Www.rotman.utoronto.ca



The Market’s Reaction to Changes in Performance RankingsJared JenningsOlin School of BusinessWashington Universityjaredjennings@wustl.eduHojun SeoOlin School of BusinessWashington Universityhojun.seo@wustl.eduMark T. Soliman Marshall School of BusinessUniversity of Southern Californiamsoliman@usc.eduJune 1, 2014Abstract:We examine how investors value changes in the relative performance ranking of the firm within the industry. We argue that the relative performance of the firm conveys information about the ability for the firm to compete in the industry. We find that investors positively (negatively) value increases (decrease) in the firm’s performance ranking, especially when the firm is initially estimated to be a low performer in the industry and when the change in the firm’s performance ranking has been stable in the past. Since the market seems to value increases in the firm’s performance ranking, we also predict and find evidence consistent with managers opportunistically excluding expenses from non-GAAP earnings to increase the firm’s performance ranking. We find that investors discount an increase in the firm’s performance ranking when it is more likely that managers are opportunistically excluding expenses from non-GAAP earnings to improve the relative performance of the firm.MARK – WE HAVE THOUGHT ABOUT CALLING THE CHANGE IN THE PERFORMANCE RANKING A RANKING SURPRISE, SINCE THAT IS WHAT IT REALLY IS. WE DID NOT KNOW IF WE WOULD GET INTO TO MUCH TROUBLE BY DOING THAT. 1.IntroductionA lucid benchmark that investors use to assess the firm’s performance is analyst expectations (Graham, Harvey, and Rajgopal, 2005). The financial press commonly compares the firm’s announced earnings to what analysts and other market participants expect. Not surprisingly, the prior literature has primarily focused on the benefits that a firm experiences when meeting or beating analyst expectations. For example, Kasznik and McNichols (2002) and Fischer, Jennings, and Soliman (2014) provide both empirical and theoretical evidence that firms meeting or beating expectations have higher returns and stock prices than firms that do not. However, the firm’s performance relative to analyst expectations is not the only benchmark that investors use to assess the firm’s performance. Market participants often compare the performance of the firm with that of other firms in the same industry. In this paper, we examine how investors value a change in the firm’s performance ranking within the industry. By doing so, we hope to better understand what information is communicated through earnings to the capital markets as well as the types of benchmarks investors use to evaluate the firm’s performance. In addition, we examine whether managers opportunistically manage earnings to affect the firm’s performance ranking and whether the market reacts less to a change in the firm’s ranking when managers are more likely to be opportunistically managing earnings.A firm’ performance can be decomposed into a firm-specific and non-firm-specific component (e.g., Waring, 1996; Rumelt 1991; McGahan and Porter 2002). The non-firm-specific component of the firm’s performance is affected by industry or market level shocks that affect all firms in the industry or market. The non-firm-specific component of the firm’s performance is likely to be primarily useful in evaluating the overall health of the industry or market. However, the component of performance that has a largest influence on the firm’s overall performance is the firm-specific component, which is the primary source for intra-industry heterogeneity (e.g., Rumelt, 1991; McGahan and Porter, 2002). Thus, the firm-specific component of firm performance is useful in evaluating the firm’s performance and competitive advantage within the industry and market (Nelson, 1991). If the competitive advantages of the firm are not readily substituted or imitated by rivals, then a change in the firm’s competitive advantage is expected to reflect an increase to expected future shareholder profits (e.g., Peteraf, 1993). We attempt to measure how investors value a meaningful change in the firm-specific component of performance by examining changes in the firm’s performance ranking within its industry. Sloan (1996) provides evidence that earnings typically exhibit high serial correlation year over year, suggesting that innovations to earnings are likely to persist. Therefore, we argue that the change in the firm’s performance ranking within the industry is likely to be a significant and potentially persistent change to the expected future performance and competiveness of the firm within the industry. Therefore, we anticipate that a change in the performance ranking of the firm within the industry is a significant information event, which ultimately conveys information to investors about the firm’s ability to generate profits for its shareholders.Using a sample of 142,581 firm/quarter observations from 1997 to 2012, we examine whether investors positively value an improvement in the firm’s performance ranking within the industry. To calculate changes in the firm’s performance ranking, we compare an initial ranking of the firm’s performance based on analyst expectations to the realized ranking of the firm’s performance based on realized earnings., Consistent with expectations, we find that the change in the firm’s performance ranking is positively associated with buy-and-hold abnormal returns during the three-day window surrounding the earnings announcement, after controlling for the firm’s earnings surprise and other firm characteristics. In fact, the positive market reaction to an increase in the firm’s performance ranking within the industry is approximately 55% of the market’s reaction to the firm’s earnings surprise. This evidence suggests that investors use both the earnings surprise as well as the relative performance of the firm within the industry to assess the firm’s ability to generate profits for shareholders. Next, we examine whether the market’s reaction to the change in the firm’s performance ranking varies based on the initial ranking of the firm’s performance in the industry. We anticipate that investors react more (less) to changes in the firm’s performance ranking when the firm is initially evaluated as a lower (higher) performer in the industry. If lower performers are unable to produce sufficient profits or to obtain the necessary capital needed to continue operating within the industry, lower performers are more likely to enter into bankruptcy or exit the market as more efficient firms enter the market. As the likelihood of bankruptcy or exiting the market increases, investors may be expecting the low performing firm’s profits to erode in the relative short run and/or may be shortening the time-series of expected future cash flows, causing the stock price to decrease. Therefore, the expected future cash flows of lower performers are likely to be more sensitive to changes in the firm’s relative performance in the industry. Consistent with expectations, we find that the market reaction to the change in the firm’s performance ranking is approximately 103% (18%) of the market’s reaction to a similar increase in the earnings surprise when the firm is initially evaluated as a low (high) performer.We also examine whether the past stability of the firm’s performance ranking changes affects investors’ market reaction to a change in the performance ranking of the firm. As previously mentioned, we argue that a change in the firm’s performance within the industry is reflective of a change to the expected future performance and competiveness of the firm within the industry. If the firm-specific component of performance has been stable (volatile) in the past, the change in the firm’s performance ranking in the current period is more (less) likely to provide an informative signal about the change in the future expected performance or competitiveness of the firm in the industry. Consistent with our prediction, we find that investors’ reaction to a change in the performance ranking of the firm is greatest (lowest) when the firm’s past ranking changes have been stable (volatile) over the preceding three years. The market reaction to the change in the firm’s performance ranking is approximately 83% (32%) of the market’s reaction to the firm’s earnings surprise when the past ranking changes have been stable (volatile).In 1998, Arthur Levitt, former SEC commissioner, expressed concern that firms were using earnings management to meet or beat analyst expectations. Since then, several papers have examined how managers might influence analyst expectations or manipulate earnings to opportunistically exceed analyst expectations. The prior literature has provided evidence consistent with managers using accrual manipulation (e.g., Abarbanell and Lehavy, 2003; Burgstahler and Eames, 2006), expectations management (e.g., Matusmoto, 2002), real activities manipulation (e.g., Roychowdhury, 2006), and non-GAAP earnings manipulation (e.g., Doyle, Jennings, and Soliman, 2013) to opportunistically exceed analyst expectations. Therefore, if the firm’s performance ranking is an important measure used by investors to evaluate firm performance, we anticipate that firms also have the incentive to opportunistically manipulate their performance ranking within the industry. Since changes in the firm’s performance ranking can change each time a firm in the industry announces earnings (which may occur frequently during the quarter), managers are likely unable to utilize many of the previously documented earnings management tools due to potentially sudden changes in the firm’s performance ranking as other firms in the industry announces earnings. As a result, we focus on the opportunistic exclusion of expenses from non-GAAP earnings (e.g., Doyle et al., 2003; Bowen et al., 2005; Doyle et al., 2013) to examine whether managers are opportunistically manipulating the firm’s performance ranking. The opportunistic exclusion of expenses from non-GAAP earnings does not require journal entries, a change in the operations of the firm, or extensive justification for excluding expenses with the auditor. We find evidence consistent with managers excluding expenses from non-GAAP earnings to improve the firm’s performance ranking within the industry. Our results appear to be driven by the more opportunistic excluded expenses (i.e., other exclusions), as documented in Doyle et al. (2013). We also find evidence that investors’ positive reaction to the improvement in the firm’s performance ranking is significantly reduced when the exclusions are more likely to be opportunistic. Specifically, we find that the positive market reaction to an improvement in the firm’s performance ranking that is coupled with the exclusion of expenses is muted when the firm is initially evaluated as a low performer, which is when investors react more strongly to changes in the firm’s performance ranking. In each of our tests described above, we control for firm size, firm growth, the magnitude of the earnings surprise, accruals, the firm’s initial ranking within the industry, and the volatility of the firm’s ranking changes within the industry over the past three years. In addition, we include industry-quarter fixed effects to capture unobservable time-varying industry characteristics (e.g., industry-level entry barriers or product market competition). We also cluster standard errors by firm and time (Peterson, 2009; Gow et al., 2013). In additional robustness tests, we explicitly control for changes in the consensus analyst recommendation and find that all of our aforementioned results are qualitatively and quantitatively similar. This is an important robustness test because prior studies find that changes in analyst