Scholarly article on topic 'Manager-Specific Effects on Earnings Guidance: An Analysis of Top Executive Turnovers'

Manager-Specific Effects on Earnings Guidance: An Analysis of Top Executive Turnovers Academic research paper on "Sociology"

0
0
Share paper
Academic journal
Journal of Accounting Research
OECD Field of science
Keywords
{""}

Academic research paper on topic "Manager-Specific Effects on Earnings Guidance: An Analysis of Top Executive Turnovers"

DOI: 10.1111/j.1475-679X.2011.00420.x Journal of Accounting Research Vol. 49 No. 5 December 2011 Printed in U.S.A.

CHIC!M№

Manager-Specific Effects on Earnings Guidance: An Analysis of Top Executive Turnovers

FRANCOIS BROCHET, * LUCILE FAUREL, t AND SARAH MCVAY*

Received 22 May 2009; accepted 21 May 2011

ABSTRACT

We investigate how managers contribute to the provision of earnings guidance by examining the association between top executive turnovers and guidance. Although firm and industry characteristics are important determinants of guidance, we conclude that CEOs participate in firm-level policy decisions, whereas CFOs are involved in the formation or discussion of guidance. Among firms that historically issued frequent guidance, breaks in guidance following CEO turnovers are relatively permanent and are potentially attributable to firm-initiated changes in guidance policy. Breaks following CFO turnovers, however, likely reflect uncertainty on the part of the newly appointed executive—they are concentrated in the two quarters following the turnover, are associated with the background of the newly appointed CFO, and extend to the relative precision of the guidance. Among firms that did not issue guidance historically, we find some evidence that newly appointed externally hired CEOs increase the likelihood of providing guidance.

* Harvard University; ^University of California, Irvine; ^University of Utah. We would like to thank an anonymous reviewer, Brian Cadman, Asher Curtis, Terry Grant, Rachel Hayes, Michael Kimbrough, Greg Miller, Mort Pincus, Doug Skinner (the Editor), Irem Tuna, Isabel Wang, and workshop participants at the 2008 AAA Conference, the 2008 FARS Conference, Harvard University, Temple University, Texas Christian University, and the University of Utah for their helpful comments and suggestions.

Copyright ©, University of Chicago on behalf of the Accounting Research Center, 2011

1. Introduction

We investigate the role of top executives in the provision and formation of quarterly management earnings forecasts (hereafter earnings guidance). Although theory suggests that managers use the provision of earnings guidance to signal their ability to anticipate changes in the economic environment (e.g., Trueman [1986]), relatively little is known about how much managers contribute to the choice to disclose, and the construction of, earnings guidance. Research has begun to investigate the general effects of specific managers by examining manager fixed effects. For example, Bertrand and Schoar [2003] document that specific managers are associated with corporate decisions, such as acquisitions or research and development expenditures, while Bamber, Jiang, and Wang [2010] reach similar conclusions when examining annual and quarterly management earnings forecasts. The specific role of top executives in the provision of earnings guidance, however, remains unclear. For example, how does a CEO or CFO's firm- or industry-specific knowledge or forecasting experience affect the provision of guidance? And are the significant fixed effects documented in Bamber, Jiang, and Wang [2010] associated with relatively temporary or permanent changes in the provision of guidance? We use the setting of top executive turnovers to examine these open questions.1

Because guidance is a very "sticky" disclosure (e.g., Gibbons, Richardson, and Waterhouse [1990], Bamber, Jiang, and Wang [2010]), we focus our analysis on firms with a discernable preexisting guidance policy, namely those that, in the prior two years, frequently issued guidance (hereafter frequent guiders) or never issued guidance (hereafter nonguiders).2 Among frequent guiders, we document breaks in earnings guidance following both CEO and CFO turnovers. The breaks in guidance following CEO turnovers are persistent, extending through the next eight quarters, while there is no evidence that the breaks in guidance following CFO turnovers extend beyond the next two quarters. Among nonguiders, we find some evidence of an increase in the provision of guidance following new appointments of

1 It is possible that turnovers coincide with firm-initiated changes in guidance policy. Moreover, firms may hire executives with particular backgrounds to facilitate their chosen guidance policy, a limitation that extends to the use of manager fixed effects. We discuss and examine the possibility that changes in top-level management and guidance issuance are endogenously determined in section 4.2.

2 Conditioning on the preexisting guidance policy allows us to better disentangle effects associated with the incoming executives, as the firm's preexisting disclosure policy is a key determinant of guidance issuance (e.g., Gibbons, Richardson, and Waterhouse [1990], Yang [2010]). We consider frequent guiders to be those that issued guidance in at least four of the eight prior quarters. Although our focus in on frequent guiders and nonguiders, we also examine infrequent guiders: those that issued guidance at least once in the prior two years, but do not meet the definition of frequent guiders. Because we use First Call to identify instances of management guidance, we may be understating the frequency of guidance issuance (Chuk, Matsumoto, and Miller [2009]). We investigate how using First Call affects our inferences in section 3.1.

externally hired CEOs, and no association between the provision of guidance and CFO turnovers.

We next investigate the role of endogeneity, as it is possible that turnovers are correlated with firm-initiated changes in guidance policy. For example, changes in guidance could be in response to concurrent bad news events such as poor performance, pending litigation or missing prior analyst forecasts (Houston, Lev, and Tucker [2009], Rogers and Van Buskirk [2009], Bamber, Jiang, and Wang [2010], Feng and Koch [2010], Chen, Matsumoto, and Rajgopal [2011]) or changes in the composition of the board (Ajinkya, Bhojraj, and Sengupta [2005], Karamanou and Vafeas [2005]). Thus, we allow for the possibility that changes in top-level management and guidance issuance are endogenously determined. We first condition on performance (using the prior two-year size-adjusted stock return) and find some evidence that the association between CEO turnovers and breaks in guidance among frequent guiders is due, at least in part, to concurrent poor performance, but this concern does not extend to CFO turnovers or turnovers among nonguiders. We next control for concurrent appointments of new board chairmen, to proxy for firm-initiated changes in guidance policy in response to changes in disclosure preferences (e.g., Richardson, Tuna, and Wysocki [2003]). This variable is not statistically significant among nonguiders; however, among frequent guiders, we find that board chairman turnover is significantly associated with permanent breaks in guidance issuance, and the inclusion of this variable weakens the effect of CEO turnovers, but not CFO turnovers, on guidance issuance. These results suggest that breaks in guidance following CEO turnovers are potentially attributable to firm-initiated changes in guidance policy, either in response to poor performance or shifts in disclosure preferences instituted by the board. These alternative explanations are not supported among CFO turnovers. To corroborate these initial findings, among frequent guiders we examine the effects of plausibly exogenous turnovers—those where the executives were hired away from the firm.3 Among CFOs, but not CEOs, we continue to document breaks in guidance following the turnovers where the outgoing executives were hired away, supporting our general conclusion that the association between CFO turnovers and breaks in guidance is not driven by correlated firm-specific shocks.

To further investigate the role of top executives in guidance issuance, we collect information about the incoming executive's background to garner evidence on the new executive's anticipated degree of firm- and industry-specific knowledge, as well as their forecasting experience. We consider whether executives are hired from within the firm or from an outside firm,

3 We expect turnovers where the outgoing executives were hired away to be uncorrelated with firm-specific shocks, such as performance. We focus on the frequent guider sample, as this sample's response to the exogenous shock is discernable via a break in guidance. In contrast, among nonguiders a similar shock will not help to partition the sample, as the status quo is silence.

and among those hired externally, whether they have prior forecasting experience or experience in the same industry as their new firm.

We find no evidence that the association between turnovers and breaks in guidance varies with the backgrounds of the newly appointed CEOs, but among both frequent guiders and nonguiders, newly appointed CEOs with prior forecasting experience tend to be associated with increases in guidance, and these associations persist for at least two years following the turnover. This finding is consistent with CEOs being associated with permanent changes in guidance policy, either because of their personal preferences (e.g., Gibbons, Richardson, and Waterhouse [1990], Bamber, Jiang, and Wang [2010]) or because their appointment coincides with firm-initiated changes in guidance policy (e.g., Richardson, Tuna, and Wysocki [2003], Houston, Lev, and Tucker [2009], Chen, Matsumoto, and Rajgopal [2011]).

Among nonguiders, the association between CFO turnovers and guidance issuance does not vary significantly with CFO backgrounds in the eight quarters following the CFO turnovers. Among frequent guiders, however, we find that changes in guidance issuance following CFO turnovers vary with the backgrounds of the newly appointed CFOs. For example, breaks are more likely when newly appointed externally hired CFOs lack forecasting experience. These breaks are concentrated in the first two quarters following the new appointment, and may represent the time needed for incoming CFOs to familiarize themselves with their new firm before forming or discussing guidance. Interestingly, four to eight quarters following the new appointment, externally hired CFOs with prior forecasting experience are associated with an increase in guidance. As with CEOs, it is not clear whether this association represents a manager-specific effect or a concurrent firm-initiated guidance policy change. Finally, we gain similar inferences when examining guidance precision among frequent guiders. Specifically, we find no evidence of a reduction in precision among newly appointed CEOs, but find consistent evidence of a reduction in precision among newly appointed CFOs. This reduction is concentrated among CFOs who are external hires.

In sum, we use top executive turnovers to investigate if there is a manager-specific component to the provision of earnings guidance. We find some evidence that CEO turnovers are associated with permanent changes in guidance policy, although we cannot disentangle this effect from firm-initiated policy changes. We find, however, that CFO turnovers are associated with temporary breaks in guidance and these breaks are, in part, associated with incoming CFOs' implied knowledge about the firm.

Our paper contributes to the management forecast literature and, more generally, to the disclosure literature. Prior research has examined both the benefits and costs of issuing earnings guidance (e.g., Coller and Yohn [1997], Feng [2006], Rogers and Van Buskirk [2009]), as well as the strategic use of guidance (e.g., Bergman and Roychowdhury [2008], Rogers and Stocken [2005], Rogers [2008]), but has just begun exploring the

role of the firm versus management, and the specific roles of different managers. As noted above, our results suggest that CEOs and CFOs have distinct effects on guidance; we conclude that CEOs participate in firmlevel policy decisions, whereas CFOs are involved in the formation of guidance. Our findings also highlight the importance of conditioning on firms' preexisting guidance policies and concurrent shocks to the firm such as performance and changes in the board. Our research design choice complements studies that find significant incremental explanatory power of executive fixed effects for variation in firms' investments (Bertrand and Schoar [2003]), financial reporting (Ge, Matsumoto, and Zhang [2011]), tax avoidance (Dyreng, Hanlon, and Maydew [2010]), and voluntary disclosure (Bamber, Jiang, and Wang [2010]). Our results suggest that, while firm- and industry-specific characteristics are the dominant factors in the provision of guidance (e.g., 63% of frequent guiders still issue earnings guidance in the quarter following an executive turnover), CFO turnovers have an economically meaningful impact on guidance among frequent guiders, reducing the likelihood of providing guidance in the next quarter by approximately 7%, on average, and by over 13% if the newly appointed CFOs do not have prior forecasting experience.

2. Hypothesis Development

A great deal of research examines the choice to issue voluntary disclosure. Corporate managers often possess private information not reflected in stock prices and can disclose that information voluntarily, for example, through earnings guidance. Firms can benefit from voluntary disclosure because it can reduce information asymmetry (Diamond and Verrecchia [1991], Coller and Yohn [1997]), reduce the cost of capital (Botosan [1997]), increase analyst following (Healy, Hutton, and Palepu [1999]), and improve a firm's reputation for transparent and credible reporting (Williams [1996]). Consistent with this, Graham, Harvey, and Raj-gopal [2005] find that over 90% of managers surveyed indicate that developing a reputation for accurate and transparent reporting is a key factor motivating their voluntary disclosures.

Issuing earnings guidance, however, can also be costly (Feng and Koch [2010]). For example, falling short of expectations can damage managerial reputation (Graham, Harvey, and Rajgopal [2005], Feng [2006]) and expose firms to legal liability (e.g., Kasznik [1999], Soffer, Thiagarajan, and Walther [2000]). For this reason, research generally finds that earnings guidance is issued less frequently when earnings are more difficult to estimate (Waymire [1985]). Moreover, earnings guidance is more prevalent when demand is higher, where demand is proxied by institutional holdings, independent boards, and analyst following (e.g., Ajinkya, Bhojraj, and Sen-gupta [2005], Karamanou and Vafeas [2005]), and is less common when managers must rely on low-quality financial reports (Feng, Li, and McVay [2009]).

