1 Impression Management and Non-GAAP Disclosure in Earnings Announcements

72 Pages·2014·0.42 MB·English

Impression Management and Non-GAAP Disclosure
in Earnings Announcements*

ENCARNA GUILLAMON-SAORIN, Universidad Carlos III de Madrid
HELENA ISIDRO, Instituto Universitário de Lisboa (ISCTE-IUL), Lisboa, Portugal
ANA MARQUES, Nova School of Business and Economics - Universidade Nova de Lisboa1

July 2014

∗ The authors appreciate the valuable comments received from Francesco Bova, Beatriz
Garcia, Ross Jennings, Elmar Venter, and the workshop participants at the 2014 Mid-year
meeting of the International Accounting Section of AAA, the 2013 Chulalongkorn
Accounting and Finance Symposium, the EAA 2012 annual meeting, the IX Workshop on
empirical research in financial accounting, Nova School of Business and Economics,
Grenoble Ecole de Management, Bristol University, Durham University, Universidad de
Navarra, Cass Business School, Bangor University, Indian Institute of Management
Bangalore, and 2013 conference on recent developments in international accounting,
University of Miami. We thank António Miguel for providing us with data on international
risk factors. This study was supported by the Foundation for Science and Technology in
Portugal (grant PTDC/EGE-GES/103770/2008). Encarna Guillamon-Saorin acknowledges
financial contribution from the Spanish Ministry of Science and Innovation (SEJ2007-
67582-C02-02/ECON, ECO2010-19314) and Comunidad Autonoma de Madrid (SEJ2008-
00059-003). Ana Marques also acknowledges financial support received from Nova
Forum. We are grateful for the excellent research assistance of Arash Aloosh and Luís
Araújo.

1Corresponding author: Nova School of Business and Economics, Campus de Campolide
1099-032 Lisboa, Portugal. Tel: [phone]. Fax: [phone]. Email:
[email protected]