recommendations may be associated with changes in the firm’s relative performance within the industry (e.g., Stickel 1985), and thus, the abnormal returns surrounding the earnings announcement could possibly be attributed to the changes in analyst recommendations rather than the changes in the firm’s performance ranking. By explicitly controlling for change in the consensus analyst recommendation, we rule out this possibility.We believe that this paper contributes to the literature in three key ways. First, we document another benchmark that investors use to evaluate the firms’ performance – the relative ranking of the firm’s performance within the industry. Put differently, our study suggests that investors evaluate the firm’s performance based on the entire distribution of earnings in a given industry to shed additional light on the competitive position of the firm within the industry. The prior literature, however, has primarily focused on the costs and benefits of meeting or beating analyst expectations (e.g., Degeorge et al., 1999; Matsumoto, 2002; Skinner and Sloan, 2002; Fischer et al, 2014), increasing earnings from the last period (e.g., Burgstahler and Dichev, 1997; Degeorge et al., 1999), or reporting earnings greater than zero (e.g., Burgstahler and Dichev, 1997). We extend the prior literature by showing that there are other important benchmarks that investors use to evaluate the firm’s performance. Earnings appear to convey useful information to investors about unexpected earnings innovations as well as changes in the firm’s competitive position within the industry.Second, we find that the opportunistic use of positive exclusions is not limited to meeting or beating analyst expectations but also manipulating the relative ranking of the firm within the industry. The vast majority of the extant literature seems to focus on management manipulating earnings to meet or beat the analysts’ consensus forecast (Doyle et al., 2013), increase earnings from the prior period (Burgstahler and Dichev, 1997; Degeorge et al., 1999), or avoid negative earnings (Burgstahler and Dichev, 1997). The prior literature pays little attention to firms that are well above or below these specific benchmarks. We provide evidence that managers have incentives to manipulate earnings even though they may be comfortably below or above these specific benchmarks.Third, our study extends the concept of relative performance evaluation (RPE) to stock valuation by providing initial evidence of RPE in capital markets. Both theoretical and empirical RPE has been mainly studied in the contracting literature. For example, RPE theory predicts that a principal evaluates an agent based on idiosyncratic performance after filtering out external shocks to improve contracting efficiency (Holmstrom, 1982). Milgrom and Roberts (1992) argue that information about the relative and ordinal performance is the most useful information in a tournament setting in an organization. Consistent with this, the prior empirical RPE studies analyze firms’ proxy statements and finds that a majority of firms compensate CEOs based on the changes in the firm’s performance ranking (e.g., Murphy, 1999; Carter et al., 2009; Bannister et al., 2011; De Angelis and Grinstein, 2014). In this paper, we document that RPE is not restricted to the contracting setting by showing that outside investors also evaluate firm performance based on the change in the firm’s performance ranking and adjust stock prices accordingly. In the next section, we develop our hypotheses. In Section 3, we describe our empirical tests. We discuss our results in Section 4. We discuss the results from additional robustness tests in Section 5. We conclude our study in Section 6. 2. Hypothesis DevelopmentAnalyst expectations are a widely used benchmark used to assess the performance of the firm. The financial media routinely cites analyst expectations when reporting the firm’s performance as a relevant benchmark in comparing expected performance to actual performance. When the actual performance of the firm is higher than expectations, firms typically experience positive abnormal returns. Kasznik and McNichols (2002) provide empirical evidence that firms that meet or beat analyst expectations experience a positive return premium on their stock. Fischer et al. (2014) provide theoretical and empirical evidence that a rational pricing bubble forms as the number of consecutive quarters that meet or beat analyst expectations increases. Therefore, meeting or beating expectations appears to positively affect the stock price of the firm, increasing management’s incentives to meet or beat analyst expectations. Consistent with this, in a survey of firm executives, Graham, Harvey, and Rajgopal (2005) find that 74% of firm executives believe that the analysts’ expectation is an important benchmark when reporting earnings. However, meeting or beating analyst expectations is likely not the only benchmark that market participants use to assess the performance of the firm. Analysts and journalists commonly evaluate the performance of the firm relative to other firms that are in the same industry (e.g., Boni and Womack, 2006; Kadan et al., 2012; Calia, 2014; Roger, 2013; Orlick, 2012). Analysts tend to incorporate firm rankings into recommendations; however, other firm, industry, and market factors also heavily influence these recommendations. The media also compares the performance of a firm to the performance of other firms in the industry. Despite the above, there has been little empirical evidence on how investors react to changes in the relative performance ranking of the firm. To the best of our knowledge, Graham et al. (2005) do not ask corporate executives as to whether the firm’s performance ranking within the industry is an important benchmark to firm executives. The absence of this question could be due to the focus of the extant accounting and finance literature on meeting or beating analyst expectations, avoiding losses, and reporting positive increases in earnings. Therefore, we do not currently have much evidence on the relative importance of the change in the firm’s performance ranking in the industry.Rumelt (1991) decomposes overall firm performance into three components: 1) the overall business cycle component, 2) the industry component, and 3) the business-specific component. Rumelt (1991) documents that the firm-specific component of performance is the most significant driver of the firm’s overall performance. He argues that the firm-specific component of performance is mainly determined by “the presence of business-specific skills, resources, reputations, learning, patents, and other intangible contributions to stable differences among business-unit returns.” McGahan and Porter (2002) analyze the variance of accounting profitability and also find consistent evidence that the firm-specific component of performance has a largest influence on overall firm performance. Waring (1996) finds that the persistence of the firm-specific component of performance substantially varies across different industries and documents that variables such as the percentage of professional workers, the degree of unionization, the percentage of consumer purchases, the number of firms within the industry, economies of scale, and R&D intensity have strong influences on the persistence of the firm-specific component of performance. Overall, prior research argues that the firm-specific component of performance is a significant predictor of the firm’s overall profitability.Based on the above discussions, we anticipate that investors primarily evaluate the firm using the firm-specific component of performance, which allows investors to better understand the relative competitive advantages held by the firm in the market and industry. If rivals cannot easily imitate the competitive advantages held by the firm, then these competitive advantages are expected to be sustained, allowing the firm to generate greater returns for shareholders (e.g., Peteraf, 1993; Barney 1986; Barney 1991). In addition, Sloan (1996) provides evidence that firm performance is likely to exhibit high serial correlation year over year, suggesting that innovations to a firm’s performance is likely to persist. Therefore, the change in the firm-specific component of performance that moves the firm’s competitive position within the industry ultimately conveys information to investors about the firm’s ability to continue as a going concern and the firm’s ability to generate profits for its shareholders. We argue that investors can obtain information regarding the change in the firm-specific component of performance that moves the competitive position of the firm within the industry by observing the change in the firm’s performance ranking within the industry. Managers pursue various strategic activities to establish competitive advantages in order to differentiate themselves from their rivals, leading to the greater performance. However, the strategic activities may not be directly observable and may not be properly evaluated by investors. Litov et al. (2012) argue that market participants face significant information problems resulting from managerial proprietary insights about the future value of the firm’s unique strategy. Litov et al. (2012) also argue that “if managers do not possess proprietary insights, and instead all opportunities are transparently obvious to the market, replication of strategies will occur and arbitrageurs will buy the resources required by the managers and sell them to the firms at prices near their value added in the manager’s strategy, thereby dissipating any value to be created by the strategy (Barney, 1986).” In addition, significant uncertainty exists as to whether the particular strategy can establish competitive advantages that generate sustainable future profits. Therefore, market participants are less likely to fully understand the implications of various strategic activities until earnings are released, allowing investors to evaluate the firm relative to other firms in the industry. Therefore, we anticipate that the firm’s earnings provide information about the competitive advantages held by the firm in the industry, which are revealed with realized earnings. We specifically predict that investors positively (negatively) value an increase (decrease) in the performance ranking of the firm within the industry. We have stated our hypothesis in alternative form below. H1 –Investors positively (negatively) value an increase (decrease) in the firm’s relative performance ranking within the industry.Our first hypothesis predicts that investors react to changes in the firm’s performance ranking because it conveys information about changes in the firm’s competitiveness in the industry. If so, we further anticipate that low performers benefit more than high performers when improving the firm’s performance ranking within the industry, resulting in a greater market reaction. Prior research documents that competition drives inefficient firms to exit the market and more efficient firms to expand their market shares at the expense of inefficient firms (e.g., Schmidt, 1997; Nickell, 1996; Jayaratne and Strahan, 1998; Stiroh and Strahan, 2003). That is, if lower performing firms are unable to produce sufficient profits to either fund or obtain the necessary capital needed to stay competitive in the industry, investors may expect the firm’s profits to erode over time and may be truncate the times series of expected future cash flows due to the increased likelihood of these firms exiting the market. As a result, if low performers improve their performance ranking within the industry, investors are more likely to interpret this improvement as a significant decrease in the likelihood of exiting the market, thereby reevaluating