Recent research has begun to shed some light on the manager-specific elements of guidance. For example, Zamora [2009] examines specific CFOs and finds that those with superior forecasting ability receive higher pay, and more generally, Baik, Farber, and Lee [2011] document a positive association between managerial ability and forecast issuance and accuracy. Most related to our paper, Bamber, Jiang, and Wang [2010] find that managers have "styles" that are associated with their propensity to issue guidance and the nature of the resulting guidance (e.g., the precision of the guidance). They find that these styles vary with the backgrounds of the executives: whether the managers have an accounting or finance background, have an MBA, or were born prior to the Great Depression. Bamber, Jiang, and Wang [2010] focus on the taste functions of managers and follow managers across firms to identify these taste functions.

In our paper, we investigate manager-specific effects on the provision of guidance, conditional on the firm's preexisting guidance policy.4 Among frequent guiders, to the extent that newly appointed managers do not have sufficient expertise to form or discuss guidance, there may be a temporary break in guidance after the appointment of a new top executive. We can then infer which managers form or discuss the guidance by their hesitation to issue guidance when they lack firm- and industry-specific knowledge or forecasting experience. Alternatively, if guidance issuance is largely determined by the taste functions of top executives, but not their forecast knowledge, a newly appointed manager might (permanently) alter the frequency of guidance. Finally, it is possible that the provision of guidance does not require CEO or CFO input, in which case top executive turnovers are not expected to be associated with changes in guidance issuance.5 This leads to our hypothesis, stated in the null form:

H1: There is no association between top executive turnovers and changes to the provision of earnings guidance.

3. Sample Selection, Data, Variable Definitions, and Descriptive Statistics

3.1 SAMPLE SELECTION AND DATA

Our full sample of firms is identified using the intersection of the Ex-ecuComp and First Call databases (to identify turnovers and earnings

4 Bhojraj, Libby, and Yang [2011] document a positive correlation between the quantity and quality of guidance, but do not examine the effects of managers. We condition on the firm's preexisting guidance policy to isolate the effect of managers on the provision of guidance (see also Yang [2010]).

5 As noted in Gibbons, Richardson, and Waterhouse [1990, p. 130], when behaving ritualis-tically, "the firm exhibits a largely passive, even rote, adherence to perceived disclosure norms and does so using routinized, bureaucratized procedures." A firm wishing to change its guidance policy may also time this policy change with a turnover to ease implementation, and may hire an executive with a suitable background. We investigate this possibility in section 4.2.

guidance, respectively), and thus our sample is skewed toward larger firms. We conduct our tests using only post-Regulation Fair Disclosure firm-quarter observations (2001 through 2008) to establish a more homogeneous institutional environment, but include observations prior to 2001 when determining a firm's preexisting guidance policy. Because many firms appear to issue guidance sporadically (McNichols [1989], Rogers and Stocken [2005]), and the impact of executive turnovers on their guidance policy (or lack thereof) may be difficult to detect, we examine three subgroups: firms with a history of frequent guidance (frequent guiders), firms with a history of sporadic guidance (infrequent guiders), and firms with no history of guidance (nonguiders). Frequent guiders are those that issue guidance in at least four different quarters over the preceding eight quarters, and two over the preceding four quarters.6 We consider eight quarters a sufficiently long period to assess whether a firm has established a guidance policy of issuing guidance frequently, and we term the issuance of guidance at least every six months, on average, as "frequent." Infrequent guiders are those that issue guidance at least once in the prior eight quarters but do not meet our definition of "frequent" guiders, and nonguiders have no instances of guidance in the prior eight quarters.7 The guidance for quarter q must occur between the day earnings are announced for fiscal quarter q — 1 and the day the fiscal quarter q ends. We exclude all earnings preannounce-ments (issued between the fiscal quarter end and the earnings announcement date of each quarter), as this type of guidance is often issued for different reasons from standard earnings guidance (Skinner [1994, 1997]) and is generally viewed as an early earnings announcement rather than a late earnings forecast (Hirst, Koonce, and Venkataraman [2008]). Using these

6 Bhojraj, Libby, and Yang [2011] define frequent guiders as those in the top two quintiles of frequency, where frequency is the number of quarters in which a firm has issued guidance, divided by the number of quarters since its first guidance issuance in the sample period. The mean frequency is 0.42 and 0.75 for firms in the top two quintiles, which is consistent with frequent guiders providing guidance two to three times a year.

7 Chuk, Matsumoto, and Miller [2009] document systematic differences between forecasts reported on the First Call database and in company press releases. To mitigate the concern that changes in guidance are a byproduct of our use of First Call, we investigate whether incidences of "missed" guidance differ among turnover firms relative to nonturnover firms for both frequent guiders and nonguiders. Specifically, to determine if First Call missed the guidance, we conduct a keyword search in Factiva (as specified by Chuk, Matsumoto, and Miller [2009]) in all quarters q + 1 coded as "no guidance" according to First Call following (1) frequent guider turnovers and (2) nonguider CEO turnovers. In addition, we perform the same keyword search for a set of no-turnover firm-quarter observations matched with turnover firm-quarter observations based on prior guidance, time period, and analyst following. Although we note missing guidance observations (approximately 14% for frequent guiders and 5% for nonguiders), the difference between the incidence of forecasts missed by First Call between turnover and nonturnover firm quarters is not significant for either frequent guiders or nonguiders (not tabulated). Thus, it is unlikely that the breaks in guidance we document are a result of First Call's collection biases. This provides support to the overall validity of our inferences. Still, the reader should exercise caution in interpreting the absolute magnitude of the effect of executive turnover (and other variables) on guidance issuance.

criteria, we identify three distinct samples: a sample of frequent guiders with 7,660 firm-quarter observations representing 775 distinct firms, a sample of infrequent guiders with 12,259 firm-quarter observations covering 1,390 distinct firms, and a sample of nonguiders with 18,271 firm-quarter observations from 1,673 distinct firms.

To identify CEO and CFO turnovers, we first identify potential executive changes from ExecuComp. Because CFOs are not always tracked by Exe-cuComp, we then confirm CFO turnovers and identify turnover quarters using PR Newswire and Wall Street Journal articles from Factiva. We identify a total of 850 (1,143) CEO (CFO) turnovers; we are able to include a maximum of 716 (983) CEO (CFO) turnovers in our regression analyses.

Our regression analyses require financial statement data, which we retrieve from the Compustat Quarterly database, and returns data, which we obtain from the CRSP database. We acquire the analyst forecast-related variables from the First Call database to remain consistent with the source of earnings guidance data. We collect restatement data from the GAO Financial Restatement Database, litigation data from the Stanford Securities Class Action Clearinghouse, and board chairman data from the BoardEx database. Sample sizes vary depending on variable requirements for each test.

3.2 VARIABLE DEFINITIONS

3.2.1. Main Dependent Variables. We define each of our variables in appendix A. Our main explanatory variable is the turnover of a top executive. CEOTurnover (CFOTurnover) is an indicator variable equal to one for firm quarters during which there is a CEO (CFO) turnover, zero otherwise. We code quarter q as a turnover quarter if the incoming executive is appointed between the day earnings are announced for fiscal quarter q — 1 and the day before earnings are announced for fiscal quarter q. We consider two main dependent variables to test our hypothesis. NextGuidanceq+1,q+8 is the number of quarters, from one to eight, until the firm issues guidance after quarter q; if the firm has not issued guidance by quarter q + 8, we set NextGuidance equal to eight (i.e., the variable is right censored). Guidanceq+\q+n is an indicator variable that is equal to one if a firm issues guidance at least once in quarters q + 1 through q + n, zero otherwise (where n is either one, two, four, or eight).

3.2.2. Firm-and Industry-Specific Determinants of Guidance Issuance. In each of our regression analyses, we include firm- and industry-specific determinants of guidance issuance. Because these policy choices tend to be "sticky" within a firm, and our focus is on the change in guidance, we include the firm's guidance history. Guidanceq is equal to one if the firm issued guidance in quarter q, and NbGuidanceq—nq—\ is equal to the number of quarters the firm issued quarterly guidance from quarters q — n through q — 1. Generally, we expect firms that historically issued guidance to continue issuing guidance.

Next we consider Litigation, which is equal to one if the firm is subject to a securities lawsuit in quarters q — 1 or q, zero otherwise. Prior studies have shown that firms are more likely to issue guidance when their ex ante litigation risk is high (Skinner [1997], Brown, Hillegeist, and Lo [2005]). However, we expect current defendants in a pending Rule 10b-5 lawsuit to be less likely to issue forward-looking statements as they would want to avoid falling short of this guidance and providing additional ammunition for the prosecution (Rogers and Van Buskirk [2009]). Restate is equal to one if the firm announces a restatement during quarters q — 1 or q, zero otherwise. We include this variable as restatements might affect a manager's ability to form, or desire to provide, guidance. Restructuring is equal to one if the firm reports restructuring charges in quarter q — 1, zero otherwise. Restructurings could impede the formation of guidance because of increased uncertainty around these events.

EPSVolat is the standard deviation of quarterly earnings per share over quarters q — 8 to q — 1. We include this variable as firms with more volatile earnings tend to be less likely to issue guidance (Waymire [1985]). Return and Loss measure performance, as firms may be less likely to issue guidance when experiencing poor performance (Miller [2002]). Return is the cumulative size-adjusted return over fiscal quarter q — 1; Loss is the percentage of quarters in which the firm reported negative earnings from quarters q — 8 to q — 1.

FSE is the percentage of quarters in which the firm fell short of analyst expectations over the preceding four quarters. Drawing on the findings of Feng and Koch [2010], we expect a negative coefficient on FSE—firms that historically reported disappointing earnings are less likely to provide guidance. AnalystFollow is the natural logarithm of one plus the number of analysts following the firm in quarter q — 1, where firms with greater analyst following are expected to be more likely to issue guidance (Ajinkya, Bhojraj, and Sengupta [2005]). We include Size (the natural logarithm of total assets as of the end of quarter q — 1), as we expect larger firms to be more likely to issue guidance, and BooktoMarket (the ratio of a firm's book value of equity to its market value of equity as of the end of quarter q — 1), as we expect growth firms to be more likely to issue guidance, since their growth prospects may necessitate guidance to aid the market's formation of earnings expectations. Also, growth firms are under greater pressure to avoid reporting disappointing earnings (Skinner and Sloan [2002]), which suggests that when analysts are overly optimistic, managers need to guide the market towards beatable expectations. Finally, because disclosure policies differ across industries (e.g., Anilowski, Feng, and Skinner [2007]), we include the percentage of firms in the firm's industry (two-digit SIC) that issued guidance in quarter q (IndProp).

3.3 DESCRIPTIVE STATISTICS

Table 1 reports descriptive statistics for the main variables in the analysis. The first, second, and third sets of columns present descriptive statistics

TABLE 1

Descriptive Statistics

Variable Sample of Frequent Guiders (to = 7,660) Sample of Infrequent Guiders (to = 12,259) Sample of Nonguiders (to= 18,271)

Mean Standard deviation Median Mean Standard deviation Median Mean Standard deviation Median

CEOTurnovejq 0.023 0.151 0.000 0.023 0.151 0.000 0.021 0.143 0.000

CFOTumoveTq 0.030 0.170 0.000 0.033 0.180 0.000 0.028 0.164 0.000

NextGuicUinceq+1,^+8 2.290 2.488 1.000 4.798 3.116 5.000 7.118 2.070 8.000

Guidunceq 0.726 0.446 1.000 0.250 0.433 0.000 0.042 0.200 0.000

Guidanceqj^ 1 0.703 0.457 1.000 0.276 0.447 0.000 0.054 0.226 0.000

Guidanceq+ i,q+2 0.799 0.401 1.000 0.382 0.486 0.000 0.089 0.284 0.000

Guidanceq+ \ f, \ 0.851 0.356 1.000 0.496 0.500 0.000 0.137 0.344 0.000

Guidanceq+1, ^+8 0.913 0.282 1.000 0.609 0.488 1.000 0.203 0.402 0.000

NbGuid^inCCq-S, q-1 5.973 1.377 6.000 1.858 0.962 2.000 0.000 0.000 0.000

ResumePrecisionq+ | 1.901 1.272 1.000 n.a. n.a. n.a. n.a. n.a. n.a.

