for a base-ten threshold is not sustainable; firms that just exceed base-ten thresholds have lower subsequent revenue growth. Given that managers engage in extra, and, on average, unsustainable efforts to increase revenues to reach base-ten thresholds, our results suggest that revenue manipulation is even more pervasive than previously documented and that lenders, investors, auditors, and regulators should apply an extra degree of skepticism when a reported revenue number just exceeds a base-ten threshold.