the erosion and length of the expected future cash flows and adjusting the stock price accordingly. We do not expect high performing firms to suffer from this problem because high performing firms’ time-series of future expected cash flows are less likely to be truncated or experience as notable erosion due to the lower likelihood of exiting the market. Thus, we anticipate that investors react more (less) to changes in the firm’s performance ranking when the firm is initially evaluated as a lower (higher) performer in the industry. We state our hypothesis in alternative form below. H2 – Investors react more strongly to a change in the firm’s relative performance ranking within the industry when the firm is initially evaluated as a low performer within the industry. Depending on the firm and/or industry characteristics, the firm’s performance ranking may fluctuate significantly. If changes to the firm’s performance ranking within the industry are volatile, investors are less likely to view changes to the firm’s performance ranking as persistent or reflective of a significant change to the competitive position of the firm within the industry. Therefore, we anticipate that investors are more likely to interpret a change in the firm’s performance ranking in the current period as a temporary fluctuation when the past volatility of changes in the firm’s performance ranking is high. If the past volatility of changes in the firm’s performance ranking is low, we anticipate that the change in the firm’s performance ranking in the current period provides investors with a more informative signal of a change in the competitiveness of the firm in the industry, leading to a greater investor response. Therefore, we predict that investors react stronger (weaker) to a change in the firm’s performance ranking when the past volatility of changes in the performance ranking is lower (higher). Our hypothesis is stated in alternative form below. H3 – Investors react more (less) strongly to a change in the firm’s performance ranking within the industry when the past volatility of changes in the firm’s performance rankings is low (high). The prior literature suggests that managers use various methods to manipulate either analyst forecasts or earnings to meet or beat analyst expectations. For example, prior studies examine whether managers use accrual manipulation (e.g., Abarbanell and Lehavy, 2003; Burgstahler and Eames, 2006), expectations management (e.g., Matusmoto, 2002), real activities manipulation (e.g., Roychowdhury, 2006), and non-GAAP earnings manipulation (e.g., Doyle et al, 2013) to opportunistically exceed analyst expectations. In addition, Graham et al. (2005) survey corporate executives and document that corporate executives have strong incentives to manipulate earnings to meet or beat analyst expectations due to the pressure from capital markets. Therefore, if the relative performance of the firm within the industry is another important benchmark that investors use to evaluate firm performance, we anticipate that firms also have incentives to opportunistically manage earnings to improvement the firm’s performance ranking within the industry. Despite the wide range of methods that managers can utilize to manipulate the firm’s performance ranking, we anticipate that managers are likely to rely on the opportunistic exclusion of expenses from non-GAAP earnings when manipulating the firm’s performance ranking (e.g., Doyle et al. 2013). Manipulating the firm’s earnings to increase its performance ranking is different from manipulating earnings to meet or exceed analyst expectations. Analyst expectations are typically determined a couple weeks prior to the announcement of earnings. Managers are able to observe the analysts’ expectations and choose the method that is most appropriate to meet or beat those expectations. However, the change in the firm’s performance ranking is dynamic in that the performance ranking is determined by both the firm’s earnings as well as the peer firms’ earnings. Put differently, the firm’s performance ranking in the industry could change anytime a firm within the industry announces earnings or when analysts revise their expectations for firms within the industry. Therefore, manager’s ability to manipulate earnings through the management of real activities, accruals, and market expectations is substantially reduced. Managing earning through the manipulation of real activities likely requires a significant amount of planning and time, which generally happens prior to the fiscal period end. Analysts expectations could be managed downward prior to observing the earnings of other firms in the industry; however, the manager would not know how much he/she would have to manage earnings expectations downward since the other firms’ earnings have not been revealed. Therefore, the management of analyst expectations is likely less effective when managers are attempting to improve the firm’s relative performance ranking. Discretionary accruals require journal entries and require planning and justification to the auditor, reducing the likelihood of the managers using discretionary accruals to manipulate the performance ranking of the firm. However, redefining non-GAAP earnings does not require journal entries or extensive justification to the auditor. Therefore, we expect that redefining non-GAAP earnings is the most effective method for managers to increase the performance ranking of the firm. We specifically predict that managers exclude expenses from non-GAAP earnings to improve the firm’s performance ranking within the industry. We state our hypothesis in alternative form below. H4 –Managers exclude expenses from non-GAAP earnings to increase the firm’s relative performance ranking within the industry.Doyle et al. (2013) find that firms that are more likely to opportunistically exclude expenses from non-GAAP earnings have earnings surprises that are less informative to investors. Doyle et al. (2013) specifically document that the market reaction to the earnings surprise is discounted when investors suspect that mangers opportunistically use income-increasing exclusions to artificially meet or beat analyst expectation. Similarly, if managers opportunistically use exclusions to affect investors’ perceptions concerning the firm’s performance ranking, we expect investors to discount the change in the firm’s performance ranking when investors suspect that managers are opportunistically excluding expenses from non-GAAP earnings to manipulate the firm’s performance ranking. We have stated the related hypothesis in alternative form below. H5 –Investors’ reaction to the firm’s relative performance ranking changes in the industry is weaker when the firm’s relative performance ranking is coupled with the exclusion of expenses from non-GAAP earnings.3. Empirical Design3.1.Market Reactions to a change in the firm’s performance rankingIn Hypothesis 1, we predict that investors positively (negatively) react to an increase (decrease) in the firm’s relative performance ranking within the industry. In the main analyses, we use the Global Industry Classification Standard (GICS) codes to define the industry to which a firm belongs. Bhojraj et al. (2003) document that firms in the same GICS classifications have higher profitability and growth correlations than firms that share the same Standard Industrial Classification (SIC) codes, North American Industry Classification System (NAICS) codes, and Fama-French classification codes. They conclude that GICS is a better industry classification to identify industry peers that compete in the same product markets. Using GICS codes to define the industry, we measure the change in the firm’s performance ranking within the industry on the date of the firm’s earnings announcement. While we could examine the change in the firm’s performance ranking at various points during the fiscal period, we choose the earnings announcement date for two reasons. First, the earnings announcement is a significant firm event that reveals information about the firm, prompting investors to evaluate and revise their expectations about the firm’s performance ranking. Second, earnings are released during the earnings announcement, allowing us to examine the market’s response to the change in the firm’s performance ranking relative to the market’s response to the earnings surprise. Therefore, we can control for the firm’s earnings surprise in the regression analyses, allowing us to isolate the effect of the change in the firm’s performance ranking. We employ the following regression model to examine investors’ response to changes in the firm’s performance ranking. 3DayReti,t =α + β1?Rankingi,t + β2Surprisei,t + β3STD_?Rankingi,t + β4Initial Rankingi,t + β5Book-to-Marketi,t + β6 ln(Sizei,t) + β7SalesGrowthi,t + β8Accrualsi,t + Industry_Quarter Dummiesj,t + εi,t(1)The subscript i, j, and t represent the firm, the industry, and fiscal quarter, respectively. The dependent variable is the 3DayReti,t variable, which is equal to the three-day market-adjusted buy-and-hold abnormal returns centered on the earnings announcement for firm i in quarter t. The main independent variable of interest is the ?Rankingi,t variable, which is measured as the change in firm i’s performance ranking within the industry on the earnings announcement date in quarter t (i.e., Realized Rankingi,t) compared to firm i’s performance ranking within the industry two days prior to the earnings announcement date in quarter t (i.e., Initial Rankingi,t). To calculate the ?Rankingi,t variable we follow three simple steps. First, we calculate the initial ranking of firm i in quarter t two days prior to the earnings announcement for firm i in quarter t (i.e., Initial Rankingi,t). To calculate the initial ranking of the firm i in quarter t, we first calculate the consensus earnings per share (EPS) forecast as the median analyst forecast issued over the preceding 90 days. We then rank (ordinal) the consensus forecast for firm i with the realized or expected earnings for all other firms sharing the same GICS code (i.e., peer firms) on the same date. If peer firms have already announced earnings, we use realized earnings. If peer firms have not already announced earnings, we calculate the consensus forecast for each peer firm in the industry following the same procedure described above. Prior to calculating the initial ranking of the firm i in quarter t, we standardize the earnings performance figures for all firms in the industry by multiplying each EPS figure by the number of shares outstanding and dividing by the lagged market capitalization for each firm. Second, we calculate the realized ranking of firm i in quarter t at the earnings announcement date (i.e., Realized Rankingi,t) similarly to how we calculated the Initial Rankingi,t variable with the one exception. Instead of using the consensus forecast for firm i, we use the realized earnings of firm i that are announced at the earnings announcement date for quarter t. We then rank firm i’s realized earnings relative to peer firms’ earnings (i.e., realized or expected earnings) in the industry to calculate the Realized Rankingi,t variable. To finally finish the calculation of the ?Rankingi,t variable, we subtract the Realized Rankingi,t variable from the Initial Rankingi,t variable and divide by the number of firms in the industry. We anticipate finding a positive coefficient on the ?Rankingi,t variable, which is consistent with investors positively valuing an improvement to the firm’s performance ranking in the industry.We also include several control variables in the model that are likely to be simultaneously associated with performance ranking changes and the market’s reaction to the earnings announcement. The Surprisei,t variable is the earnings surprise for