A Horizon [ 4j i j _ [ ?+1 _ ?+4] 3.516 26.023 0.917 n.a. n.a. n.a. n.a. n.a. n.a.

Litigatioriq-\q 0.025 0.158 0.000 0.030 0.170 0.000 0.019 0.137 0.000

Restatßq-\q 0.036 0.185 0.000 0.036 0.186 0.000 0.021 0.144 0.000

Restructuringq-\ 0.325 0.468 0.000 0.278 0.448 0.000 0.184 0.388 0.000

EPSVolatq-8t q-1 0.326 2.261 0.147 0.367 0.876 0.164 1.024 16.826 0.199

Returriq-1 0.011 0.186 0.003 0.025 0.237 0.014 0.034 0.203 0.026

LoSSq-S,q-l 0.146 0.264 0.000 0.205 0.311 0.000 0.196 0.294 0.000

FSEq-i,q-l 0.198 0.229 0.250 0.256 0.266 0.250 0.259 0.275 0.250

AnalystFolloWq-1 2.196 0.764 2.303 2.010 0.801 2.079 1.863 0.974 2.079

IndPrOpq 0.280 0.124 0.265 0.240 0.111 0.236 0.207 0.108 0.203

Sizßq-1 7.638 1.509 7.539 7.554 1.728 7.337 7.533 1.910 7.360

BooktoMarketq-\ 0.466 0.333 0.401 0.542 0.464 0.451 -0.430 90.485 0.470

o o ffi M H

£ ¡a

See appendix A for variable definitions.

for the samples of frequent guiders, infrequent guiders, and nonguiders, respectively. Across all three samples, about 2% (3%) of the firm-quarter observations experience a CEO (CFO) turnover. Among frequent guiders, approximately 80% (85%) of sample observations issue earnings guidance in the next two (four) quarters. These figures are substantially lower in the infrequent guider and nonguider samples (by construction) with 38% (50%) of observations issuing guidance in the next two (four) quarters among infrequent guiders and only 9% (14%) among nonguiders. These results highlight the importance of holding the preexisting disclosure policy constant in order to isolate any manager-specific effects.

Turning to the firm- and industry-specific determinants of guidance issuance, in the sample of frequent guiders, approximately 3% (4%) of the sample observations experience litigation (restatements), approximately 33% experience a restructuring, and, on average, losses are recognized in close to 15% of the preceding eight quarters. Most firms meet analyst expectations, with only about 20% of firm quarters among frequent guiders historically falling short of expectations—this is consistent with prior research examining firms issuing guidance (e.g., Houston, Lev, and Tucker [2009]).

In table 2, we provide univariate statistics on guidance issuance by turnover (CEO, CFO, or no turnover) and by the background of the incoming executive (internal or external hire, and, among external hires, prior forecasting experience and industry affiliation; see appendix A). Referring first to the column of frequent guiders, we see that firms that provide guidance regularly and are not experiencing a turnover tend to issue guidance in the next two quarters (NextGuidance is 2.262). Those with a CEO (CFO) turnover tend to issue guidance a little later, with NextGuidance of 3.093 (2.643), both of which are statistically different from nonturnover firms. Among both CEO and CFO turnovers, NextGuidance is not statistically different between external hires and internal hires, however, among external hires, those with prior experience issue guidance more quickly than those without prior experience.

Among infrequent guiders (nonguiders), firms without a top executive turnover issue guidance in the next five (seven) quarters, on average. NextGuidance is not statistically different for firms experiencing a CEO or CFO turnover for infrequent guiders or nonguiders. Among infrequent guiders, externally hired CFOs with no prior experience are slower to issue guidance than those with prior experience, and among nonguiders, externally hired CEOs from a different industry are slower to issue guidance than those from the same industry. Overall, we do find univariate evidence that turnovers affect the frequency of guidance, and that this varies differentially with the executive's background.

In figure 1, we graph the frequency of guidance partitioned by frequent, infrequent, and nonguiders as well as by turnover (CEO, CFO, or no turnover) to visually illustrate the economic significance of the univari-ate statistics in table 2. Similar to table 2, among infrequent guiders and

TABLE 2

Univariate Comparisons of Guidance Issuance (NextGuidanceq+^q+g) and Top Executive Turnover

Frequent Guiders

Infrequent Guiders

Nonguiders

No turnover 2.262 4.791 7.119

CEO turnover Internal External Prior experience No prior experience Same industry Different industry 3.093a 3.1 94a 2.813 1.286a 3.073cc 2.600 3.167a 5.136 5.017 5.324 5.143 5.337 5.674 5.103 7.102 7.139 7.036 6.545 7.129 6.761 7.329d

CFO turnover Internal External Prior experience No prior experience Same industry Different industry 2.643a 2.551 2.741a 1.767 2.975aa'cc 2.457 2.733a 5.010 5.063 4.963 4.000 5.156cc 5.370 4.754 7.164 7.144 7.188 6.962 7.217 7.303 7.131

O o ffi M H

£ ¡a

See appendix A for variable definitions.

^^Statistically different from No Turnover at p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively.

B,bb,bStatistically different from Internal at p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively.

C,cc'Statistically different from Prior Experience at p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively.

D,dd'Statistically different from Same Industry at p < 0.01, p < 0.05, and^fr < 0.10 (two-tailed), respectively.

q-4 q-J q-2 q- 1 q q+J q + 2

Fig. 1.—Guidance Issuance

nonguiders the impact of turnovers appears to be economically weak. Although there is evidence of mean reversion, there is no clear visual effect of turnovers. Among frequent guiders, while there is an overall reduction in guidance frequency, consistent with mean reversion (recall that to be in this sample firms must have issued guidance in at least four of the last eight quarters and two of the last four quarters), it is clear that this reduction is sharper among both CEO and CFO turnover firms. The effect appears to be economically stronger and the reduction appears to persist to a greater extent for CEO turnovers relative to CFO turnovers; guidance frequency among CFO turnovers converges with that of the nonturnover sample by quarter q + 4.

4. Regression Analysis

4.1 TEST OF HYPOTHESIS: TOP EXECUTIVE TURNOVERS AND GUIDANCE ISSUANCE

Our null hypothesis states that there is no association between top executive turnovers and the issuance of earnings guidance. We test this hypothesis using the following duration analysis:

NextGuidance = f (CEOTurnover, CFOTurnover, Guidance,

NbGuidance, Litigation, Restate, RRestructuring, EPSVolat, Return, Loss, FSE, AnalystFollow, IndProp, Size, BooktoMarket, YearFixedEffects).

We conduct the duration analysis using a semiparametric, discrete time Cox proportional hazard model where the dependent variable NextGuidanceq+1q+8 is the number of quarters, from one to eight, until the firm issues guidance after quarter q (i.e., a value of one [two] indicates that the firm issues guidance in quarter q + 1 [q + 2] for the first time after quarter q).8 The general form of the model is as follows:

h(NextGuidanceq+8) = ho(NextGuidanceq+8)exp(/3 X), (2)

where ho is the baseline hazard function, /3 is a vector of unknown regression estimates and X is a vector of observable covariates, as identified in equation (1). The model is said to be semiparametric because the baseline hazard function is unknown (hence nonparametric), but the functional form of the covariates' effects is specified (hence parametric). The baseline hazard function ho(n) is the common probability that firms—holding all covariates equal to zero—will issue guidance n quarters after quarter q.

We are interested in the coefficients on CEOTurnover and CFOTurnover. A negative coefficient on either variable would indicate that firms experiencing an executive turnover are slower to issue guidance in the next eight quarters, relative to nonturnover firms. We include each of the firm-and industry-specific determinants of guidance issuance introduced in section 3.2.2 and defined in appendix A. We also include calendar-year fixed effects to capture market-level time trends in guidance issuance. Since our sample consists of repeated observations of the same firms, we assess the significance of our regression coefficients using standard errors clustered by firm (Petersen [2009]).

For completeness, we consider all three samples in table 3, but will focus on the frequent guiders and nonguiders in subsequent analyses as these two samples have a more discernable preexisting guidance policy. The first column of results presents frequent guiders, the second infrequent guiders, and the final nonguiders. Both CEO and CFO turnovers are negatively associated with future guidance issuance among frequent guiders; although the coefficient on CEO turnover is almost double that of the coefficient on CFO turnover (-0.279 vs. -0.145), the coefficients are not statistically different from one another under an F-test. Among infrequent guiders, CFO turnovers exhibit a marginally significant negative association with future guidance issuance, although again the F-test indicates that the coefficient is not statistically different from the coefficient on CEO turnovers. Finally, among nonguiders, neither turnover event is associated with future guidance issuance when we consider the aggregate variable NextGuidanceq+1,q+8. In economic terms, frequent guiders experiencing a CEO (CFO) turnover

8 We do not consider guidance issued beyond eight quarters after quarter q. Thus, our hazard model has an adjustment for right censoring at the end of quarter q + 8. Other papers using a Cox proportional hazard model in accounting studies include Beatty, Ke, and Petroni [2002] for earnings management, O'Brien, McNichols, and Weilin [2005] for analyst recommendations, and Tse and Tucker [2010] for earnings warnings.

TABLE 3

Hazard Model Analysis of Guidance Issuance and Top Executive Turnover

Dependent Variable: NextGuidanceq+1,q+8 Frequent Guiders Infrequent Guiders Nonguiders

Coefficient Hazard Coefficient Hazard Coefficient Hazard

p-value Ratios p-value Ratios p-value Ratios

CEOTurnoverq —0.279 0.76 0.015 1.02 0.163 1.18

0.01 0.86 0.18

CFOTurnoverq —0.145 0.87 —0.125 0.88 —0.048 0.95

0.07 0.09 0.66

Guidanceq 0.687 1.99 0.963 2.62 1.377 3.96

0.01 0.01 0.01

NbGuidanceq —8, q— 1 0.106 1.11 0.210 1.23 n.a. n.a.

0.01 0.01 n.a.

Litigationq -1q —0.210 0.81 —0.190 0.83 0.033 1.03

0.02 0.02 0.80

Restateq — \q 0.011 1.01 0.028 1.03 0.143 1.15

0.89 0.68 0.28

Restructuringq — 1 —0.008 0.99 —0.025 0.98 0.359 1.43

0.79 0.41 0.01

EPSVolatq— 8,q—1 —0.091 0.91 —0.085 0.92 —0.135 0.87

0.01 0.01 0.01

Returnq _1 0.229 1.26 0.206 1.23 0.153 1.17

0.01 0.01 0.08

Lossq-8,q-1 —0.184 0.83 —0.179 0.84 —0.188 0.83

0.01 0.01 0.03

FSEq—4, q—1 —0.255 0.78 —0.415 0.66 —0.909 0.40

0.01 0.01 0.01

AnalystFollowq—1 0.015 1.02 0.239 1.27 0.514 1.67

0.56 0.01 0.01

IndPropq 0.488 1.63 0.641 1.90 2.051 7.78

0.01 0.01 0.01

Sizeq — 1 0.054 1.06 —0.111 0.90 —0.177 0.84

0.01 0.01 0.01

BooktoMarketq — 1 —0.229 0.80 —0.201 0.82 —0.069 0.93

0.01 0.01 0.08

Year fixed effects Included Included Included

Number of 7,232 11,294 14,782

observations

Number of CEO/ 171/214 273/375 272/394

CFO turnovers

The duration of time in the hazard model is the time until the firm issues quarterly earnings guidance after quarter q, with adjustment for right censoring at the end of an eight-quarter period. Bolded coefficients and p-values are statistically significant (two-tailed p-values < 0.10); p-values are based on standard errors that have been clustered by firm and are presented in italics below the coefficients. NextGuidanceq+1,q+8 is the number of quarters, from one to eight, until the firm issues quarterly earnings guidance after quarter q. CEOTurnoverq (CFOTurnoverq) is an indicator variable equal to one if there is a change in the CEO (CFO) during quarter q, zero otherwise. See appendix A for additional variable definitions.

A'aa'aStatisticallydifferent from CEOTurnover atp < 0.01,p < 0.05, andp < 0.10 (two-tailed), respectively.

are 24% (13%) less likely to issue guidance over the next eight quarters than frequent guiders not experiencing a turnover.

