Prior research in accounting indicates the importance placed by market participants on two features associated with earnings the trend in a firm’s earnings over time, and whether the firms’ earnings meet or beat analysts’ forecasts. For example, Barth, Elliott, and Finn (1999) find that when earnings show consistent increases over time (i.e., exhibit a positive trend), firms are rewarded with a higher price-earnings multiple. Bartov, Givoly, and Hayn (2002) reveal a market valuation premium placed on firms that meet or beat analysts’ earnings forecasts, and Kasznik and McNichols (2002) show that beating these forecasts consistently is particularly valuable. Most of the existing research has separately examined either earnings’ trends or meet/beat performance, with less focus on examining how the market reacts to both factors in combination.
The purpose of our study is to investigate how investors’ react to the combination of earnings trends and performance relative to analysts’ forecasts over multiple periods. We show this reaction is not static, but rather depends on the intertemporal consistency of those two earnings features.
Understanding how investors react to both factors over multiple time periods is important because market participants routinely evaluate firms in light of multi-period data about a firm’s earnings trend and their performance relative to analysts’ forecasts (Tan, Libby and Hunton, 2007). While research on the reaction to earnings trends takes a multi-period approach (e.g., Barth, et al., 1999), most research investigating the market reaction to missing or meeting analyst forecasts studies the effect of this performance for the current period only (e.g., Bartov, et al., 2002).
Thus, even though market participants have information on both trend and meet/beat performance over time, prior research has not fully capitalized on this institutional fact. Further, since the two features may not be independent, conclusions drawn from either set of studies may need to be modified when both features are considered in combination. In short, it is unclear how investors incorporate multi-period information about earnings trend and performance relative to analyst’s forecasts into their judgments about a firm.
To address the question of the how investors use earnings trends and multi-year benchmark information, we undertake the following in this paper: First, we draw on psychology theory to identify possible judgment strategies that investors could use when evaluating both earnings features over multiple periods. Judgment strategy refers to the manner in which individuals combine information about a firm’s earnings trend and benchmark performance to form judgments about firm valuation, desirability as an investment, and so on (Ashton, 1982). Specifically, we generate three possibilities: a dominance strategy wherein only one of the earnings features influences the judgments, an additive strategy wherein both features are used but have independent effects on judgment, and an interactive strategy wherein both features are used but interact in their effect on judgments.
Second, we conduct four experiments to investigate the conditions under which these strategies are used. Multiple experiments are necessary to test our hypothesis that how investors react to these two earnings features depends on their consistency over time. Consistency is important because we argue that it affects the informativeness of the earnings feature (Tucker and Zarowin, 2006).
In each experiment, participants are presented with multi-period information about two attributes of earnings specifically, they are told earnings per share (i.e., trend information) and whether the companies beat or missed analysts’ earnings forecasts (hereafter, benchmark performance). They are provided this information for a five-year time period and for four firms. These firms are identical except for the earnings trend and benchmark performance documented for each. Participants evaluate each of the four firms along several dimensions that pertain to market-related measures, including value, desirability as an investment, and future prospects. These measures are relevant because they map into the those studied in archival studies, including market returns (based on investor valuations and desire to invest) and their relation to future earnings (Bartov, et al., 2002; Kasznik and McNichols 2002).
Download
PDF Ebook Earnings Trends and Beating Analysts’ Forecasts: Are Both Considered Together Over Time?