firm i in quarter t, which is equal to the IBES-reported Actual EPS (IBES item VALUE) figure less the median consensus analyst forecast divided by stock price at the end of quarter t-1. The STD_?Rankingi,t variable is the standard deviation of the ?Rankingi,t variable over the previous 12 quarters (we require a minimum of 8 quarter observations to calculate the variable). The Initial Rankingi,t variable is the performance ranking of firm i two days prior to the earnings announcement date for quarter t and is described above in detail. The Book-to-Marketi,t variable is calculated by dividing the book value of equity (Compustat data item seqq) by the market value of equity (Compustat data item prccq multiplied by Compustat data item cshoq) at the end of quarter t. The SalesGrowthi,t variable is equal to net sales (Compustat data item saleq) in quarter t divided by net sales in quarter t-4. The ln(Sizei,t) variable is equal to the natural logarithm of the market value of equity at the end of quarter t. Accrualsi,t is measured as firm i’s GAAP EPS (Compustat item epspxq) less cash flows from operations per share (Compustat item oancfy divided by Compustat item cshprq) in quarter t divided by the stock price (Compustat item prccq) at the end of quarter t-1. For all our tests, we include industry/calendar quarter dummies to control for unobservable time-varying industry characteristics such as industry-level entry barriers or the level of product market competition in the industry. We cluster the standard errors by calendar quarter and firm to correct for cross-sectional and serial-correlation in the standard errors (Petersen, 2009; Gow et al., 2013).In an additional test, all independent variables in equation (1) are decile-ranked to facilitate the comparison between the market’s reaction to the earnings surprise (Surprisei,t) and the market’s reaction to the change in the performance ranking (?Rankingi,t). By ranking the independent variables into deciles, we can compare the market’s reaction to a change in the performance ranking to that of the earnings surprise (Abarbanell and Bushee, 1998). We create decile-ranked variables by ranking the variable into deciles (i.e., 0 through 9) and dividing by 9. Therefore, the decile-ranked variables range from 0 to 1. All decile-ranked variables are preceded by a “D_” prefix. Since the D_?Rankingi,t (D_Surprisei,t) variable is decile-ranked, the coefficient on the D_?Rankingi,t (D_Surprisei,t) variable represents the market’s reaction to an increase from the 1st to 10th decile of the change in performance ranking (earnings surprise) deciles. Using the decile-ranked variables, we are now able to compare the difference between the market’s reaction to an increase in the firm’s performance ranking and the market’s reaction to an increase in the firm’s earnings surprise as each variable moves from its 1st to 10th decile. 3.2.Differential responses to the firm’s performance ranking changesWhen examining Hypothesis 2 and 3, we focus on the model specification in which the independent variables are decile-ranked to facilitate the comparison of the effect of earnings surprise and change in the performance ranking on returns. In Hypothesis 2, we predict that the market reaction to an improvement in the relative ranking of the firm within its industry is stronger (weaker) when the firm is initially evaluated as a low (high) performer in the industry. To test this hypothesis, we use the following regression model.3DayReti,t =α + β1D_?Rankingi,t + β2D_Surprisei,t + β3D_Initial Rankingi,t + β4D_Initial Rankingi,t× D_?Rankingi,t + β5D_Initial Rankingi,t× D_Surprisei,t + β6D_STD_?Rankingi,t + β7D_Initial Rankingi,t + β8D_Book-to-Marketi,t + β9D_ln(Sizei,t) + β10D_SalesGrowthi,t + β11D_Accrualsi,t + Industry_Quarter Dummiesj,t + εi,t(2)All variables are as previously defined. As a reminder, the “D_” prefix denotes a decile-ranked variable. We continue to expect a positive coefficient on the D_?Rankingi,t variable, which would suggest that the market positively values a change in the firm’s performance ranking when the initial ranking variable is in its 1st decile (i.e., the D_Initial Rankingi,t variable is equal to zero). To test H2, we interact the D_Initial Rankingi,t variable with the D_?Rankingi,t variable and expect a negative coefficient on this interaction, suggesting that the market reaction to an improvement in the relative performance ranking of the firm is weaker when the firm is initially evaluated as a high performer in the industry. We also interact the D_Initial Rankingi,t variable with the D_Surprisei,t variable to facilitate the comparison between changes in the firm’s performance ranking and the earnings surprise when the initial performance ranking of the firm is high and low. However, we do not make an explicit prediction on the interaction between the D_Initial Rankingi,t and D_Surprisei,t variables.Hypothesis 3 predicts that the market reaction to an improvement in the relative ranking of the firm is stronger (weaker) when the changes in the firm’s performance ranking have been stable (volatile) in the recent past, making the current quarter’s ranking change more (less) notable and informative to investors. To test this hypothesis, we use the following model.3DayReti,t =α + β1D_?Rankingi,t + β2D_Surprisei,t + β3D_STD_?Rankingi,t + β4D_STD_?Rankingi,t ×D_?Rankingi,t + β5D_STD_?Rankingi,t × D_Surprisei,t + β6D_STD_?Rankingi,t + β7D_Initial Rankingi,t + β8D_Book-to-Marketi,t + β9D_ln(Sizei,t) + β10D_SalesGrowthi,t + β11D_Accrualsi,t + Industry_Quarter Dummiesj,t + εi,t(3)All variables are as previously defined. Once again, we decile rank all independent variables to facilitate the comparison between the market’s reaction to a change in the firm’s performance ranking and the market’s reaction to the earnings surprise when the volatility in the past changes in the performance ranking varies. We continue to expect a positive coefficient on the D_?Rankingi,t variable, which provides support on how the market values an increase in the performance ranking when the volatility in past performance ranking changes is in the 1st decile (i.e., the D_STD_?Rankingi,t variable is equal to zero). The interaction between the D_STD_?Rankingi,t and the D_?Rankingi,t variables is our test of H3 and provides evidence on how the market reacts to a change in the performance ranking of the firm when the volatility of the firm’s past performance ranking changes moves from the 1st to 10th decile. We anticipate a negative coefficient on the interaction between the D_STD_?Rankingi,t and the D_?Rankingi,t variables. Similar to the preceding test, we also interact the D_STD_?Rankingi,t and the D_Surprisei,t variables to compare the market’s reaction to changes in the firm’s performance ranking to the market’s reaction to the earnings surprise. 3.3. Managers’ Use of Exclusions and Changes in Industry RankingsNow we turn to the Hypothesis 4, which predicts that managers opportunistically use non-GAAP exclusions to affect the relative performance ranking of the firm within the industry. To test this prediction, we use the following regression model, which uses the ?Rankingi,t variable as the dependent variable.?Rankingi,t =α + β1Pos Excl Usei,t + β2Book-to-Marketi,t + β3SalesGrowthi,t + β4ln(Size) + β5Profitablei,t + β6ROAi,t + β7MBEi,t + β8ln(NUMESTi,t) + β9STD_?Rankingi,,t + β10Initial Rankingi,t + Industry_Quarter Dummiesj,t + εi,t(4)The independent variable of interest is the Pos Excl Usei,t variable, which is an indicator variable equal to one if exclusions are positive; otherwise zero. Positive exclusions represent expenses that are excluded from non-GAAP earnings but included in GAAP earnings. Consistent with prior studies, we define exclusions as IBES-reported Actual EPS less GAAP EPS (e.g., Doyle et al. 2013). GAAP EPS is defined as earnings per share before extraordinary items and discontinued operations, using either basic (Compustat item epspxq) or diluted EPS (Compustat item epsfxq), depending on the IBES basic/diluted flag. If managers opportunistically exclude expenses from non-GAAP earnings to maintain or improve the performance ranking of the firm, we would expect to observe the existence of positive exclusions to be associated with an improvement to the change in the firm’s performance ranking (i.e., a positive coefficient on the Pos Excl Usei,t variable). If analysts understand and expect the expenses that managers exclude from non-GAAP earnings, analysts should also exclude these expenses from their forecasts. Therefore, if the magnitude or existence of expenses excluded from non-GAAP earnings are expected by analysts, then management’s use of positive exclusions should not mechanically result in an improvement to the performance ranking of the firm since the firm’s initial and realized ranking are prepared on the same basis (Doyle et al., 2013). However, since management can manipulate exclusions, managers have the opportunity to include recurring expenses that were not expected by analysts, potentially increasing the performance ranking of the firm. Similar to the extant literature (Doyle et al., 2013), we divide positive exclusions into expected and unexpected exclusions, which we proxy for using special items and other exclusions, respectively. Special items are defined as operating income per share (Compustat Item opepsq) less GAAP EPS before extraordinary items (Compustat item epspxq or epsfxq based on IBES basic/diluted flag). We then define other exclusions as total exclusions less special items, which capture the unexpected income-increasing exclusions. Even though special items are typically regarded as unusual or nonrecurring items, we anticipate that analysts can better anticipate and estimate the expenses that are classified as special items compared to other exclusions (Doyle et al., 2003; Doyle et al., 2013). Doyle et al. (2003) argue that other exclusions predict negative future operating cash flows, suggesting that other exclusions have recurring expense properties. This evidence is consistent with management strategically classifying exclusions as other exclusions.To test whether unexpected exclusions are the primary mechanism that managers use to opportunistically increase the relative performance ranking of the firm, we replace the Pos Excl Usei,t variable with the Pos Special Item Usei,t and Pos Other Excl Usei,t variables. The Pos Special Item Usei,t (Pos Other Excl Usei,t) variable is an indicator variable equal to one if special items (other exclusions) are positive; zero otherwise. If managers primarily use other exclusions to influence the firm’s performance ranking then we would expect to observe significantly positive coefficient on the Pos Other Excl Usei,t variable and an insignificant coefficient on the Pos Special Item Usei,t variable. However, it is possible that we find a positive coefficient on the Pos Special Item Usei,t variable if analysts are not able to perfectly anticipate and identify special items without any bias. Regardless of whether the coefficient on the Pos Special Item Usei,t variable is positive, we expect the coefficient on the Pos Special Item Usei,t variable to be significantly lower than the coefficient on the Pos Other Excl Usei,t variable.Since we expect the market to react more to changes in the relative performance ranking of the low performers (H2), we also anticipate that low performers in the industry are more likely than high performers to manipulate earnings with the intention to improve the firm’s relative performance ranking. Therefore, we divide the full sample into three subsamples based on initial ranking terciles and estimate equation (4) for each subsample. If the low performers in the industry are more likely to use income-increasing exclusions to improve their performance