1
Impression Management and Non-GAAP Disclosure
in Earnings Announcements
Abstract
We study the market reaction to the disclosure of non-GAAP earnings measures in
earnings announcement press releases that are combined with a high level of
impression management. Our sample includes observations from the largest
European firms during the period of 2003 to 2009. We hand-collect and code non-
GAAP information and impression management information and create an
impression management score that captures the communication style of non-GAAP
disclosures. We find that managers who exclude recurring earnings components
from their self-constructed non-GAAP earnings figures use higher levels of
impression management. This finding is consistent with the idea that managers use
impression management strategically to hide the persistence of non-GAAP
adjustments. Market reaction tests suggest that investors are able to identify
managers’ intentions and react negatively to the disclosure of non-GAAP
information that is surrounded by high impression management. Moreover,
investors’ reaction is stronger in countries with more sophisticated users and more
stringent regulations and enforcement. Our results are robust to several impression
management definitions, and estimations methods.
Keywords: pro forma; disclosure tone; investor protection; sophisticated markets.
JEL classification: M41
2
1. Introduction
We study the market reaction to the disclosure of non-GAAP earnings
measures that are accompanied by impression management communication
techniques, and how market reaction varies across countries. We find evidence
consistent with managers using high levels of impression management to mask the
recurring nature of some non-GAAP adjustments. We also find that investors
perceive managers’ disclosure as strategic and penalize firms for that behavior. The
country-level tests indicate that investors’ reaction to the disclosure of non-GAAP
earnings with a high level of impression management is more negative in countries
with more sophisticated users and more stringent regulations and enforcement.
From an information perspective accounting earnings are important because
they convey information that can change investors’ beliefs about future earnings
(Watts and Zimmerman 1986). Investors are aware that current earnings contain
transitory and price-irrelevant components, and knowing the varying persistence of
earnings components can therefore become an informational advantage. As
markets value persistent earnings (Collins and Kothari 1989), firms have incentives
to separate permanent and transitory earnings components. However, measurement
and disclosure of earnings is constrained by GAAP and subject to monitoring. In
their search for more flexible ways to convey information about earnings
persistence managers have turned to non-GAAP disclosures in earnings press
3
releases. These earnings measures are calculated by excluding transitory
components from GAAP earnings and are disclosed in the firm’s earnings
announcement press releases.
Market participants perceive non-GAAP earnings measures to be
informative about future earnings (Bradshaw and Sloan 2002; Bhattacharya, Black,
Christensen, and Larson 2003). Nevertheless, because such measures are manager-
adjusted and non-audited, there are concerns that firms use non-GAAP disclosure
strategically to positively bias investors’ perceptions about the performance of the
firm (Andersson and Hellman 2007; Bhattacharya, Black, Christensen, and
Mergenthaler 2007; Cormier, Lapointe-Antunes, and Magnan 2011). That concern
is greater if regulation and monitoring of non-GAAP disclosures are not present.
One issue is that managers can adjust GAAP earnings for recurring expenses
resulting in non-GAAP earnings that overstate permanent earnings. In the spirit of
Frankel, McVay, and Soliman (2011) and Jennings and Marques (2011) we
analyze the cross-sectional variation in the persistence of non-GAAP adjustments
to make inferences about informative versus opportunistic intentions. Consistent
with these studies we find that non-GAAP adjustments have a recurring nature.
By disclosing non-GAAP measures in earnings press releases managers use
a communication channel that affords them great flexibility in terms of content and
format of the message. Press releases are widely used by the business community,
4
and also create ample opportunity for discretionary disclosure. There is empirical
evidence of strategic use of communication techniques in earnings press releases,
such as positive language tone to mislead investors about the firm’s future
performance (Lang and Lundholm 2000; Huang, Teoh, and Zhang 2014). These
communication techniques are usually referred to as impression management (Neu
1991; Neu, Warsame, and Pedwell 1998). We hand-collect and code data from the
earnings announcements press releases for a sample of large European firms. We
measure the level of impression management surrounding non-GAAP using a
score, derived from content analysis, which captures three specific communication
techniques: disclosure tone, emphasis, and performance comparisons.
The low litigation environment and absence of strict regulation on non-
GAAP disclosure that characterizes many European markets facilitate the use of
discretionary information in earnings announcements. In contrast to the US, where
Regulation G stipulates stringent rules on non-GAAP reporting and where the SEC
frequently monitors financial disclosures, in Europe there are only
recommendations issued by the Committee of European Securities Regulators
(CESR 2005). The recommendation encourages European-listed firms that choose
to disclose non-GAAP financial measures to do it “in a way that is appropriate and
useful for investors’ decision making.” However, a survey conducted by the
European Financial Reporting Advisory Group (EFRAG 2009) concludes that non-
GAAP disclosures in Europe are inconsistent and obscure. Capital market
5
sophistication and regulation quality are also relatively less developed in Europe
than in the US, which suggests that the potential market effects of misleading
disclosures can be high, particularly for less sophisticated investors. All this makes
Europe an interesting setting to conduct our analysis.
We initially assess whether there is a firm-level style, i.e., whether firms
consistently disclose non-GAAP measures with a similar level of impression
management techniques. We find no evidence of an impression management firm
style. We assign firms to 20 (and 10) impression management portfolios and find
that only a small percentage (7% for the 10 portfolio analysis) consistently falls
into the same group over the years. We also confirm that managers communicate
using more impression management around non-GAAP disclosures when the non-
GAAP adjustments have greater persistence. This suggests that managers seek to
mask the persistence of the adjustments through the use of communications
strategies1.

1 One could argue that given the difficulty to identify who writes the press release (Merkl-Davies
and Brennan 2007; Garcia Osma and Guillamon-Saorin 2011), it is unlikely that the person
preparing the press release is the same person who determines the non-GAAP exclusions and its
presentation in the press release even though the manager is accountable for its content. Given the
lack of evidence in the literature, we can assume that the quality of firm communication is an
equilibrium outcome (Ball 2006), which implies consistent quality levels across the range of
reported information prepared within a firm, regardless of the number of parties involved
(Gronstedt 1996). This leads to the general expectation that impression management and Non-
GAAP information are positively associated. In line with our prediction, earlier research provides
evidence indicating that earnings management is positively associated with the use of impression
management practices (Godfrey, Mather, and Ramsay 2003; Aerts and Cheng 2011), and disclosure
tone (Huang et al. 2014).
6
Our first hypothesis assesses whether the market reaction to non-GAAP
disclosures around the earnings announcement date is affected by impression
management. We find that market participants react negatively to non-GAAP
disclosures when these disclosures are surrounded by a high level of impression
management. Our findings suggest this disclosure strategy is interpreted by the
market as an attempt to mask the persistence of the non-GAAP adjustments, and
penalize the firm. Investors perceive that non-GAAP earnings measures are an
overstatement of the firm’s performance and discount the adjustments made. This
is in line with earlier evidence indicating that investors incorporate the tone of
managers’ communications in their valuation decisions (Baginski, Demers, Wang,
and Yu 2011; Demers and Vega 2011; Davis, Piger, and Sedor 2012), but that they
are capable of detecting managers’ strategic use of communication techniques
(Barton and Mercer 2005; Huang et al. 2014). Another important finding of our
study is that we show that the market reaction is not limited to language tone in
earnings communications. Differently from other studies, we explore and find
market effect for other communication techniques, namely emphasis and
performance comparisons.
We subsequently evaluate the extent to which countries’ institutional and
market conditions influence investors’ reaction to the combination of non-GAAP