Turning next to the firm- and industry-specific determinants of guidance issuance, both measures of historical guidance are positive and significant,

consistent with guidance being a relatively "sticky" disclosure choice. Litigation is negatively associated with guidance among frequent and infrequent guiders, consistent with Rogers and Van Buskirk [2009], while Restate is not a significant explanatory variable. It could be that litigation firms want to avoid providing potential "fuel" for the litigators (in the event they miss their own guidance; Rogers and Van Buskirk [2009]), while restaters might be attempting to restore faith in their abilities (Farber [2005]) and thus are reluctant to stop providing guidance. Restructuring is generally insignificant, but positive and significant among nonguiders, inconsistent with our expectations. Earnings volatility (EPSVolat) is negative and significant, consistent with Waymire [1985]. The coefficients on both Return and Loss support our conjecture that poorly performing firms are less likely to provide guidance (Miller [2002]), and the coefficient on FSE is consistent with the findings of Feng and Koch [2010]; firms are less likely to provide guidance if they fell short of analysts' expectations in the past. AnalystFollow is positive and significant among infrequent guiders and nonguiders, as expected, but is not significant among frequent guiders. The proportion of firms in the industry that provide guidance (IndProp) is positive and significant across all three samples. Size is positive and significant among frequent guiders, as expected, but negative and significant among infrequent guiders and nonguiders, inconsistent with our expectations; recall, however, that several of our variables are correlated with size (e.g., analyst following, guidance history). Finally, BooktoMarket is negative and significant as predicted.

4.2 ENDOGENOUS NATURE OF TURNOVERS

Executive turnovers are correlated with other shocks to the firm such as poor performance. Although we explicitly consider events such as litigation, poor performance, or falling short of prior analyst forecasts, which might be causing the changes in guidance (e.g., Rogers and Van Buskirk [2009], Feng and Koch [2010]), in this section we investigate, for both frequent guiders and nonguiders, the possibility that concurrent shocks to the firm lead to both the turnover and the change in guidance policy. For example, a firm-specific performance change may lead to both an executive turnover and a change in guidance.

Since Miller [2002] documents an association between performance and voluntary disclosure and we also expect performance variables to be correlated with executive turnover (Weisbach [1988], Parrino [1997]), we first partition the analysis by historical performance (based on the prior two-year size-adjusted stock return) and examine the association between turnovers and guidance within terciles of past performance. Results are presented in table 4, with frequent guiders and nonguiders in Panels A and B, respectively.9 Referring first to frequent guiders, we see that CEO turnovers

9 For brevity, we include but do not tabulate the firm- and industry-specific determinants. Complete tables are available from the authors upon request.

TABLE 4

Supplemental Hazard Model Analyses of Guidance Issuance

Dependent Variable: NextGuidanceq+iJq+s

Past Performance Terciles

1 (Low)

3 (High)

Coefficient Hazard Coefficient Hazard Coefficient Hazard p-value Ratios p-value Ratios p-value Ratios Panel A: Partitions of frequent guiders based on the prior two-year size-adjusted stock return

CEOTurnover,

CFOTurnover,

Additional determinants Year fixed effects Number of observations Number of CEO/CFO turnovers

-0.408 0.67 0.01

—0.029a 0.97 0.84

Included Included

2,410 86/76

—0.217 0.81

—0.064 0.94 0.63

Included Included

0.033 1.03 0.88 —0.406a 0.67 0.01

Included Included

Panel B: Partitions of nonguiders based on the prior two-year size-adjusted stock return

CEOTurnovr

CFOTurnover,

Additional determinants Year fixed effects Number of observations Number of CEO/CFO turnovers

0.084 0.69 0.108 0.59

0.651 0.01

0.027aa 0.90

—0.093 0.66 —0.238 0.23

Included Included

4,645 109/164

Included Included

5,056 87/129

Panel C: Concurrent change in the chairman of the board of directors

Frequent Guiders

Included Included

5,072 76/101

Nonguiders

Coefficient Hazard Coefficient Hazard

p-value Ratios p-value Ratios

CEOTurnover, —0.175 0.84 0.192 1.21

0.09 0.12

CFOTurnover, —0.145 0.87 —0.077 0.93

0.07 0.51

NewChairman, —0.278 0.76 —0.139a 0.87

0.01 0.28

Additional determinants Year fixed effects Number of observations Number of CEO/CFO turnovers Number of chairman of the board turnovers

Included Included

7,232 171/214 187

Included Included

14,782 272/394 341

( Continued)

TABLE 4 — Continued

Panel D: Hired-away executive turnovers among frequent guiders

Dependent Variable: NextGuidanceq+1,q+8

Coefficient Rvalue

Hazard Ratios

CEOTurnoverHiredAwayq CEOTurnoverNotHiredAwayq CFOTurnoverHiredAwayq CFOTurnoverNotHiredAwayq

Additional determinants Year fixed effects Number of observations Number of CEO/CFO hired away Number of CEO/CFO not hired away

0.525 0.28 -0.310b 0.01 -0.557 0.01 -0.030C 0.74

1.69 0.73 0.57 0.97

Included Included

7,232 7/39 164/175

The duration of time in the hazard models is the time until the firm issues guidance after quarter q, with adjustment for right censoring at the end of an eight-quarter period. Bolded coefficients and p-values are statistically significant (two-tailed p-values < 0.10); p-values are based on standard errors that have been clustered by firm and are presented in italics below the coefficients. NextGuidanceq+1,q+8 is the number of quarters, from one to eight, until the firm issues guidance after quarter q. CEOTurnoverq (CFOTurnoverq) is an indicator variable equal to one if there is a change in CEO (CFO) during quarter q, zero otherwise. The additional firm- and industry-specific determinants included are: Guidanceq, NbGuidanceq-8,q-1 (when applicable), Litigationq_1,q, Restateq_1,q, Restructuringq_ь EPSVolatq-8,q-1, Returnq_ь Lossq-8,q-1, FSEq-4,q-1, AnalystFollowq_ь IndPropq, Sizeq_ь and BooktoMarketq_1. In panel C, NewChairmanq is an indicator variable equal to one if there is a newly appointed chairman of the board of directors in quarter q, zero otherwise. See appendix A for additional variable definitions.

A,aa,aStatisticallydifferent from CEOTurnover atp < 0.01,p < 0.05, andp < 0.10 (two-tailed), respectively.

B,bb,Statistically different from CEOTurnoverHiredAway at p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively.

C,cc,cStatistically different from CFOTurnoverHiredAway at p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively.

are associated with breaks in guidance only among the lowest tercile of past performance, while CFO turnovers play a role only in the top tercile of past performance. Thus, there is some evidence that, among frequent guiders, CEO turnovers are affected by concurrent poor firm performance, but this does not extend to CFO turnovers. Referring next to nonguiders, we see that the significant association between turnover and guidance provision for CEOs is concentrated in the middle performance tercile. Thus, while poor performance is correlated with both CEO turnovers and breaks in guidance among frequent guiders, exceptionally strong performance does not similarly correlate with CEO turnovers and initiations of guidance among nonguiders.10

10 We also examine changes in earnings volatility and analyst forecast dispersion around executive turnovers, to control for changes in firm-specific uncertainty, and control for mergers and equity offerings, as managers might provide more guidance prior to these events (e.g., Lang and Lundholm [2000]). Finally, we consider the changes in the determinants (vs. levels) in each of our regression estimations. These alternative specifications do not change the inferences from our main results (not tabulated).

Second, to shed light on the validity offirm-initiated changes in guidance policy in response to changes in disclosure preferences, we examine concurrent board chairman turnovers (e.g., Ajinkya, Bhojraj, and Sengupta [2005], Karamanou and Vafeas [2005], Richardson, Tuna, and Wysocki [2003]). A new board chairman may stipulate a change in guidance policy, and this event may coincide with the turnover of a top executive. We identify a total of 187 (341) firm-quarter observations among our frequent guiders (nonguiders) with a change in chairman (from the BoardEx database, which compiles biographical information on corporate directors and senior executives of major U.S. and foreign companies). In panel C of table 4, we reestimate table 3 including NewChairman, an indicator variable that is equal to one if there is a newly appointed board chairman in quarter q, and zero otherwise. We do not tabulate the firm- and industry-specific determinants of guidance issuance, as the inferences are virtually identical to those reported in table 3. Among frequent guiders, there is a reduced likelihood of providing guidance following a board chairman turnover, while the impact of board chairman turnover is not significant among the nonguider sample. Turning to the coefficients on CEO and CFO turnover, we see that among frequent guiders, CFO turnover has the same coefficient and p-value as in table 3 after controlling for the effect of board chairman turnover; however, the coefficient on CEO turnover falls to -0.175 with a p-value of 0.09 (from -0.279 with a p-value of 0.01). Thus, among frequent guiders there is some evidence that the change in guidance policy following new CEO appointments is partially attributable to concurrent changes in the board chairman, while breaks in guidance following CFO turnovers are largely unaffected by this concurrent event.11

Third, to further disentangle manager-specific effects from concurrent shocks to the firm among frequent guiders, we examine the circumstances of the turnover to better infer the underlying cause of the joint decision to appoint a new manager and cease providing guidance. Specifically, we consider whether the outgoing executive is hired away by another firm— arguably a relatively exogenous turnover. Among frequent guiders, this should help us determine whether breaks in guidance following turnovers result from a lack of firm- or industry-specific knowledge on the part of the incoming executive. To identify whether an outgoing manager is hired away, we read the press release announcing the executive's departure and note when the press release mentions the executive's appointment in a new

11 The correlation between CEO (CFO) turnover and chairman turnover is 0.40 (0.01) among frequent guiders. Of the 75 firm quarters with a contemporaneous change in both roles, 59 of the incoming CEOs and chairmen are the same individuals. Clearly, in these instances we cannot separate the CEO from general firm policy. We reestimate the hazard model using three partitions: new chairman who is not the new CEO, new chairman who is the new CEO, and new CEO who is not the new chairman. Both new CEO indicator variables are weakly associated with breaks in guidance (p < 0.10), suggesting that the CEO has an effect on guidance in both instances (not tabulated).

firm (examples are provided in appendix B). Finding that the previously documented associations hold among turnovers where the outgoing executive is hired away by another firm mitigates concerns of correlated omitted variables such as performance or other shocks. We find 7 (39) instances where outgoing CEOs (CFOs) are hired away from frequent guider firms. Results are presented in panel D of table 4; we again include, but do not tabulate, the additional determinants. Among outgoing CEOs who are hired away, there is no association between CEO turnover and breaks in guidance; however, because there are only seven instances where this variable is equal to one, the power of the test is low. Among CFOs, however, the breaks in guidance are present only among hired-away CFOs. These findings suggest that concurrent shocks to the firm are not the underlying driver of the association between CFO turnovers and breaks in guidance among frequent guiders.

4.3 LOGISTIC REGRESSION ANALYSIS

Although we find some evidence of breaks in guidance following CEO and CFO appointments in table 3, the results based on the hazard model make it difficult to assess the permanency of the breaks. To the extent that breaks following CFO turnovers are in response to uncertainty on the part of the CFO, we expect these breaks to be relatively temporary. Moreover, in the prior section we find some evidence that breaks following CEO turnovers may be firm initiated, in which case these breaks may be more permanent. To gain additional insight about the timing of any changes in the provision of guidance, we next partition the quarters and present results from four separate logistic regressions. We use the following logit model:

where y is a vector of coefficients and X is the same vector of explanatory variables as in equations (1) and (2). The first through fourth columns examine the association between CEO and CFO turnovers and guidance issued in quarter q + 1, quarters q + 1 to q + 2, quarters q + 1 to q + 4, and quarters q + 1 to q + 8, respectively. This allows for a better depiction of which specific quarters are associated with guidance issuance following the turnovers. Also, to better assess the economic significance of the results, mean marginal effects are reported along with p-values of the estimated coefficients.

We first consider frequent guiders in panel A of table 5. In quarter q + 1, CFO turnovers exhibit a significant and negative association with the provision of guidance, although the coefficients on CEO and CFO turnovers are not statistically different from one another under an F-test. Including quarter q + 2 results in an insignificant coefficient on CFO turnover, while the negative coefficient on CEO turnover becomes significant, although the two coefficients are again not statistically different under an F-test. Looking

TABLE 5

Logit Analysis of Guidance Issuance and CEO (CFO) Turnover

Dependent Variable: Guidanceq+i,q+ n

Quarter Quarters Quarters Quarters

q + 1 q + 1, q + 2 q + 1, q + 4 q + 1, q + 8

Panel A: Frequent guiders

CEOTurnover, -0.048 -0.061 -0.082 -0.065

0.10 0.01 0.01 0.01

CFOTurnover, -0.067 -0.036 -0.014aa -0.012aa

0.02 0.12 0.51 0.53

Guidance, 0.211 0.183 0.154 0.137

0.01 0.01 0.01 0.01

NbGuidanceq—8,q—1 0.039 0.023 0.010 0.018

0.01 0.01 0.09 0.01

NbGuidanceq—n,q—1 0.074 0.049 0.028 n.a.