ranking, we expect to observe the most positive coefficient on the Pos Excl Usei,t and Pos Other Excl Usei,t variables when examining those firms that are in the 1st initial ranking tercile (i.e., low performers in the industry). We anticipate that the magnitude of the coefficients on the Pos Excl Usei,t and the Pos Other Excl Usei,t variables to decrease monotonically as we move from the low performers (the 1st Initial Rankingi,t tercile) to the high performers (the 3rd Initial Rankingi,t tercile), suggesting that firms are more likely to use income-increasing exclusions to opportunistically increase the relative performance ranking of the firm when the benefits are highest to the firm. Our independent variables in equation (4) are defined as follows: the Profitablei,t variable is an indicator variable equal to one if firm i's IBES-reported Actual EPS in quarter t is positive, zero otherwise; the ROAi,t variable controls for firm performance and is equal to firm i’s IBES-reported Actual EPS in quarter t divided by total assets per share at the end of quarter t (Compustat item atq divided by Compustat item cshoq); the MBEi,t variable is intended to control the effect of positive exclusions on the likelihood of meeting or beating analyst expectations (Doyle et al. 2013) and is an indicator variable equal to one if firm i’s earnings surprise in quarter t (i.e.,Surprisei,t) is non-negative, zero otherwise; the ln(NUMESTi,t) variable is the natural logarithm of the number of analysts following the firm i in quarter t.3.4. The effect of exclusions on the market reactions to the performance ranking changesIn H5, we predict that the market’s response to the change in the performance ranking will be discounted if the market can identify firms that are more likely to be manipulating earnings to improve the firm’s performance ranking. We estimate the below regression to test H5. 3DayReti,t =α + β1?Rankingi,t + β2Pos Excl Usei,t + β3?Rankingi,t×Pos Excl Usei,t + β4STD_?Rankingi,t + β5Initial Rankingi,t + β6Book-to-Marketi,t + β7 ln(Sizei,t) + β8SalesGrowthi,t + β9Accrualsi,t + Industry_Quarter Dummiesj,t + εi,t(5)The coefficient on the interaction between the ?Rankingi,t and Pos Excl Usei,t variables is the primary coefficient of interest. If the firm’s use of positive exclusions increases the likelihood that firms are opportunistically manipulating earnings, then a negative coefficient on the interaction between the ?Rankingi,t and Pos Excl Usei,t variables would suggest that investors are discounting changes in the relative performance ranking of the firm when the likelihood of manager manipulation is higher. We also substitute the Pos Excl Usei,t variable for the Pos Other Excl Usei,t and Pos Special Items Usei,t variables to see whether the unexpected exclusions are driving the results. When testing H4, we separate the sample into Initial Rankingi,t terciles to examine whether the association between the positive exclusions and the change in the relative performance ranking is strongest in the 1st tercile and weakest in the 3rd tercile. We anticipated this weakening from the 1st to 3rd Initial Rankingi,t tercile because of the expected heightened investor reaction to a change in the relative ranking of the firm when the firm is a low performer (H2). Therefore, we also anticipate that the market discounts the change in the relative ranking of the firm most when the likelihood of manipulation is the highest, which we expect is in the 1st Initial Rankingi,t tercile. 4. Data and Descriptive StatisticsData for our empirical tests were obtained from the intersection of I/B/E/S, COMPUSTAT, and CRSP. To calculate the median EPS consensus analyst forecasts, we use individual analysts’ quarterly EPS forecasts taken from the unadjusted I/B/E/S detail file. We also obtain the realized earnings and earnings announcement dates from I/B/E/S. We start collecting data in 1995 because individual analyst forecasts are relatively sparse prior to 1995 (Clement et al. 2011). Since one of our main control variables, STD_?Rankingi,t, requires at least past 2 years of analyst forecast data, our sample period ranges from 1997 to 2012. We retrieve quarterly financial statement data from COMPUSTAT and daily stock return data from CRSP. We require at least 10 firm-quarter observations in each industry in each quarter to calculate the performance ranking changes for each firm in the industry. We only keep firm/quarter observations with fiscal quarter ends of March, June, September, and December. We also require firm/quarter observations with sufficient data to calculate the independent and dependent variables in each regression. Our final sample consists of 142,581 firm-quarter observations ranging from 1997 to 2012. The number of observations in any particular test varies depending on the availability of data necessary for that particular test. All continuous variables are winsorized at 1% and 99%.Panel A of Table 1 presents descriptive statistics for the full sample. We note that the average (median) change in the firm’s performance ranking (?Rankingi,t) is 0.011 (0.000). The average (median) earnings surprise (Surprisei,t) is 0.000 (0.010). We note that the average and median changes in the ?Rankingi,t and Surprisei,t variables are reasonably close to zero. We expected an average and median value close to zero for the ?Rankingi,t variable given its construction. The zero values for the average and median values for the Surprisei,t variable suggests that analyst forecasts are relatively unbiased. Mean and the median values of other variables are similar to those reported in prior research. For instance, the mean of MBE is 0.667, suggesting that the majority of firms (66.7%) meet or beat their earnings expectations. The average (median) Book-to-Market ratio is 0.626 (0.493) and the average (median) sales growth is 1.148 (0.964), which are both similar values to those found in prior studies (e.g., Doyle et al. 2013).Table 2 reports the correlations among main variables in our study. The 3DayReti,t variable is positively correlated with the ?Rankingi,t and Surprisei,t variables (Spearman correlation 0.22 and 0.20 respectively), suggesting that investors positively respond to the improvement in the firm’s performance ranking in the industry as well as unexpected earnings. The ?Rankingi,t variable is also positively correlated with the Surprisei,t variable (Spearman correlation 0.62), the Profitablei,t variable (Spearman correlation 0.22), and the ROAi,t variable (Spearman correlation 0.16), suggesting that firms that have a greater surprise, are profitable, and have better performance are more likely to experience an increase in its performance ranking. This further highlights the need to control for various measures of the firm’s performance in our multivariate regressions analyses. 5.Empirical Results5.1.Results for H1: Market reactions to performance ranking changes In Hypothesis 1, we predict that stock returns surrounding earnings announcement dates are positively correlated with the firm’s performance ranking changes within the industry after controlling the effect of earnings surprise and several other control variables. Table 3 presents the results from equation (1) with the three-day buy-and-hold abnormal returns surrounding the earnings announcement date as the dependent variable and the firm’s performance ranking changes as the primary independent variable of interest. In column (1) and (2), we omit the control variables (except for the industry/calendar quarter fixed effects) from the analysis and include the ?Rankingi,t and Surprisei,t, variables in separate regressions. We find a positive and significant (1% level) coefficient on both the ?Rankingi,t variable as well as the Surprisei,t, variable in column (1) and (2), respectively. This evidence is consistent with investors positively (negatively) valuing an increase (decrease) in the firm’s performance ranking and an unexpected increase (decrease) in earnings. In column (3), we include both the ?Rankingi,t and Surprisei,t, variables together in the same regression with only industry/calendar quarter fixed effects. We find positive and significant coefficients on both variables, consistent with investors positively valuing an improvement in the firm’s relative performance ranking as well as unexpected earnings. This evidence also suggests that the change in the firm’s performance ranking conveys incremental information over the earnings surprise. It is also worth noting that the magnitude of coefficient on earnings surprise is reduced significantly when the ?Rankingi,t variable is included in the model, suggesting that a significant portion of relation between the earnings surprise and the market’s reaction is explained by a change in the firm’s relative performance ranking in the industry. In column (4) of Table 3, we continue to find that the coefficients on the ?Rankingi,t and Surprisei,t, variables are still significantly positive at the 1% level when including the control variables described in equation (1). To gauge the relative effect of a change in the relative performance ranking of the firm and the earnings surprise, we report the regression results using decile-ranked independent variables in column (5). The coefficient on the decile-ranked ?Rankingi,t variable is equal to 0.028 and continues to be significant at the 1% level. Similarly, the coefficient on the Surprisei,t, variable is significant at the 1% level and is equal to 0.051. The coefficient on each of the decile-ranked independent variables can be interpreted as the change in dependent variable as the independent variable moves from the 1st to 10th decile. Therefore, the positive coefficient on the decile-ranked ?Rankingi,t (Surprisei,t) variable implies that an increase in the ?Rankingi,t (Surprisei,t) variable from the 1st to 10th decile results in a 2.7% (5.1%) three-day buy and hold abnormal return, which we believe is economically significant. The effect of the change in the firm’s performance ranking (?Rankingi,t) on the firm’s returns appears to be approximately 54.9% (= 0.028/0.051) of the effect of the firm’s earnings surprise. In summary, the results in Table 3 suggest that the improvement in the firm’s performance ranking is an important metric that investors use to assess the firm’s performance. These results also suggest that investors evaluate firm performance based on the entire distribution of earnings in the industry.5.2.Results for H2 and H3: Market reactions to performance ranking changes conditioning on the initial rankings and past performance ranking changesWe next examine Hypothesis 2, which predicts that the extent to which investors respond to the firm’s performance ranking change is expected to be greater (weaker) when the firm is initially evaluated as a low (high) performer in the industry. Table 4 present the regression results. Consistent with the results in the Table 3, the coefficient on the D_?Rankingi,t variable is equal to 0.036 and is significantly positive at 1% level, suggesting an increase from the 1st to 10th performance ranking decile results in a 3.6% increase in returns when the initial ranking of the firm is in its lowest decile. Consistent with H2, we find that the coefficient on the D_Initial Rankingi,t × D_?Rankingi,t interaction is equal to -0.023 and significantly negative at 1% level, indicating that the market’s reaction is less sensitive for firms that are in the 10th Initial Rankingi,t decile. We note that the sum of the coefficients on the D_?Rankingi,t variable and D_Initial Rankingi,t × D_?Rankingi,t interaction is still positive (1.1% = 3.7% - 2.6%) and significant at the 1% level. This evidence supports our hypothesis that investors’ response to the change in the firm’s performance ranking is greater when the firm is initially