7
disclosure and high impression management in earnings announcements.
International literature documents that macro factors such as the level of investors’
protection, the quality of enforcement mechanisms, and the sophistication of
market participants affect reporting transparency (Bushman, Piotroski, and Smith
2004; Holthausen 2009; Lang, Lins, and Maffett 2012). Better regulatory
environments and the presence of more sophisticated investors capable of
processing and monitoring financial information are likely to curb managers’
aggressive non-GAAP disclosures. Investors are more likely to have the means and
motivation to discount aggressive disclosures in those environments. Our results
are aligned with this argument. We find that the market reaction to non-GAAP
disclosures surrounded by a high level of impression management is more negative
in countries with more sophisticated market participants (i.e. financial analysts and
institutional investors). We also find the same result in countries having more
efficient systems in place to protect minority investors. Our results are robust to
selection and endogeneity bias and to alternative definitions of the impression
management score.
This study contributes to the voluntary disclosure literature in three ways.
First, it shows that managers complement disclosures of non-GAAP financial
measures with impression management techniques. While earlier studies on
management communication focus mostly on language tone and its effect on
markets, we investigate a wide range of impression management techniques
8
including qualitative and quantitative information, and assess the market reaction
to the combination of these techniques with non-GAAP disclosures. Second, we
provide evidence that country-level factors affect how markets react to non-GAAP
adjustments when these are surrounded by a high level of impression management.
Third, we document the use of impression management techniques in narratives in
an international context, and to the best of our knowledge, ours is the first study to
so.
In addition to these contributions to academic research, our study also has
policy implications. In February of 2014 the ESMA-European Securities and
Markets Authority (which replaced CESR) issued a consultation paper with
guidelines for the disclosure of alternative performance measures. The objective of
ESMA is to strengthen the transparency principles contained in the 2005
recommendation. We believe that our findings are useful to European regulators
concerned with matters such as designing these guidelines.
The remainder of the paper is organized as follows. In the next section we
discuss earlier findings and develop the hypotheses. In Section 3 we present the
sample, the hand-collection, and coding of the data. Section 4 explains the research
design. In Section 5 we present the main results. Section 6 reports additional
analyses and Section 7 concludes.
9
2. Literature review and hypotheses development
When making capital allocation decisions investors use earnings
information to predict future earnings and future returns. The accuracy of such
predictions depends on earnings persistence, and earlier research shows that market
participants reward persistence (Collins and Kothari 1989). The possibility of
market rewards, as well as reputation and compensation motives, gives managers
an incentive to provide capital markets with self-constructed earnings measures
that can inform investors about the transitory nature of certain items included in
GAAP earnings. Consistent with this idea, earlier studies on the informativeness of
non-GAAP earnings report that investors perceive non-GAAP earnings to be more
informative about future earnings than GAAP earnings (Bradshaw and Sloan 2002;
Bhattacharya et al. 2003).
However, because non-GAAP earnings are created by managers, are not
monitored, and are based mostly on the exclusion of expenses, there is a potential
risk that investors will be misled. While some managers may adjust GAAP
earnings to provide a better measure of permanent earnings than GAAP earnings,
others may exclude recurring earnings components in an attempt to enhance
investors’ perceptions of the firm’s profitability. Whether or not the users of non-
GAAP information can see through managers’ intentions depends upon the
sophistication of the investors (Christensen, Drake, and Thornock 2014). If
investors are able to evaluate the accuracy of the adjustments made by managers
10