0.01 0.01 0.01 n.a.

Litigationq — 1,q -0.044 -0.030 -0.045 -0.024

0.17 0.36 0.10 0.27

Restateq— 0.021 0.009 0.009 0.005

0.50 0.75 0.75 0.83

Restructuring q— 1 -0.003 -0.015 -0.010 0.007

0.83 0.30 0.44 0.47

EPSVolatq—8,q—1 -0.023 -0.018 -0.016 -0.010

0.22 0.15 0.12 0.16

Returnq—1 0.074 0.059 0.064 0.057

0.01 0.01 0.01 0.01

Loss,—8,q—1 -0.057 -0.054 -0.061 -0.058

0.06 0.06 0.02 0.01

FSEq—4,q—1 -0.090 -0.082 -0.051 -0.047

0.01 0.01 0.04 0.02

AnalysFolloWq—1 0.010 0.005 0.016 0.019

0.40 0.68 0.20 0.05

IndPropq 0.199 0.167 0.117 0.037

0.01 0.01 0.03 0 . 35

Sizeq — 1 0.012 0.012 0.005 -0.012

0.02 0.04 0.35 0.01

BooktoMarketq — 1 -0.073 -0.059 -0.042 -0.018

0.01 0. 01 0.02 0.17

Year fixed effects Included Included Included Included

Number of observations 7,226 7,210 7,127 6,799

Number of observations 5,092 5,770 6,080 6,206

with Guidanceq+1,q+n = 1

Number of CEO/CFO 170/214 169/213 168/209 155/198

turnovers

McFadden pseudo-R2 12.22% 14.97% 14.51% 22.09%

Panel B: Nonguiders

CEOTurnoverq -0.021 0.007 0.030 0.047

0.21 0.69 0.14 0.05

CFOTurnoverq -0.006 0.003 0.004 -0.019aa

0.64 0.84 0.82 0.32

Guidanceq 0.105 0.155 0.218 0.282

0.01 0.01 0.01 0.01

(Continued)

TABLE 5 — Continued

Dependent Variable: Guidanceq +i,q+n

Quarter Quarters Quarters Quarters

q + 1 q + 1, q + 2 q + 1, q + 4 q + 1, q + 8

Litigationq -\q -0.010 -0.004 0.008 0.023

0.49 0.84 0.76 0.45

Restateq — \q -0.009 -0.001 0.009 0.026

0.67 0.97 0.75 0.42

Restructuringq — 1 0.021 0.031 0.051 0.062

0.01 0.01 0.01 0.01

EPSVolatq— 8,q—1 -0.001 -0.003 -0.011 -0.026

0.65 0.39 0.17 0.04

Returnq — 1 -0.008 -0.003 0.005 0.031

0.43 0.77 0.70 0.06

Lossq—8, q— 1 -0.010 -0.014 -0.018 -0.026

0.30 0.33 0.42 0.40

FSEq—4,q— 1 -0.054 -0.085 -0.115 -0.149

0.01 0.01 0.01 0.01

AnalystFollowq—1 0.029 0.042 0.062 0.090

0.01 0.01 0.01 0.01

IndPropq 0.109 0.167 0.245 0.354

0.01 0.01 0.01 0.01

Sizeq — 1 -0.010 -0.014 -0.021 -0.031

0.01 0.01 0.01 0.01

BooktoMarketq — 1 -0.004 -0.006 -0.007 -0.009

0.35 0.33 0.39 0.51

Year fixed effects Included Included Included Included

Number of observations 14,769 14,730 14,522 13,770

Number of observations 910 1,487 2,255 3,139

with Guidanceq+1,q+n = 1

Number of CEO/CFO 272/394 272/394 267/389 250/366

turnovers

McFadden pseudo-R2 17.32% 17.58% 18.00% 16.91%

Mean marginal effects of the logit models are reported. p-values of the logit coefficients are based on standard errors that have been clustered by firm and are presented in italics below the marginal effects. Bolded marginal effects and p-values are statistically significant (two-tailed p-values < 0.10). Guidanceq+\iq+n is an indicator variable equal to one if the firm issues quarterly earnings guidance in at least one quarter from quarters q + 1 through q + n, zero otherwise. CEOTurnoverq (CFOTurnoverq) is an indicator variable equal to one if there is a change in CEO (CFO) during quarter q, zero otherwise. See appendix A for additional variable definitions.

A,aa,aStatisticallydifferent from CEOTurnover atp < 0.01,p < 0.05, andp < 0.10 (two-tailed), respectively.

out four and eight quarters, however, we see that only CEO turnovers continue to be negatively associated with the prevalence of earnings guidance, and the coefficients on CEO and CFO turnovers are statistically different (p < 0.05 for both horizons under an F-test). Thus, the association with CFO turnovers is temporary, lasting one to two quarters, while the association with CEO turnovers is more permanent, persisting through the next two years. In terms of economic significance, the marginal effects of CEO or CFO turnover on the likelihood of guidance issuance in the next few quarters are in the range of -5% to -8%, which is comparable to the effects of adverse economic events such as reporting a string of losses (about -6%) or negative earnings surprises (approximately -5% to -9%, depending

on the horizon), but falls below the economic significance of preexisting guidance policy and industry. For example, a firm that issues guidance in quarter q is 21% more likely to issue guidance in quarter q + 1, while a firm that belongs to an industry where firms have a history of guidance issuance is 20% more likely to issue guidance in quarter q + 1 than if it were in an industry where firms generally do not provide guidance.

We next consider nonguiders in panel B of table 5. In quarters q + 1 through q + 4, neither CEO nor CFO turnovers are associated with the provision of guidance, consistent with the hazard model results in table 3. Looking out eight quarters, however, we see that CEO turnovers are positively associated with the prevalence of earnings guidance, and the coefficients on CEO and CFO turnovers are statistically different (p < 0.05 under an F-test). Although, the marginal effect is economically significant at 5%, firms' preexisting guidance policies and industries are again the determinants with the greatest economic significance.

4.4 ANALYSIS OF NEWLY APPOINTED EXECUTIVES' BACKGROUNDS

In this section, we investigate the association between guidance and incoming managers' backgrounds to better link manager-specific effects on guidance. We begin by hand-collecting details about the incoming executive (e.g., if the newly appointed executive is an internal or external hire, is hired from the same industry, or has forecasting experience; see appendix A). We then examine how the breaks in guidance vary with incoming executive characteristics. If our findings are due to managerial uncertainty (i.e., a lack of knowledge about the new firm), we expect our results to be stronger when the incoming executives are external hires, especially when they are hired from a different industry or lack forecasting experience.

4.4.1. Frequent Guiders. In panel A of table 6, we report the results relating to guidance issued over the next one, two, four, and eight quarters for frequent guiders. As in table 5, mean marginal effects and p-values of the estimated coefficients are reported; we do not tabulate the firm- and industry-specific determinants of guidance issuance, as the inferences are virtually identical to those reported in table 5. In the first model, we distinguish between incoming executives who were internally promoted (74% and 52% of CEO and CFO turnovers, respectively) and those hired from another firm. Among CEOs, internal (external) appointments are associated with breaks in guidance two (four) to eight quarters following new appointments, and the coefficients on internal and external hires are not statistically different in any of the four windows. Among CFOs, external appointments are associated with breaks in guidance in the quarter following new appointments, but the effect is not statistically different from that of internal hires.

The remainder of our partitions focus on the characteristics of the 47 (110) externally hired CEOs (CFOs). In the second model, our cross-sectional variable is prior forecasting experience. We identify 6 (30) externally hired CEOs (CFOs) that have prior forecasting experience. As

TABLE 6

Logit Analysis of Guidance Issuance and Incoming Executive Background

Dependent Variable: Guidanceq+i q+n

Model I

Model II

Model III

Quarters Quarters Quarters Quarters Quarters Quarters Quarters Quarters Quarters

Quarter 7 + 1, 7+1, 7+1, Quarter 7+1, 7 + 1, 7+1, Quarter 7 + 1, 7+1, 7+1,

7 + 1 7+2 7 + 4 7+8 7+1 7+2 7 + 4 7 + 8 7+ 1 7+2 7 + 4 7+8

Panel A: Frequent guiders

InternalCEOq -0.038 -0.059 -0.085 -0.065 -0.039 -0.060 -0.085 -0.065 -0.038 -0.059 -0.085 -0.065

0.25 0.03 0.01 0.01 0.24 0.03 0.01 0.01 0.24 0.03 0.01 0.01

ExternalCEOq -0.079 0.19 -0.067 0.20 -0.070 0.09 -0.064 0.03

InternalCFOq -0.027 -0.019 -0.030 -0.034 -0.027 -0.019 -0.030 -0.033 -0.027 -0.020 -0.031 -0.034

0.51 0.58 0.32 0.18 0.51 0.58 0.33 0.19 0.50 0.56 0.31 0.18

ExternalCEOq -0.107 0.01 -0.053 0.11 0.002 0.94 0.013 0.62

PriorExperienceCEOq 0.135 0.53 0.014 0.94 4.084A 0.01 0.736A 0.01

NoPriorExperienceCEOq -0.110 0.09 -0.079 0.14 -0.095° 0.02 -0.080° 0.01

PriorExperienceCFOq -0.014 0.86 0.072 0.40 8.188B 0.01 1.508B 0.01

O o ffi M H

£ ¡a

(Continued)

TABLE 6—

Continued

Dependent Variable: Guidanceq+\q+n

Model I Model II Model III

Quarter 7+1 Quarters Quarters 7+1, 7+1, 7+2 7+4 Quarters 7 + 1, 7+8 Quarter 7+ 1 Quarters Quarters 7 + 1, 7 + 1, 7+2 7+4 Quarters 7+1, 7 + 8 Quarter 7 + 1 Quarters 7 + 1, 7 + 2 Quarters 7+1, 7 + 4 Quarter; 7 + 1, 7 + 8

NoPriorExperienceCFOq —0.139b 0.01 —0.084d —0.035d 0.03 0.30 —0.008° 0.79

SamelndusCEOq -0.027 0.73 -0.021 0.74 -0.038 0.45 -0.061 0.09

DifflndusCEOq -0.171 0.09 -0.137 0.13 -0.117 0.10 -0.068 0.16

SamelndusCFOq -0.057 0.42 -0.056 0.31 -0.021 0.66 -0.013 0.76

DifflndusCFOq -0.131b 0.01 -0.051 0.22 0.015 0.73 0.026 0.44

Additional determinants Included Included Included

Year fixed effects Included Included Included

Number of observations 7,226 7,210 7,127 6,799 7,226 7,210 7,127 6,799 7,226 7,210 7,127 6,799

Number of observations 5,092 5,770 6,080 6,206 5,092 5,770 6,080 6,206 5,092 5,770 6,080 6,206

with Guidanceq+ 2,q+n — 1 McFadden pseudo-R2 12.25% 14.98% 14.52% 22.14% 12.30% 15.03% 14.59% 22.20% 12.28% 14.99% 14.54% 22.15%

•n •n

(Continued)

TABLE 6 — Continued

Dependent Variable: Guidanceq+i q+n

Model I

Model II

Model III

Quarters Quarters Quarters Quarters Quarters Quarters Quarters Quarters Quarters

Quarter 7+1, 7+1, 7 + 1, Quarter 7+1, 7 + 1, 7+1, Quarter 7+1, 7+1, 7 + 1,

7 + 1 7 + 2 7 + 4 7 + 8 7+1 7 + 2 7 + 4 7+8 7+1 7 + 2 7 + 4 7+8

Panel B: Nonguiders

InternalCEOq -0.041 -0.018 0.016 0.014 -0.041 -0.018 0.016 0.014 -0.041 -0.018 0.017 0.014

0.07 0.42 0.50 0.62 0.07 0.42 0.50 0.62 0.07 0.42 0.50 0.62

ExternalCEOq 0.016a 0.55 0.059aa 0.03 0.059 0.07 0.109aa 0.01

InternalCFOq 0.003 0.000 0.002 -0.006 0.003 0.000 0.002 -0.006 0.003 0.000 0.002 -0.006