evaluated as a low performer. We also find a significantly positive coefficient on the D_Surprisei,t, variable (0.035) and a significantly positive coefficient on the D_Initial Rankingi,t × D_Surprisei,t, (0.039), indicating that the market reaction to the earnings surprise is lowest in the 1st Initial Rankingi,t decile and greatest in the 10th Initial Rankingi,t decile. The market’s reaction to the change in the firm’s performance ranking appears to be approximately 103% (18%) of the market’s reaction to the earnings surprise when the firm is initially evaluated as a low (high) performer. In summary, this evidence is consistent with the Hypothesis 2, which predicts that investors value an improvement in the firm’s performance ranking most when the firm is initially evaluated as a low performer in the industry.Next, we examine Hypothesis 3, which predicts that investors’ response to a change in the firm’s performance ranking is more (less) pronounced when the firm’s past volatility in performance ranking changes is low (high). Table 5 presents the estimation results. Consistent with the Hypothesis 3, we find that the coefficient on the D_?Rankingi,t variable is equal to 0.038 and significantly positive at the 1% level while the coefficient on the D_STD_?Rankingi,t × D_?Rankingi,t interaction is equal to -0.020 and significantly negative at the 1% level, indicating that the market reaction to the firm’s performance ranking change is greatest in the 1st STD_?Rankingi,t decile (3.8%) and decreases as we move to the 10th STD_?Rankingi,t decile (1.8% = 0.038 – 0.020). We also note that the sum of the coefficients on the D_?Rankingi,t variable and D_STD_?Rankingi,t × D_?Rankingi,t interaction is still positive and significant at the 1% level. The market’s reaction to the change in the firm’s performance ranking is approximately 83% (32%) of the market’s reaction to the earnings surprise when the firm’s past volatility in performance ranking changes is low (high).5.3.Results for H4: Earnings manipulation and firm performance ranking changesIn Hypothesis 4, we predict that management has incentives to manipulate earnings to affect the firms’ performance ranking within the industry. We present our results using equation (4) in Table 6. Consistent with our expectation, we find a positive coefficient on the Pos Excl Usei,t variable, which is significant at the 1% level in column (1) of Panel A. This evidence is consistent with managers using exclusions to manipulate the performance ranking of the firm. We then examine whether managers are more likely to opportunistically manage earnings when the benefits from doing so are greater. We predict and find evidence consistent with the market reacting more to changes in the performance ranking for low performers (Table 4), as suggested in Hypothesis 2. Therefore, we divide the full sample into terciles based on the Initial Rankingi,t variable and estimate equation (4) for each subsample. Columns (2), (3), and (4) in Panel A reports the estimation results using the firm/quarter observations in the 1st (Low group), 2nd (Med group), and 3rd (High group) Initial Rankingi,t terciles, respectively. Consistent with our prediction, we find the most significantly positive coefficient on the Pos Excl Usei,t variable is in the 1st Initial Rankingi,t tercile (column 2). We note that coefficient on the Pos Excl Usei,t variable decreases monotonically as we move to the 3rd initial ranking tercile in column (4). We statistically test the difference between the coefficient on the Pos Excl Usei,t variable in column (2) and (4) and find them to be statistically different at the 1% level. This evidence is consistent with managers manipulating the relative performance ranking of the firm when the benefits to doing so are the highest.In Panel B of Table 6, we substitute the Pos Excl Usei,t variable with the Pos Other Excl Usei,t and Pos Spec Items Usei,t variables to see whether unexpected exclusions are driving the results. Using the full sample, in column (1) we show that the coefficient on the Pos Other Excl Usei,t variable is positive and is significantly greater (1% level) than the coefficient on the Pos Special Items Usei,t variable. Similar to the results in Panel A, we find that the coefficient on the Pos Other Excl Usei,t variable is greatest in the 1st Initial Rankingi,t tercile (column 2) and appears to decrease monotonically as we move from the 1st to the 3rd Initial Rankingi,t tercile. It is also worth noting that the coefficients on the Pos Special Items Usei,t variable in column (2) through (4) become insignificant across all subsamples. This evidence corroborates our findings that managers are more likely to use positive exclusions to improve the firm’s performance ranking when the benefits are the greatest. 5.4.Results for H5: Differential market responses to the performance ranking change when the market suspects the opportunistic use of exclusionsIn this section, we examine H5, which predicts that the market discounts its response to the change in the firm’s performance ranking when the likelihood of manipulation is the highest. Table 7 presents the regression results using equation (5). Using the full sample, we find a significantly positive coefficient on the ?Rankingi,t variable in column (1) of Panel A, consistent with H1. Interestingly, we also find a significantly positive coefficient on the ?Rankingi,t × Pos Excl Usei,t interaction, which is contrary to expectations. It is possible that exclusions reflect both informative and opportunistic managerial choices. Therefore, the managerial incentives for using positive exclusions likely vary significantly. Similar to Table 6, we divide the full sample into three subsamples based on Initial Rankingi,t terciles and examine Hypothesis 5 using each subsample. Column (2) in Panel A reports the estimation results using the observation falling in the 1st Initial Rankingi,t tercile. We find that the coefficient on the ?Rankingi,t × Pos Excl Usei,t interaction is negative and significant at the 1% level. The coefficient of -0.025, suggests that the market’s reaction to an improvement in the firm’s performance ranking is discounted approximately 17.5% (-0.175 = -0.025 / 0.143) when using positive exclusions relative to the market reaction when positive exclusions are not present. We note that the coefficient on the interaction between the .?Rankingi,t and Pos Excl Usei,t variables is positive and significant in columns (3) and (4). This evidence seems to suggest that investors positively value the use of positive exclusions when examining all firms other than the lowest initial performers in the industry, suggesting that exclusions are performing more of an informational role when the incentives for managers to manipulate the performance ranking are lower.In Panel B of Table 7, we substitute the Pos Excl Usei,t variable for the Pos Other Excl Usei,t and Pos Special Items Usei,t variables. Using the full sample (presented in column 1), we find a positive and significant coefficient on the ?Rankingi,t variable, which is once again consistent with H1. We now find an insignificant and negative coefficient on interaction between the ?Ranking and Pos Other Excl Use variables. After splitting the sample into Initial Rankingi,t terciles (columns 2 through 4), we find a negative and significant (1% level) coefficient on the interaction between the ?Rankingi,t and Pos Other Excl Usei,t variables. This evidence continues to suggest that the market discounts improvements in the performance ranking of the firm when the likelihood of manipulation is the highest. The coefficient of -0.038 suggests that the market discounts the change in the performance ranking approximately 25% discount (-0.25 = -0.038 / 0.151) when positive other exclusions are used. Similar to Panel A, we continue to find evidence that the interaction between the ?Rankingi,t and Pos Other Excl Usei,t variables is positive and significant in column (3) and (4), suggesting that the market does not discount improvements in the firm’s performance ranking when the incentives for manipulation are low. When splitting the sample into Initial Rankingi,t terciles, we only find the interaction between the ?Rankingi,t and Pos Special Items Usei,t variables to be significant in column (2), providing no consistent evidence on how investors value management’s use of positive special items.6. Robustness tests6.1. The changes in analyst recommendations surrounding earnings announcementAs an additional robustness test, we include changes in the consensus analyst recommendations to ensure that we are not documenting the change in analyst rankings that might be occurring around the earnings announcement. Stickel (1985) provides evidence that the market reacts to changes in relative recommendations by Value Line Investments. Boni and Womack (2006) provide evidence that stock recommendations are associated with the rankings of firms within an industry. However, stock recommendations do not only reflect changes in the firm’s performance ranking. Stock recommendations could also reflect industry trends or conditions. Nevertheless, it is possible that a change in analyst recommendations coincide with changes in the relative performance of the firm within the industry. Therefore, we re-run each of our tests previously reported in this paper including the change in the consensus analyst recommendation as an additional independent variable to ensure that the abnormal returns surrounding the earnings announcement are not attributed to the changes in relative recommendations by analysts. All of our results are qualitatively and quantitatively similar after controlling analyst recommendation changes surrounding the earnings announcement. 7. Conclusion In this study, we predict and find evidence that investors positively value improvements in the firm’s performance ranking within the industry. We measure the change in the firm’s performance ranking within the industry by considering how a realized ranking based on the released EPS figure at the earnings announcement date is different from an initially expected ranking based on analyst consensus EPS forecasts calculated immediately prior to the earnings announcement. Using this measure, we specifically find that the buy-and-hold market-adjusted abnormal returns surrounding the earnings announcement are positively associated with the change in the firm’s performance ranking within the industry after controlling the earnings surprise and other control variables. We also predict and find that if the firm is initially evaluated as a low (high) performer within the industry or has a history of stable (volatile) performance ranking changes, the market response to the change in the firm’s performance ranking is stronger (weaker). Finally, we predict and find that managers opportunistically exclude expenses from non-GAAP earnings to influence the investors’ perception regarding the firm’s performance ranking. However, it appears that investors price protect themselves by discounting the change in the firm’s performance ranking that is associated with income-increasing exclusions.This study provides empirical evidence that earnings realizations convey useful information to investors regarding the firm’s ability to compete within the industry. The prior research has examined the information conveyed at specific points in the earnings distribution. We attempt to provide evidence that investors use the entire distribution of earnings within the industry to assess the firm’s performance. We also believe that the evidence in this paper provides one explanation for why firms might engage in earnings management activities even when they are comfortably above or below traditional benchmarks. Lastly, we further the relative performance literature by providing additional evidence on how investors value firms based on relative benchmarks within the industry. ReferencesAbarbanell, J. and Lehavy, R. 2003. “Biased Forecasts or Biased Earnings? The Role of Reported Earnings in Explaining Apparent Bias and Over/underreaction in Analysts’ Earnings Forecasts.” Journal of Accounting and Economics 36 (1): 105–46.Abarbanell, J. and Bushee J. B. 1998. “Abnormal Returns to a Fundamental Analysis Strategy.” The Accounting Review, 19–45.Bhojraj, S., Lee, C., and Oler, D. K. 2003. “What’s My Line? A Comparison of Industry Classification Schemes for Capital Market Research.” Journal of Accounting Research 41 (5): 745–74.Boni, L. and Womack, K. L. 2006. “Analysts, Industries, and Price Momentum.” Journal of Financial and Quantitative Analysis 41 (01): 85–109.Bowen, R. M., Davis, A. K., and Matsumoto, D. A.. 2005. "Emphasis on pro forma versus GAAP earnings in quarterly press releases: Determinants, SEC intervention, and market reactions." The Accounting Review 80, no. 4: 1011-1038.Burgstahler, D., and Dichev, I., 1997. "Earnings management to avoid earnings decreases and losses." Journal of accounting and economics 24, no. 1: 99-126.Burgstahler, D. and Eames, M. 2006. “Management of Earnings and Analysts’ Forecasts to Achieve Zero and Small Positive Earnings Surprises.” Journal of Business Finance & Accounting 33 (5‐6): 633–52.Bushman, R., E. Engel, and A. Smith. 2006. An analysis of the relation between the stewardship and valuation roles of earnings. Journal of Accounting Research 44: 53-83.Calia, M., 2014. “Fiat Chrysler, GM sales jump, Ford sales slip”, The Wall Street Journal, 1 MayDe Angelis, D., and Grinstein, Y., 2014. “Relative Performance Evaluation in CEO Compensation: A Non-Agency Explanation.” Available at SSRN 2432473.Doyle, J. T., Lundholm, R. J., and Soliman, M. T., 2003. "The predictive value of expenses excluded from pro forma earnings." Review of Accounting Studies 8, no. 2-3: 145-174.Carter, M. E., Ittner, C. D., and Zechman, S. 2009. “Explicit Relative Performance Evaluation in Performance-Vested Equity Grants.” Review of Accounting Studies 14 (2-3): 269–306.Degeorge, F., Patel, J., and Zeckhauser, R. 1999. “Earnings Management to Exceed Thresholds*.” The Journal of Business 72 (1): 1–33.Doyle, J., Jennings, J., and Soliman, M. 2013. “Do Managers Define Non-GAAP Earnings to Meet or Beat Analyst Forecasts?” Journal of Accounting and Economics 56 (1): 40–56.Paul, J. 1992. On the efficiency of stock-Based compensation. Review of Financial Studies 5:471-502.Fischer, P., Jennings, J., and Soliman, M. 2014. “Meeting, Beating, and Bubbles” Working paper.Gjesdal, F. 1981. Accounting for stewardship. Journal of Accounting Research: 208-231.Gow, I. D., Ormazabal, G., and Taylor, D. J., 2010. "Correcting for cross-sectional and time-series dependence in accounting research." The Accounting Review 85, no. 2: 483-512.Graham, J. R., Harvey, C. R., and Rajgopal, s., 2005. "The economic implications of corporate financial reporting." Journal of Accounting and Economics 40, no. 1: 3-73.Jayaratne, J. and Strahan, P. E. 1998. “Entry Restrictions, Industry Evolution, and Dynamic Efficiency: Evidence from Commercial Banking.” Journal of Law and Economics. 41: 239.Holmstrom, B. 1982. “Moral Hazard in Teams.” The Bell Journal of Economics, 324–40.Jayaratne, J., and Strahan, P. E., 1998. "Entry restrictions, industry evolution, and dynamic efficiency: Evidence from commercial banking." Journal of Law and Economics. 41Kadan, O., Madureira, L., Wang, R., and Zach., T. 2012. “Analysts’ Industry Expertise.” Journal of Accounting and Economics 54 (2): 95–120.Kasznik, R. and McNichols, M. F., 2002. “Does Meeting Earnings Expectations Matter? Evidence from Analyst Forecast Revisions and Share Prices.” Journal of Accounting Research 40 (3): 727–59.Lambert, Richard A. 2001. “Contracting Theory and Accounting.” Journal of Accounting and Economics 32 (1): 3–87.Levitt, A. 1998 “The 'Numbers Game" – Remarks of Chairman Arthur Levitt at the N.Y.U. Center for Law and Business, New York, N.Y.Litov, L. P., Moreton, P., and Zenger, T. R., 2012. "Corporate strategy, analyst coverage, and the uniqueness paradox." Management Science 58, no. 10: 1797-1815.Matsumoto, D. A. 2002. “Management’s Incentives to Avoid Negative Earnings Surprises.” The Accounting Review 77 (3): 483–514.McGahan, A. M., and Porter, M. E., 2002. "What do we know about variance in accounting profitability?" Management Science 48, no. 7: 834-851.Milgrom, P., Roberts, J., 1992. Economics, Organization & Management. Prentice Hall, Englewood Cliffs, NJ.Murphy, K. J. 1999. “Executive Compensation.” Handbook of Labor Economics 3: 2485–2563.Nelson, R. R., 1991. "Why do firms differ, and how does it matter?" Strategic management journal 12, no. S2: 61-74.Nickell, S. J. 1996. “Competition and Corporate Performance.” Journal of Political Economy 104 (4): 724.Orlik, T., 2012. “Lenovo's Uncertain Prize for Taking First Place in Global PCs”, The Wall Street Journal, 28 October, Peteraf, M. A., 1993. "The cornerstones of competitive advantage: A resource‐based view." Strategic management journal 14, no. 3: 179-191.Petersen, M. A., 2009. "Estimating standard errors in finance panel data sets: Comparing approaches." Review of financial studies 22, no. 1: 435-480.Rogers, C., 2013. “Chrysler Earnings To Weigh On an IPO”, The Wall Street Journal, 28 OctoberRoychowdhury, S. 2006. “Earnings Management through Real Activities Manipulation.” Journal of Accounting and Economics 42 (3): 335–70.Rumelt, R. P., 1991. "How much does industry matter?" Strategic Management Journal 12, no. 3: 167-185.Schmidt, K. M., 1997. "Managerial incentives and product market competition." The Review of Economic Studies 64, no. 2: 191-213. Stiroh, K. J., and Strahan, P. E., 2003. "Competitive dynamics of deregulation: Evidence from US banking." Journal of Money, Credit, and Banking 35, no. 5: 801-828.Skinner, D. J. and Sloan, R. G. 2002. “Earnings Surprises, Growth Expectations, and Stock Returns or Don’t Let an Earnings Torpedo Sink Your Portfolio.” Review of Accounting Studies 7 (2-3): 289–312.Sloan, R. G. 1996. “Do Stock Prices Fully Reflect Information in Accruals and Cash Flows about Future Earnings?” The Accounting Review, 289–315.Stickel, S. E. 1985. “The Effect of Value Line Investment Survey Rank Changes on Common Stock Prices.” Journal of Financial Economics 14 (1): 121–43.Stiroh, K. J, and Strahan, P. E. 2003. “Competitive Dynamics of Deregulation: Evidence from US Banking.” Journal of Money, Credit, and Banking 35, no. 5: 801–828.Waring, G. F., 1996. "Industry differences in the persistence of firm-specific returns." The American Economic Review: 1253-1265.Appendix A. Variable DefinitionsVariablesDescriptions3DayReti,t3DayReti,t is firm i’s three-day buy-and-hold stock returns centered on the earnings announcement in quarter t less three-day value-weighted CRSP market returns over the same window.?Rankingi,t?Rankingi,t is measured by firm i’s realized ranking at the earnings announcement in quarter t less firm i’s initial ranking two days prior to the earnings announcement in quarter t. Realized (Initial) ranking is equal to firm i’s performance ranking within 6-digit GICS industry based on firms’ realized (expected) earnings divided by lagged market capitalization. Realized (expected) earnings are measured by IBES-reported Actual EPS (the consensus IBES median analyst forecast) multiplied by the number of shares outstanding. If industry peer firms have already announced earnings, peer firms’ announced earnings are used to determine firm i’s realized and initial rankings; otherwise peer firms’ expected earnings (i.e., the consensus IBES median analyst forecast) are used.Initial Rankingi,tInitial Rankingi,t is firm i’s initial ranking in the industry in quarter t and it is described above.STD_?Rankingi,tSTD_?Rankingi,t is measured by the standard deviation of the ?Rankingi,t variable using past 12 quarters observations (a minimum of 8 quarters observations is required). ln(NUMESTi,t)ln(NUMESTi,t) is equal to the natural logarithm of the number of analysts following firm i in quarter t.Surprisei,tSurprisei,t is the firm i’s earnings surprise in quarter t as measured by firm i’s IBES-reported Actual EPS (IBES item VALUE) less the consensus median EPS forecast in quarter t. If the dependent variable in the regression equation is stock returns, the earnings surprise is scaled by lagged price (Compustat item prccq). We calculate the median EPS consensus based on individual analyst forecasts, which are required to be reported within a 90-day window preceding the daily date to ensure that our analyst consensus is not based on stale forecasts. We exclude individual analyst forecasts if I/B/E/S excludes the forecasts from calculating IBES-reported median EPS consensus. If the daily median EPS consensus analyst forecast is missing, we supplement our data by using IBES-reported median EPS consensus forecasts (i.e., IBES item MEDEST)MBEi,tMBEi,t is an indicator variable equal to one if the Surprisei,t variable is positive; zero otherwiseln(SIZEi,t)ln(Sizei,t) is equal to the natural logarithm of firm i’s market value of equity at the end of quarter t (Compustat item prccq multiplied by Compustat item cshoq).Book-to-Marketi,tBook-to-Marketi,t is measured by firm i’s book value of equity (Compustat item seqq) divided by the market value of equity at the end of quarter t.SalesGrowthi,tSalesGrowthi,t is equal to firm i’s net sales in quarter t (Compustat item saleq) divided by net sales in quarter t-4.Accrualsi,tAccruals is measured as firm i’ GAAP EPS (Compustat item epspxq) less cash flows from operation per share in quarter t (Compustat item oancfy divided by Compustat item cshprq) divided by price at the end of quarter t-1 (Compustat item prccq).Profitablei,tProfitablei,t is an indicator variable equal to one if firm i’s IBES-reported Actual EPS in quarter t is positive; zero, otherwise.ROAi,tROAi,t is measured by firm i’s IBES-reported Actual EPS in quarter t divided by total assets per share at the end of quarter t (Compustat item atq divided by cshprq or cshfdq depending on the IBES basic/diluted flag).Pos Excl Usei,tPos Excl Usei,t is indicator variable equal to one if exclusions are positive; zero otherwise. Exclusions are defined as firm i’s IBES-reported Actual EPS less GAAP EPS before extraordinary items in quarter t (Compustat item epspxq or epsfxq depending on IBES basic/diluted flag).Pos Special Items Usei,tPos Special Items Usei,t is an indicator variable equal to one if special items are positive; zero otherwise. Special items are defined as firm i’s operating income per share (Compustat item opepsq) less GAAP EPS before extraordinary items in quarter t.Pos Other Excl Usei,tPos Other Excl Usei,t is an indicator variable equal to one if other exclusions are positive; zero otherwise. Other exclusions are defined as exclusions less special items.Table 1 Descriptive statisticsThis table presents descriptive statistics for all sample firms with available information. The sample period ranges from 1997 to 2012. All variables are defined in the appendix. All continuous variables are winsorized at the 1% and 99% level.?NMeanStdP25MedianP753DayReti,t142,5810.0010.085-0.0400.0000.042Initial Rankingi,t142,5810.5140.2740.2860.5170.744?Rankingi,t142,5810.0110.134-0.0130.0000.048STD_?Rankingi,t142,5810.1070.0820.0450.0830.146NUMESTi,t142,5815.7065.4802.0004.0008.000Surprisei,t142,5810.0000.155-0.0200.0100.040MBEi,t142,5810.6670.4710.0001.0001.000ln(SIZEi,t)142,5816.6811.7695.4236.5927.834Book-to-Marketi,t142,5810.6260.5710.2870.4930.788Sales Growthi,t142,5811.1480.4420.9641.0811.228Accrualsi,t142,581-0.0580.146-0.083-0.031-0.002Profitablei,t142,5810.7750.4171.0001.0001.000ROAi,t142,5810.0030.0390.0010.0090.020Table 2 CorrelationsThis table presents Pearson (Above) / Spearman (Below) correlations. Correlations that are significant at 1% level are bolded.