0.86 0.99 0.92 0.81 0.87 1.00 0.93 0.79 0.86 0.99 0.92 0.81

ExternalCEOq -0.021 0.37 0.006 0.79 0.006 0.82 -0.036 0.25

PriorExperienceCEOq 0.072aa 0.08 0.129a 0.01 0.114 0.06 0.214A 0.01

NoPriorExperienceCEOq -0.002 0.94 0.036c 0.28 0.043 0.26 0.084 0.04

PriorExperienceCFOq 0.007 0.87 -0.001 0.98 -0.030 0.68 -0.068 0.39

O o ffi M H

£ ¡a

(Continued)

TABLE 6—

Continued

Dependent Variable: Guidanceq+i^q+n

Model I Model II Model III

Quarter 7+1 Quarters Quarters 7+1, 7+1, 7+2 7+4 Quarters 7+1, 7 + 8 Quarter 7 + 1 Quarters Quarters 7+1, 7+1, 7+2 7+4 Quarters 7+1, 7+8 Quarter 7+ 1 Quarters 7+1, 7 + 2 Quarters 7+1, 7 + 4 Quarters 7 + 1, 7+8

NoPriorExperienceCFOq -0.029 0.31 0.008 0.013 0.77 0.66 -0.029 0.40

SamelndusCEO q 0.019 0.61 0.088a 0.01 0.074 0.09 0.110a 0.03

DifflndusCEOq 0.011 0.76 0.004 0.92 0.037 0.42 0.108a 0.03

SamelndusCFOq 0.009 0.77 0.004 0.91 -0.036 0.48 -0.091 0.12

DifflndusCFOq -0.045 0.20 0.007 0.81 0.024 0.42 -0.012 0.74

Additional determinants Included Included Included

Year fixed effects Included Included Included

Number of observations 14,769 14,730 14,522 13,770 14,769 14,730 14,522 13,770 14,769 14,730 14,522 13,770

Number of observations 910 1,487 2,255 3,139 910 1,487 2,255 3,139 910 1,487 2,255 3,139

with Guidanceq+\ q+n = 1 McFadden pseudo-R2 17.38% 17.63% 18.01% 16.94% 17.41% 17.66% 18.02% 16.95% 17.40% 17.65% 18.02% 16.95%

Mean marginal effects of the logit models are reported. Rvalues of the logit coefficients are based on standard errors that have been clustered by firm and are presented in italics below the marginal effects. Bolded marginal effects and Rvalues are statistically significant (two-tailed Rvalues < 0.10). Guidance^is an indicator variable equal to one if the firm issues quarterly earnings guidance in at least one quarter from quarters q + 1 through q + n, zero otherwise. See appendix A for additional variable definitions. ^^Statistically different from InternalCEO at p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively. B'bb'bStatistically different from InternalCFO at p < 0.01, p < 0.05, and jfr < 0.10 (two-tailed), respectively.

C,cc'Statistically different from PriorExperienceCEO at p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively.

D,dd'Statistically different from PriorExperienceCFO at p < 0.01, p < 0.05, and jfr < 0.10 (two-tailed), respectively.

E,ee,eStatistically different from SamelndusCEO at p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively.

F,ff'f Statistic ally different from SamelndusCFO 21 p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively.

•n •n

in model I, internal appointments are often associated with breaks in guidance following new CEO appointments and are not associated with breaks in guidance following new CFO appointments.

Among external hires, the background of the CEO appears to matter. The coefficient on PriorExperienceCEO is not significant one or two quarters ahead, and is positive and significant four and eight quarters out (i.e., these CEOs increase the propensity to guide, consistent with the taste effects documented by Bamber, Jiang, and Wang [2010]). The coefficient on NoPriorExperienceCEO is significantly negative across three of the four horizons (all except two quarters out). In terms of statistical differences across coefficients, CEOs with forecasting experience are significantly more likely to provide guidance than CEOs without forecasting experience four and eight quarters following new appointments (p < 0.01 under an F-test).

Among externally hired CFOs, the prior experience of the CFO also appears to play a role. Similar to PriorExperienceCEO, PriorExperienceCFO is not significant one or two quarters out, and is again positive four and eight quarters out. CFOs with prior forecasting experience are significantly more likely to issue guidance than CFOs without experience two, four, and eight quarters ahead (p-value <0.10, 0.01, and 0.01 for two, four, and eight quarters ahead, respectively). The coefficient on NoPriorExperienceCFO is significantly negative in the first two quarters, and becomes insignificant after four quarters. In the quarter following the new appointment, newly appointed CFOs with no prior forecasting experience are less likely to issue guidance than internal hires (p < 0.10 under an F-test), but this difference is not significant subsequently. Thus, there is some evidence that breaks in guidance following CFO turnovers are more transitory than those following CEO turnovers, and the temporary breaks are associated with inexperienced hires, although it is also possible that firms wishing to implement a change in guidance policy hire executives with specific backgrounds.

In our third model, we partition the externally hired executives based on the industry backgrounds of the newly appointed executives. Of the external hires, 64% (32%) of the CEOs (CFOs) are hired from within the same industry. If the firm-specific expertise of the incoming executives affects their decision to issue guidance, we expect executives hired from within the same industry to be more likely to issue guidance (as they will be more knowledgeable about the industry and able to immerse themselves in the details of their new firm more quickly), and thus less likely to be associated with breaks in guidance. Referring to the final set of columns in panel A of table 6, the coefficients on DiffIndusCEO and DiffIndusCFO are both negative and significant one quarter ahead, and insignificant thereafter, while SameIndusCEO is negative and significant eight quarters out. The difference in the coefficients between same and different industries, however, is not statistically significant under an F-test in any of the quarters.

Overall, the results across the alternative partitions weakly support a lack of expertise on the part of the incoming CFOs playing a role in the break in guidance, while we find no support for this among CEOs. The marginal

effect of CFO turnover on the likelihood of guidance issuance in the next quarter is -13.9% for incoming CFOs with no prior forecasting experience, which is twice as large as the average marginal effect ofCFO turnovers among frequent guiders documented in table 5.

4.4.2. Nonguiders. In panel B of table 6, we report the results relating to guidance issued over the next one, two, four, and eight quarters for nonguiders.12 As with frequent guiders in panel A, in the first model we distinguish between incoming executives who were internally promoted (64% and 55% of CEO and CFO turnovers, respectively) and those hired from another firm. We find that firms with externally hired CEOs are significantly more likely than nonturnover firms to initiate guidance in the two, four, and eight quarters following the appointment. Furthermore, the coefficient on external CEO hires is significantly greater than the one on internal CEO hires for two and eight quarters ahead (/»-value < 0.05). Among CFOs, however, we do not find an association between guidance provision and turnovers among internal or external hires.

We next partition on forecasting experience in model II. We identify 22 (26) externally hired CEOs (CFOs) that have prior forecasting experience (16% and 11% of all externally hired CEOs and CFOs, respectively). Conditional on being an external hire, newly appointed CEOs with prior forecasting experience are more likely to initiate guidance than internal hires one, two, and eight quarters following the turnover, and are more likely to initiate guidance than external hires without prior forecasting experience in the two quarters following the turnover. Again, CFO turnovers are not associated with the provision of guidance for either subset of external hires (those with or without forecasting experience).

In our third model, we partition the externally hired executives based on the industry backgrounds of the newly appointed executives. Of the external hires, 51% (33%) of the CEOs (CFOs) are hired from within the same industry. The final set of columns in panel B of table 6 shows that, among externally hired CEOs, those coming from the same industry are positively associated with the initiation of guidance in the two to eight quarters following their appointment. Under an F-test, this effect is not statistically different from newly appointed CEOs hired from a different industry, but is statistically different from internally promoted CEOs two and eight quarters following the turnover. As in prior tests of nonguiders, there is no difference between externally hired CFOs based on their background (in this case, industry expertise).

Overall, the results across the alternative partitions of nonguiders generally provide evidence that newly appointed externally hired CEOs are often

12 We also collect the newly appointed executive backgrounds for infrequent guiders— those with a less discernable preexisting guidance policy. We find no significant differences among CEOs based on their background, and find that newly appointed CFOs with no prior forecasting experience are less likely to issue guidance than newly appointed CFOs with prior forecasting experience (not tabulated).

associated with an initiation of guidance, especially when they have prior forecasting experience (7.2-21.4%).

4.5 GUIDANCE PRECISION

Thus far, our findings suggest that CEOs participate more in firm-level policy decisions, while CFOs are more involved in the formation and discussion of guidance. As a final corroboration of these findings, we examine, among frequent guiders, how quickly the firm resumes its prior level of guidance precision, as King, Pownall, and Waymire [1990] argue that guidance precision reflects the accuracy of management's private information about earnings. We consider point forecasts to be the most precise, and qualitative forecasts the least precise (Pownall, Wasley, and Waymire [1993]; Baginski and Hassell [1997]). Specifically, we code the precision of guidance as follows: three for a point estimate, two for a range estimate, one for an open-ended interval (minimum or maximum), and zero for qualitative guidance.13 ResumePrecisionq+i^+4 is the number of quarters, from one to four, until the firm issues guidance after quarter q that is at least as precise as the average precision of guidance issued in quarters q - n to q - 1; if the firm has not resumed its prior level of precision by quarter q + 4, we set ResumePrecision equal to four (i.e., the variable is right censored).

If the breaks are related to managerial uncertainty or lack of expertise, then we expect new managers opting to provide guidance to issue less-precise guidance to compensate for their lack of knowledge about their new firm. Alternatively, if the breaks are not related to managerial uncertainty or lack of expertise, we would not expect to see less-precise guidance following turnovers, on average. To investigate this, we estimate the following model among frequent guiders:

ResumePrecisionq+1q+4 = f (CEOTurnover, CFOTurnover, Precision, PrecisionVolat, Horizon, Litigation, Restate, Restructuring, EPSVolat, Return, Loss, FSE, (4) AnalystFollow, IndProp, Size, BooktoMarket, YearFixedEffects).

As with guidance issuance, we estimate a Cox proportional hazard model. To the extent that manager-specific uncertainty plays a role in the determination of the precision of guidance, we expect a negative coefficient on CEO/ CFOTurnover.

13 To the extent that First Call is more likely to miss qualitative guidance (e.g., Anilowski, Feng, and Skinner [2007], Chuk, Matsumoto, and Miller [2009]), and new executives issue more qualitative guidance, this might suggest that they have stopped guidance when in fact they have provided less-precise guidance. This should bias against finding lower guidance precision for these firms. Nonetheless, we conduct a similar validity check to the one described in footnote 7 and find no evidence that the precision of the forecasts we hand-collect from Factiva are significantly lower than the precision of First Call forecasts for quarter q + 1 (a result that holds separately for turnover and nonturnover observations; not tabulated).

Providing less-precise guidance is an alternative to the more extreme decision not to issue guidance, but it has similar determinants (Ajinkya, Bho-jraj, and Sengupta [2005]); thus we include the same firm- and industry-specific determinants as those in equation (1). We replace Guidance and NbGuidance with historical precision (Precision) and the volatility of historical precision (PrecisionVolat) to capture firm-specific challenges affecting precision that are unrelated to the turnover. We also include Horizon, the average number of days between the issuance of earnings guidance and the fiscal quarter end, as we expect the guidance to be less precise if it is issued earlier in the quarter (Baginski and Hassell [1997]).14

The results are presented in table 7. We find that, among frequent guiders, CEO turnovers are not associated with reduced precision, but CFO turnovers are. The difference between CEO and CFO turnovers is statistically significant, with a p-value of less than 0.01. Specifically, firms with CFO turnovers are 22% less likely than firms without CFO turnovers to reach the firm's prior level of precision in the next year.

We also examine, but do not tabulate, the association between changes in precision and the characteristics of the newly appointed executives' backgrounds. Among CEOs, we find no difference in precision across internal versus external hires. Among CFOs, however, we find that reductions in precision are concentrated among external hires, and the difference between internal and external hires is statistically significant, with a p-value of less than 0.10. We next condition on the externally hired executives' forecasting experience. We find that CEO turnovers are not associated with changes in precision, regardless of CEOs' forecasting experience, while CFOs tend to reduce the precision of their guidance (relative to nonturnover firms) when the newly appointed CFO has no forecasting experience, although the difference between CFOs with and without forecasting experience is not statistically significant under an F-test. Finally, regardless of whether newly appointed CEOs are hired from the same or a different industry, there is no association between CEO turnover and changes in precision. There is evidence of reduced precision for all externally hired CFOs, however, regardless of whether they are hired from within or outside of the industry.

Overall, we find no evidence that newly appointed CEOs provide relatively imprecise guidance. Newly appointed CFOs, however, tend to provide less-precise guidance when we expect them to be experiencing the greatest uncertainty: when they are hired from another firm.

14 Extending the tests until quarter q + 8 does not affect our results (not tabulated). Note that since we include Horizon, we effectively exclude observations for which there is no guidance issued from quarters q + 1 to q + 4. Thus our precision tests are subject to a selection issue insofar as firms/managers do not issue guidance prior to quarter q + 4. We conduct a Heckman two-stage analysis as a robustness check to capture the decision to report guidance; inferences remain unchanged (not tabulated).

TABLE 7

Hazard Model Analysis of Guidance Precision andExecutive Turnover Among Frequent Guiders

Dependent Variable: ResumePrecisionq+1,q+4

Coefficient p-value Hazard Ratios

CEOTurnoverq 0.091 1.10

CFOTurnoverq -0.250a 0.78

Precisionq- 4, q-1 -0.539 0.58

PrecisionVolatq- 4, q- 1 -0.631 0.53

AHorizon[q-4,q-1],[q+1,q+4] -0.001 1.00

Litigationq -1q 0.036 1.04

Restateq —1,q -0.083 0.92

Restructuringq - 1 0.021 1.02

EPSVolatq-4,q-1 -0.009 0.99

Returnq _1 0.170 1.19

ALoss [ q-4, q-1],[ q+1, q+4] -0.196 0.82

AFSE [q-4,q-1],[q+1,q+4] 0.092 1.10

AAnalystFollow[q-4,q—1], [q+1,q+4] 0.051 1.05

IndPropq 0.196 1.22

Sizeq -1 -0.007 0.99

BooktoMarketq -1 -0.215 0.81

Year fixed effects Included

Number of observations 6,078

Number of CEO/CFO turnovers_170/214_

The duration of time in the hazard model is the time until the firm issues quarterly earnings guidance after quarter q that is at least as precise as the quarterly earnings guidance issued in quarters q — n to q — 1, with adjustment for right censoring at the end of a four-quarter period. Bolded coefficients and p-values are statistically significant (two-tailed p-values <0.10); p-values are based on standard errors that have been clustered by firm and are presented in italics below the coefficients. ResumePrecisionq+1,q+4 is the number of quarters, from one to four, until the firm issues quarterly earnings guidance after quarter q with a precision "score" greater than or equal to the average precision score of the quarterly earnings guidance issued in quarters q — n to q — 1 (where the precision "score" is set to zero for qualitative guidance, one for an open-ended interval, two for a range estimate, and three for a point estimate). See appendix A for additional variable definitions.

A,aa,a Statistically different from CEOTurnover at p < 0.01, p < 0.05, and p < 0.10 (two-tailed), respectively.

5. Conclusion

We examine the relation between CEO and CFO turnovers and quarterly management earnings guidance to infer manager-specific effects on

the provision and formation of guidance. While firm- and industry-specific determinants explain most of the guidance decision, we do find some evidence that top executive turnovers are associated with the provision ofguid-ance. Although we document an association between CEO turnovers and the provision of guidance, we are not able to disentangle this effect from firm-initiated policy changes. Conversely, among CFOs, we find no evidence that, following turnovers, breaks in guidance among frequent guiders are driven by firm-initiated changes in guidance policy; rather, these breaks are consistently associated with the background of the incoming CFO.15 Thus, we conclude that temporary breaks in guidance following CFO turnovers result from a lack of firm- and industry-specific forecasting knowledge on the part of the incoming CFO. Interestingly, among firms that historically have not provided earnings guidance, we document that new appointments of externally hired CEOs are associated with an initiation of earnings guidance, consistent with both taste issues on the part of the incoming CEO (Gibbons, Richardson, and Waterhouse [1990], Bamber, Jiang, and Wang [2010]) and specific CEOs being hired to support firm-initiated guidance policy changes.

A caveat of our study is that executive turnovers and changes in guidance might be jointly determined by contemporaneous events that are not captured by our firm- and industry-specific determinants of guidance issuance. In addition to examining performance terciles and concurrent changes in the board chairman for frequent guiders and nonguiders, among frequent guiders we examine plausibly exogenous executive turnovers—those where the outgoing executive is hired away by another firm. We consistently find evidence of breaks in guidance following CFO turnovers among frequent guiders, but, as noted above, the results are less clear regarding the role of the CEO.

Overall, our results speak to the impact of individual managers on firmlevel disclosure. While we document that the choice to provide guidance is predominately firm-specific (e.g., 63% of frequent guiders still issue earnings guidance in the quarter following an executive turnover), CFO

15 We also examine public announcements of decisions to stop issuing guidance for our sample firms (e.g., Chen, Matsumoto, and Rajgopal [2011]). Public announcers represent a small subset of all firms interrupting guidance—we identify 51 public announcements among our sample firms. We examine the association between CEO and CFO turnovers and these public announcements and find that CEO turnovers are significantly likely to coincide (i.e., to occur in the same quarter) with public announcements (not tabulated). We find no statistically significant evidence, however, that public announcements are likely to precede CEO turnovers (by one to four quarters). Hence, it is difficult to determine whether publicly announced changes in guidance policy coinciding with CEO turnovers are initiated by the firm or by the new CEO. Among CFO turnovers, we find no association between turnovers and public announcements. This additional evidence favors the interpretation that the breaks following CFO turnovers do not result from firm-initiated changes in guidance policy. We also include an indicator for public announcements as an additional determinant in equation (1), which does not affect our inferences (not tabulated).

turnovers have an economically meaningful impact on guidance among frequent guiders, reducing the likelihood of providing guidance in the next quarter by approximately 7%, on average, and by over 13% if the newly appointed CFOs do not have prior forecasting experience. Among nonguiders, we find some evidence that newly appointed externally hired CEOs initiate guidance; however, we cannot fully disentangle the manager-specific effects from firm-initiated changes in guidance policy. We conclude that CEOs participate in firm-level policy decisions, while CFOs are involved in the formation and discussion of guidance.

Future research might explore whether other disclosure policies (guidance being only a specific subset of voluntary disclosure) vary systematically with managerial style, ability, or any other individual-level fixed effect (Bertrand and Schoar [2003], Bamber, Jiang, and Wang [2010], Baik, Farber, and Lee [2011]). For example, do more confident newly appointed managers provide supplemental information to credibly signal that they have sufficient knowledge to form their forecast (e.g., Hutton, Miller, and Skinner [2003], Baginski, Hassell, and Kimbrough [2004], Hirst, Koonce, and Venkataraman [2007])? Do individual managers' tastes extend to pro forma reporting? If so, are the effects more concentrated among CEOs or CFOs? We leave these questions for future research.

APPENDIX A

Variable Definitions

MAIN VARIABLES

CEOTurnoverq (CFOTurnoverq): indicator variable equal to one if there is a change in CEO (CFO) during quarter q, zero otherwise.

NextGuidanceq+1,q+8: number of quarters, from one to eight, until the firm issues quarterly earnings guidance after quarter q (i.e., a value of one [two] indicates that the firm issues quarterly earnings guidance in quarter q + 1 [q + 2] for the first time after quarter q).

Guidanceq+iq+n: indicator variable equal to one if the firm issues quarterly earnings guidance in at least one quarter from quarters q + 1 through q + n, zero otherwise.

ResumePrecisionq+1 q+4: number of quarters, from one to four, until the firm issues quarterly earnings guidance after quarter q with a precision "score" greater than or equal to the average precision score of the quarterly earnings guidance issued in quarters q - n to q - 1, where the precision "score" is set to zero for qualitative guidance, one for an open-ended interval (minimum or maximum), two for a range estimate, and three for a point estimate (i.e., a value of one [two] indicates that the firm issues guidance in quarter q + 1 [q + 2] that is, for the first time after quarter q, at least as precise as in quarter[s] q — 1 [q — 2 to q — 1]).

VARIABLES TO INVESTIGATE THE ENDOGENOUS NATURE OF TURNOVERS

NewChairmanq: indicator variable equal to one if there is a newly appointed chairman of the board of directors in quarter q, zero otherwise.

CEOTurnoverHiredAwayq (CFOTurnoverHiredAwayq): indicator variable equal to one if the outgoing CEO (CFO) in quarter q leaves immediately for another company (i.e., when press releases mention the CEO's [CFO's] departure along with his/her appointment in a new firm), zero otherwise.

CEOTurnoverNotHiredAwayq (CFOTurnoverNotHiredAwayq): indicator variable equal to one if the outgoing CEO (CFO) in quarter q is not hired immediately by another company, zero otherwise.

INCOMING EXECUTIVE CHARACTERISTICS

InternalCEOq (InternalCFOq): indicator variable equal to one if the incoming CEO (CFO) in quarter q is promoted within the same firm, zero otherwise.

ExternalCEOq (ExternalCFOq): indicator variable equal to one if the incoming CEO (CFO) in quarter q is hired from another firm, zero otherwise.

PriorExperienceCEOq (PriorExperienceCFOq): indicator variable equal to one if the incoming CEO (CFO) in quarter q held a CEO (CFO) position in another firm that has issued at least one quarterly earnings forecast in quarters q — 4 to q — 1, zero otherwise.

NoPriorExperienceCEOq (NoPriorExperienceCFOq): indicator variable equal to one if the incoming CEO (CFO) in quarter q has no prior experience as defined above.

SameIndusCEOq (SameIndusCFOq): indicator variable equal to one if the incoming CEO (CFO) in quarter q either comes directly from a firm (or subsidiary of a conglomerate) within the same industry or has held a top executive position in a firm within the same industry in the recent past, zero otherwise.

DiffIndusCEOq (DiffIndusCFOq): indicator variable equal to one if the incoming CEO (CFO) in quarter q neither comes directly from a firm (or subsidiary of a conglomerate) within the same industry nor has held any top executive position in a firm within the same industry in the recent past, zero otherwise.

FIRM- AND INDUSTRY-SPECIFIC DETERMINANTS OF GUIDANCE ISSUANCE

Guidanceq: indicator variable equal to one if the firm issues quarterly earnings guidance in quarter q, zero otherwise.

NbGuidanceq—n,q—\: number of quarters from quarters q — n through q — 1 during which the firm issued quarterly earnings guidance, zero otherwise.

AHorizon[q—4 q— 1],[q+1,q+4]: change, from quarters [q — 4, q — 1] to quarters [q + 1, q + 4], in the average number of days between a management quarterly earnings forecast and the end of the fiscal quarter.

Litigationq—1q: indicator variable equal to one if the firm is subject to a securities lawsuit in quarters q — 1 or q (per the Stanford Securities Class Action Clearinghouse), zero otherwise.

Restateq—: indicator variable equal to one if the firm announced a restatement during quarters q — 1 or q (per the GAO Financial Restatement Database), zero otherwise.

Restructuringq—1: indicator variable equal to one if the firm reports restructuring charges in quarter q — 1 (as measured by a nonzero value for the variable RCAQin Compustat), zero otherwise.

EPSVolatq—8q—1: standard deviation of quarterly earnings per share over quarters q — 8 through q — 1.

Returnq—1: cumulative size-adjusted return over quarter q — 1.

Lossq—8,q—1: percentage of quarters during which the firm reported negative earnings over quarters q — 8 to q — 1.

FSEq—4 q—1: percentage of quarters, over quarters q — 4 to q — 1, during which the firm fell short of the consensus analyst expectations (mean of the latest forecasts from each broker in the FirstCall database) upon announcing quarterly earnings.

AnalystFolloWq_ 1: natural logarithm of one plus the number of analysts following the firm during quarter q — 1.

IndPropq: percentage of the firm's industry peers (based on two-digit SIC) that issue quarterly earnings guidance during quarter q.

Sizeq—1: natural logarithm of total assets as of the end of quarter

q — 1.

BooktoMarketq—x: ratio of a firm's book value of equity to its market value of equity as of the end of quarter q — 1.

Years: indicator variables for calendar years.