(1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(1)3DayReti,t-0.050.220.000.010.200.240.010.010.050.010.120.11(2)Initial Rankingi,t0.05--0.15-0.050.100.070.080.22-0.080.060.000.500.41(3)?Rankingi,t0.28-0.02--0.030.040.610.560.05-0.060.070.070.220.16(4)STD_?Rankingi,t-0.01-0.05-0.01--0.23-0.10-0.14-0.280.25-0.05-0.14-0.07-0.04(5)NUMESTi,t0.020.090.08-0.24-0.080.110.56-0.130.020.050.160.16(6)Surprisei,t0.300.070.86-0.040.11-0.540.12-0.130.090.170.290.25(7)MBEi,t0.250.080.77-0.150.120.82-0.14-0.120.090.090.280.23(8)ln(SIZEi,t)0.030.220.08-0.260.560.150.14--0.340.050.050.370.35(9)Book-to-Marketi,t0.01-0.02-0.050.26-0.15-0.06-0.11-0.31--0.16-0.28-0.16-0.07(10)Sales Growthi,t0.090.120.12-0.100.060.140.140.12-0.24-0.110.060.05(11)Accrualsi,t-0.01-0.080.04-0.110.040.040.06-0.05-0.230.13-0.04-0.03(12)Profitablei,t0.130.490.26-0.040.180.270.280.37-0.060.19-0.09-0.65(13)ROAi,t0.130.520.26-0.160.190.300.310.38-0.320.270.000.72-Table 3 Investor response to the effect of the change in the performance rankingThis table presents estimation results from the regression of three-day buy-and-hold abnormal returns surrounding the earnings announcement (3DayReti,t) on the firm's performance ranking changes (?Rankingi,t), control variables, and calendar quarter/industry fixed effects. In column (5), all variables are decile-ranked. All variables are defined in the Appendix. All continuous variables are winsorized at the 1% and 99% levels. Following Peterson (2009), we cluster the standard errors by time and firm to correct the standard errors for cross-sectional and serial-correlation. *, **, and *** represent significance level at the 10%, 5%, and 1% respectively. VariablesPred(1)(2)(3)(4)(5)Decile-Ranked?Rankingi,t+0.145***-0.124***0.134***0.028***(35.759)-(28.494)(31.171)(14.854)Surprisei,t+-0.668***0.287***0.244***0.051***-(19.902)(11.182)(10.066)(22.182)STD_?Rankingi,t+ / ?---0.0030.000---(0.776)(0.097)Initial Rankingi,t+---0.024***0.008***---(15.696)(7.273)Book-to-Marketi,t+ / ?---0.004***0.006***---(6.073)(5.343)ln(Sizei,t)+ / ?----0.000-0.001---(-1.368)(-0.507)Sales Growthi,t+---0.008***0.017***---(10.205)(17.303)Accrualsi,t?----0.005-0.005***---(-1.489)(-4.445)Constant-0.001***0.002***0.000-0.022***-0.053***(-3.760)(16.610)(1.434)(-10.189)(-28.421)Industry-Quarter F.E.YesYesYesYesYesObservations142,581142,581142,581142,581142,581Adjusted R-squared?0.0680.0430.0710.0790.102Table 4 The effect of the firm’s position in the industryThis table presents estimation results from the regression of three-day buy-and-hold abnormal returns surrounding the earnings announcement (3DayReti,t) on the decile-ranked change in the firm's performance ranking (D_?Rankingi,t), decile-ranked control variables, and calendar quarter/industry fixed effects. All variables in this table are decile-ranked and defined in the Appendix. Following Peterson (2009), we cluster the standard errors by time and firm to correct the standard errors for cross-sectional and serial-correlation. *, **, and *** represent significance level at the 10%, 5%, and 1% respectively. VariablesPred(1)D_?Rankingi,t+0.036***(12.192)D_Surprisei,t+0.035***(12.587)D_Initial Rankingi,t × D_?Rankingi,t?-0.023***(-4.485)D_Initial Rankingi,t × D_Surprisei,t+ / ?0.039***(8.515)D_STD_?Rankingi,t+ / ?0.000(0.151)D_Initial Rankingi,t+0.000(0.221)D_Book-to-Marketi,t+ / ?0.006***(5.327)D_ln(Sizei,t)+ / ?-0.000(-0.235)D_Sales Growthi,t+0.017***(17.094)D_Accrualsi,t?-0.005***(-4.245)Constant-0.050***(-25.280)Industry-Quarter F.E.YesObservations142,581Adjusted R-squared?0.103Table 5 The effect of the volatility of past changes in the firm’s performance rankingThis table presents estimation results from the regression of three-day buy-and-hold abnormal returns surrounding the earnings announcement (3DayReti,t) on the decile-ranked change in the firm's performance ranking (D_?Rankingi,t), decile-ranked control variables, and calendar quarter/industry fixed effects. All variables in this table are decile-ranked and defined in the Appendix. Following Peterson (2009), we cluster the standard errors by time and firm to correct the standard errors for cross-sectional and serial-correlation. *, **, and *** represent significance level at the 10%, 5%, and 1% respectively. VariablesPred(1)D_?Rankingi,t+0.038***(10.223)D_Surprisei,t+0.046***(12.685)D_STD_?Rankingi,t × D_?Rankingi,t?-0.020***(-3.640)D_STD_?Rankingi,t × D_Surprisei,t+ / ?0.011**(2.268)D_STD_?Rankingi,t+ / ?0.005**(2.537)D_Initial Rankingi,t+0.008***(7.057)D_Book-to-Marketi,t+ / ?0.006***(5.379)D_ln(Sizei,t)+ / ?-0.001(-0.547)D_Sales Growthi,t+0.017***(17.351)D_Accrualsi,t?-0.005***(-4.449)Constant-0.056***(-24.785)Industry-Quarter F.E.YesObservations142,581Adjusted R-squared?0.102Table 6 The effects of the use of positive exclusions on the change in the performance rankingThis table presents estimation results from the regression of the change in the firm's performance ranking (?Rankingi,t) on the use of positive exclusions (Pos Excl Usei,t), control variables, and calendar quarter/industry fixed effects. In panel B, we replace the use of positive exclusions with the use of positive other exclusions (Pos Other Excl Usei,t) and the use of positive special items (Pos Special Items Usei,t). We divide the full sample into Low, Medium, and High groups based on terciles of the Initial Rankingi,t variable. All variables are defined in the Appendix. Following Peterson (2009), we cluster the standard errors by time and firm to correct the standard errors for cross-sectional and serial-correlation. *, **, and *** represent significance level at the 10%, 5%, and 1% respectively. Panel A Results based on Positive Exclusions UseVariablesPred(1)?(2)?(3)?(4)Initial RankingFullLowMedHighPos Excl Usei,t+0.005***0.007***0.006***-0.001(6.463)(6.012)(4.928)(-0.439)Book-to-Marketi,t+ / ?0.006***-0.0000.009***-0.004**(4.649)(-0.295)(2.874)(-2.029)Sales Growthi,t+0.008***0.010***0.007***0.008***(8.745)(8.895)(4.472)(4.340)ln(Sizei,t)+ / ?-0.004***-0.005***-0.003***-0.003***(-7.290)(-9.251)(-5.376)(-5.929)Profitablei,t+0.097***0.110***0.136***0.206***(20.735)(21.917)(20.117)(16.812)ROAi,t+0.333***0.464***0.790***0.309***(13.272)(16.272)(15.167)(4.113)MBEi,t+0.142***0.093***0.168***0.142***(59.831)(44.986)(48.520)(54.696)ln(NUMESTi,t)+ / ?-0.002***-0.003***-0.002**-0.002**(-3.127)(-3.340)(-2.231)(-2.153)STD_?Rankingi,t+ / ?0.060***0.337***0.186***-0.281***(4.084)(26.101)(11.534)(-17.796)Initial Rankingi,t?-0.180***-0.278***-0.183***-0.166***(-24.826)(-18.629)(-20.743)(-23.086)Constant-0.061***-0.035***-0.140***-0.133***(-10.006)(-6.323)(-13.241)(-8.417)Industry-Quarter F.E.YesYesYesYesObservations142,58147,67647,54047,365Adjusted R-squared?0.431?0.437?0.498?0.545?Coeff. (High - Low)Pos Excl Usei,t0.007***(4.600)?????????Panel B Results based on Positive Other Exclusions and Positive Special Items UseVariablesPred(1)?(2)?(3)?(4)Initial RankingFullLowMedHighPos Other Excl Usei,t+0.010***0.014***0.011***0.002(13.672)(12.006)(7.915)(1.614)Pos Special Items Usei,t+ / ?0.003***0.001-0.000-0.002(4.232)(0.515)(-0.316)(-1.215)Book-to-Marketi,t+ / ?0.006***-0.0000.008***-0.004**(4.474)(-0.316)(2.765)(-2.162)Sales Growthi,t+0.008***0.010***0.007***0.009***(8.748)(9.052)(4.530)(4.418)ln(Sizei,t)+ / ?-0.004***-0.005***-0.003***-0.003***(-7.689)(-9.281)(-5.489)(-6.031)Profitablei,t+0.097***0.110***0.136***0.206***(20.712)(21.882)(20.201)(16.789)ROAi,t+0.327***0.455***0.789***0.313***(13.058)(16.178)(15.240)(4.186)MBEi,t+0.142***0.093***0.168***0.142***(60.069)(45.099)(48.800)(54.783)ln(NUMESTi,t)+ / ?-0.002***-0.003***-0.002**-0.002**(-3.409)(-3.489)(-2.334)(-2.158)STD_?Rankingi,t+ / ?0.059***0.335***0.184***-0.282***(4.017)(26.055)(11.471)(-17.745)Initial Rankingi,t?-0.181***-0.275***-0.183***-0.166***(-25.069)(-18.692)(-20.770)(-23.057)Constant-0.062***-0.037***-0.140***-0.133***(-10.172)(-6.733)(-13.284)(-8.404)Industry-Quarter F. E.YesYesYesYesObservations142,58147,67647,54047,365Adjusted R-squared?0.432?0.438?0.499?0.545?Coeff. (High - Low)Pos Other Excl Usei,t0.012***(6.801)Pos Special Items Usei,t0.002(1.388)?????????Table 7 Investor response to the ranking change with the use of positive exclusionsThis table presents estimation results from the regression of three-day buy-and-hold abnormal returns surrounding the earnings announcement (3DayReti,t) on the change in the firm's performance ranking (?Rankingi,t), the use of positive exclusions (Pos Excl Usei,t), the interaction of those two variables, control variables, and calendar quarter/industry fixed effects. In panel B, we replace the use of positive exclusions with the use of positive other exclusions (Pos Other Excl Usei,t) and the use of positive special items (Pos Special Items Usei,t). We divide the full sample into Low, Medium, and High groups based on terciles of the Initial Rankingi,t variable. All variables are defined in the Appendix. Following Peterson (2009), we cluster the standard errors by time and firm to correct the standard errors for cross-sectional and serial-correlation. *, **, and *** represent significance level at the 10%, 5%, and 1% respectively.Panel A Results based on Positive Exclusions UseVariablesPred(1)??(2)?(3)?(4)Initial RankingFullLowMedHigh? Rankingi,t+0.130***0.143***0.157***0.124***(28.118)(20.001)(21.552)(17.278)Pos Excl Usei,t?-0.007***-0.005***-0.009***-0.006***(-12.119)(-5.345)(-9.546)(-6.011)? Rankingi,t × Pos Excl Usei,t?0.013***-0.025**0.027***0.033***(3.235)(-2.411)(3.974)(4.500)Surprisei,t+0.250***0.269***-0.0970.272***(10.413)(8.809)(-1.370)(3.894)STD_?Rankingi,t+ / ?0.002-0.026***-0.012*0.039***(0.509)(-4.516)(-1.656)(5.282)Initial Rankingi,t+0.024***0.031***0.022***0.032***(15.777)(6.307)(4.794)(6.742)Book-to-Marketi,t+ / ?0.005***0.004***0.006***0.008***(6.638)(4.083)(3.670)(5.821)ln(Sizei,t)+ / ?-0.0000.0010.001*-0.001**(-0.114)(1.526)(1.770)(-2.320)Sales Growthi,t+0.008***0.006***0.009***0.007***(9.972)(6.217)(6.109)(5.105)Accrualsi,t?-0.008**-0.011***-0.009-0.005(-2.447)(-2.723)(-1.319)(-1.039)Constant-0.022***-0.022***-0.026***-0.029***(-9.993)(-6.686)(-6.265)(-5.398)Industry-Quarter F. E.YesYesYesYesObservations142,68747,61147,51547,561Adjusted R-squared?0.080?0.076?0.097?0.081?Coeff. (High - Low)? Rankingi,t × Pos Excl Usei,t-0.058***(-4.257)?????????Panel B Results based on Positive Other Exclusions and Special Items UseVariablesPred??(1)?(2)?(3)Initial RankingFullLowMedHigh?Rankingi,t+0.129***0.151***0.152***0.123***(24.267)(18.624)(19.383)(14.136)Pos Other Excl Usei,t?-0.007***-0.006***-0.009***-0.006***(-10.325)(-4.506)(-9.036)(-5.530)? Rankingi,t × Pos Other Excl Usei,t?-0.003-0.038***0.013*0.023***(-0.573)(-4.440)(1.653)(2.894)Pos Special Items Usei,t+ / ?0.001**0.002**0.0000.001(2.033)(2.127)(0.530)(1.602)? Rankingi,t × Pos Special Items Usei,t+ / ?0.015***-0.0030.023***0.013(3.086)(-0.333)(3.917)(1.389)Surprisei,t+0.254***0.272***-0.0940.273***(10.454)(8.845)(-1.321)(3.928)STD_?Rankingi,t+ / ?0.004-0.025***-0.0110.040***(0.954)(-4.270)(-1.477)(5.440)Initial Rankingi,t+0.023***0.030***0.022***0.031***(15.196)(6.124)(4.635)(6.740)Book-to-Marketi,t+ / ?0.005***0.004***0.006***0.008***(6.617)(3.993)(3.844)(5.873)ln(Sizei,t)+ / ?-0.0000.0000.000-0.001***(-0.824)(1.106)(1.329)(-2.797)SalesGrowthi,t+0.007***0.006***0.009***0.007***(9.888)(6.078)(6.096)(5.039)Accrualsi,t?-0.007**-0.010**-0.007-0.003(-1.981)(-2.378)(-0.995)(-0.736)Constant-0.021***-0.022***-0.026***-0.029***(-9.753)(-6.485)(-6.150)(-5.446)Industry-Quarter F. E.YesYesYesYesObservations142,58147,67647,54047,365Adjusted R-squared?0.080?0.078?0.096?0.081?Coeff. (High - Low)? Ranking × Pos Other Excl Use-0.057***(-5.075)? Ranking × Pos Special Items Use-0.017(-1.483)????????? ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download