APPENDIX B

Data Collection and Turnover Characteristics

We use the ExecuComp database to initially identify the occurrence of CEO and CFO turnovers. We then refine our sample and retrieve turnover dates by searching Dow Jones, PR, and Reuters Newswires as well as Wall Street Journal articles via Factiva. We collect information regarding the background of the incoming executives. More precisely, we identify whether the incoming executive is promoted within the firm or hired from another firm, and, among those hired from outside the firm, if the incoming executive is hired from the same or different industry and if the incoming executive held a similar position in another firm that issued guidance. We examine a total of 1,993 executive (CEO or CFO) turnovers across frequent guiders, infrequent guiders, and nonguiders; fewer observations are included in our regression analysis. The frequency breakdown of these turnovers is as follows:

Frequent Guiders Infrequent Guiders Nonguiders

CEO CFO CEO CFO CEO CFO

Turnovers Turnovers Turnovers Turnovers Turnovers Turnovers

Turnovers 180 227 288 409 382 507

Internal 133 117 176 192 244 278

(Percentage of (74%) (52%) (61%) (47%) (64%) (55%)

turnovers)

External 47 110 112 217 138 229

(Percentage of (26%) (48%) (39%) (53%) (36%) (45%)

turnovers)

Prior experience 6 30 8 36 22 26

(Percentage of (13%) (27%) (7%) (17%) (16%) (11%)

external)

No prior experience 41 80 104 181 116 203

(Percentage of (87%) (73%) (93%) (83%) (84%) (89%)

external)

Same industry 30 35 43 73 71 76

(Percentage of (64%) (32%) (38%) (34%) (51%) (33%)

external)

Different industry 17 75 69 144 67 153

(Percentage of (36%) (68%) (62%) (66%) (49%) (67%)

external)

See appendix A for definitions of turnover characteristics.

To aid in our investigation of the potentially endogenous nature of turnovers, for frequent guiders we also identify a subset of relatively exogenous turnovers where the outgoing executive is hired away. Specifically, when press releases mention the executive's departure along with his/her appointment in a new firm, we classify these observations as "hired away" turnovers. Among frequent guiders, we identify 7 (164) CEO turnovers where the outgoing CEO was hired away (not hired away), and 39 (175) CFO turnovers where the outgoing CFO was hired away (not hired away). Below are three examples of turnovers: one classified as hired away and the other two as not hired away.

HIRED-AWAY TURNOVER

"[...] American Greetings Corp. (AM) named Stephen J. Smith, currently treasurer and head of investor relations, chief financial officer, replacing Michael J. Merriman who is leaving to become chief executive officer of

Lamson & Sessions Co. (LMS). [...]"

"American Greetings Announces CFO Succession and Third Repurchase Plan," press release, 10/26/2006, Dow Jones Newswires.

NOT HIRED-AWAY TURNOVERS

"[...] Milacron Inc. said on Friday that its board of directors had elected Ronald Brown chairman and chief executive officer. Brown replaces Daniel

Meyer, Milacron's chairman and chief executive since 1990, who will retire at the end of May. [...]"

"Milacron says appoints Brown as chairman, CEO," press release, 2/9/2001, Reuters News.

"[...] Phillips-Van Heusen Corp. (PVH) named Emanuel Chirico chief executive succeeding Mark Weber, who has left the company 'by agreement with the board.' [...]"

"Phillips-Van Heusen Names Emanuel Chirico CEO," press release, 2/27/2006, Dow Jones Newswires.

REFERENCES

Ajinkya, B.; S. Bhojraj; and P. Sengupta. 'The Association Between Outside Directors, Institutional Investors and the Properties of Management Earnings Forecasts." Journal ofAccounting Research 43 (2005): 343-76.

Anilowski, C.; M. Feng; and D. Skinner. "Does Earnings Guidance Affect Market Returns? The Nature and Information Content of Aggregate Earnings Guidance." Journal of Accounting and Economics 44 (2007): 36-63.

Baginski, S., and J. Hassell. "Determinants of Management Forecast Precision." The Accounting Review 72 (1997): 303-12.

Baginski, S.; J. Hassell; and M. Kimbrough. 'Why Do Managers Explain Their Earnings Forecasts?" Journal of Accounting Research 42 (2004): 1-29.

Baik, B.; D. Farber; and S. Lee. "CEO Ability and Management Earnings Forecasts." Contemporary Accounting Research (2011): Forthcoming.

Bamber, L.; J. Jiang; and I. Wang. "What's My Style? The Influence of Top Managers on Voluntary Corporate Financial Disclosure." The Accounting Review 85 (2010): 1131-62.

Beatty, A.; B. Ke; and K. Petroni. "Earnings Management to Avoid Earnings Declines Across Publicly and Privately Held Banks." The Accounting Review 77 (2002): 547-70.

Bergman, N., and S. Roychowdhury. "Investor Sentiment and Corporate Disclosure." Journal of Accounting Research 46 (2008): 1057-972.

Bertrand, M., and A. Schoar. "Managing with Style: The Effect of Managers on Firm Policies." The Quarterly Journal of Economics (2003): 1169-208.

Bhojraj, S.; R. Libby; and H. Yang. "Analyzing Guidance at the Firm Level: The Effect of Reputation-Building and Learning-by-Doing on Guidance Frequency and Guidance Properties." Working paper, Cornell University, 2011.

Botosan, C. "Disclosure Level and the Cost of Capital." The Accounting Review 72 (1997): 323-49.

Brown, S.; S. Hillegeist; and K. Lo. "Management Forecasts and Litigation Risk." Working paper, University of British Columbia, 2005.

Chen, S.; D. Matsumoto; and S. Rajgopal. "Is Silence Golden? An Empirical Analysis of Firms That Stop Giving Quarterly Earnings Guidance." Journal of Accounting and Economics 51 (2011): 134-50.

Chuk, E.; D. Matsumoto; and G. Miller. "Assessing Methods of Identifying Management Forecasts: CIG vs. Researcher Collected." Working paper, University of Washington, 2009.

Coller, M., and T. Yohn. "Management Forecasts and Information Asymmetry: An Examination of Bid-Ask Spreads." Journal of Accounting Research 35 (1997): 181-91.

Diamond, D., and R. Verrecchia. "Disclosure, Liquidity and the Cost of Capital." Journal of Finance 46 (1991): 1325-59.

Dyreng, S.; M. Hanlon; and E. Maydew. 'The Effects of Executives on Corporate Tax Avoidance." The Accounting Review 85 (2010): 1163-89.

Farber, D. "Restoring Trust After Fraud: Does Corporate Governance Matter?" The Accounting Review 80 (2005): 539-61.

Feng, M. "Why Do Managers Meet or Slightly Beat Earnings Forecasts in Equilibrium?" Working paper, University of Pittsburgh, 2006.

Feng, M., and A. Koch. "Once Bitten, Twice Shy: The Relation Between Outcomes of Earnings Guidance and Management Guidance Strategy." The Accounting Review 85 (2010): 1951-84.

Feng, M.; C. Li; and S. McVay. "Internal Control and Management Guidance." Journal of Accounting and Economics 48 (2009): 190-209.

Ge, W.; D. Matsumoto; and J. Zhang. "Do CFOs Have Styles of Their Own? An Empirical Investigation of the Effect of Individual CFOs on Financial Reporting Practices. Contemporary Accounting Research (2011): Forthcoming.

Gibbons, M.; A. Richardson; and J. Waterhouse. "The Management of Corporate Financial Disclosure: Opportunism, Ritualism, Policies, and Processes." Journal of Accounting Research 32 (1990): 75-102.

Graham,J.; C. Harvey; and S. Rajgopal. "The Economic Implications of Corporate Financial Reporting." Journal of Accounting and Economics 40 (2005): 3-73.

Healy, P.; A. Hutton; and K. Palepu. "Stock Performance and Intermediation Changes Surrounding Sustained Increases in Disclosure." Contemporary Accounting Research 16 (1999): 485-520.

Hirst, E.; L. Koonce; and S. Venkataraman. "How Disaggregation Enhances the Credibility of Management Earnings Forecasts." Journal of Accounting Research 45 (2007): 811-37.

Hirst, E.; L. Koonce; and S. Venkataraman. "Management Earnings Forecasts: A Review and Framework." Accounting Horizons 22 (2008): 315-38.

Houston, J.; B. Lev; and J. Tucker. "To Guide or Not to Guide? Causes and Consequences of Stopping Quarterly Earnings Guidance." Contemporary Accounting Research 27 (2009): 143-85.

Hutton, A.; G. Miller; and D. Skinner. "The Role of Supplementary Statements with Management Earnings Forecasts." Journal of Accounting Research 41 (2003): 867-90.

Karamanou, I., and N. Vafeas. "The Association Between Corporate Boards, Audit Committees, and Management Earnings Forecasts: An Empirical Analysis." Journal of Accounting Research 43 (2005): 453-86.

Kasznik, R. "On the Association Between Voluntary Disclosure and Earnings Management." Journal of Accounting Research 37 (1999): 57-81.

King, R.; G. pownall; and G. Waymire. "Expectations Adjustments via Timely Earnings Forecast Disclosure: Review, Synthesis and Suggestions for Future Research." Journal of Accounting Literature (1990): 113-44.

Lang, M., and R. Lundholm. "Voluntary Disclosure and Equity Offerings: Reducing Information Asymmetry or Hyping the Stock?" Contemporary Accounting Research 17 (2000): 623-62.

McNichols, M. "Evidence of Information Asymmetries from Management Earnings Forecasts and Stock Returns." The Accounting Review 64 (1989): 1-27.

Miller, G. "Earnings Performance and Discretionary Disclosure." Journal ofAccounting Research 40 (2002): 173-204.

O'Brien, P.; M. McNichols; and H. Weilin. "Analyst Impartiality and Investment Banking Relationships." Journal of Accounting Research 43 (2005): 623-50.

Parrino, R. "CEO Turnover and Outside Succession: A Cross-Sectional Analysis." Journal of Financial Economics 46 (1997): 165-97.

Petersen, M. "Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches." Review of Financial Studies 22 (2009): 435-480.

pownall, G.; C. Wasley; and G. Waymire. "The Stock Price Effects of Alternative Types of Management Earnings Forecasts." The Accounting Review 68 (1993): 896-912.

Richardson, S.; A. Tuna; and P. Wysocki. "Accounting for Taste: Board Member Preferences and Corporate Policy Choices." Working paper, University of Pennsylvania, 2003.

Rogers, J. "Disclosure Quality and Management Trading Incentives." Journal ofAccounting Research 46 (2008): 1265-96.

Rogers, J., and P. Stocken. "Credibility of Management Forecasts." The Accounting Review 80 (2005): 1125-62.

Rogers,J., and A. Van Buskirk. "Shareholder Litigation and Changes in Disclosure Behavior."

Journal of Accounting and Economics 47 (2009): 136—56.

Skinner, D. "Why Firms Voluntarily Disclose Bad News." Journal of Accounting Research 32 (1994): 38-60.

Skinner, D. "Earnings Disclosures and Stockholder Lawsuits." Journal of Accounting and Economics 23 (1997): 249-82.

Skinner, D., and R. Sloan. "Earnings Surprises, Growth Expectations, and Stock Returns, or Don't Let an Earnings Torpedo Sink Your Portfolio." Review of Accounting Studies 7 (2002): 289-312.

Soffer, L.; R. Thiagarajan; and B. Walther. "Earnings Preannouncement Strategies." Review of Accounting Studies 5 (2000): 5-26.

Trueman, B. "Why Do Managers Voluntarily Release Earnings Forecasts?" Journal of Accounting andEconomics 8 (1986): 53-71.

Tse, S., and J. Tucker. "Within-Industry Timing of Earnings Warnings: Do Managers Herd?" Review of Accounting Studies 15 (2010): 879-914.

Waymire, G. "Earnings Volatility and Voluntary Management Forecast Disclosure." Journal of Accounting Research 23 (1985): 268-95.

Weisbach, M. "Outside Directors and CEO Turnover." Journal of Financial Economics 20 (1988): 431-60.

Williams, P. "The Relation Between a Prior Earnings Forecast by Management and Analyst Response to a Current Management Forecast." The Accounting Review 71 (1996): 103-13.

Yang, H. "Capital Market Consequences to Managers' Voluntary Disclosure Styles." Working paper, University of Pennsylvania, 2010.

Zamora, V. "Do Managers Benefit from Superior Forecasting?" Working paper, Seattle University, 2009.