Full length articleDoes government report readability matter? Evidence from market reactions to AAERs
Introduction
The SEC strives to continuously (1) improve communications with investors, (2) improve the information environment in markets, and (3) facilitate investors’ decision making. In October 1998, the SEC adopted a plain English mandate [SEC Rule 421(d)]. The Plain English Rule was introduced to “[ensure] the orderly and clear presentation of complex information so that investors have the best possible chance of understanding it” (SEC, 1998, p. 5). This rule was accompanied by the publication of A Plain English Handbook: How to Create Clear SEC Disclosure Documents, guiding securities lawyers and companies to reduce legalese and jargon. Although there is a large body of literature studying the readability of corporate disclosures (see Loughran and McDonald, 2016 for a review), little research has investigated the readability of government documents. While there is ample evidence suggesting that regulatory actions affect capital markets and investors’ decision making (Christensen et al., 2010, Feroz et al., 1991, McDowell, 2005, Nourayi, 1994), prior research has not investigated whether government report readability influences investors’ decision making. We extend this literature by focusing on government disclosures.
Following a long stream of complaints by the public on the poor readability of US government documents,1 the Plain Writing Act was signed into law (Public Law 111–274) in October 2010.2 The Act requires federal government agencies, including the SEC, to use plain writing in “every covered document of the agency that the agency issues or substantially revises” (Public Law 111–274, Section 4), where plain writing is defined as writing that is “clear, concise, well-organized, and follows other best practices appropriate to the subject or field and intended audience” (Public Law 111–274, Section 3).
Despite growing regulatory efforts to improve communication in public documents, empirical evidence to date indicates that reports issued by governmental or regulatory bodies are low in readability (Adelberg, 1982, Bradbury et al., 2018, Kinnersley and Fleischman, 2001, Smith and Richardson, 1999). For example, in 2013, Editor Software (“The Plain English Company”) completed an audit of 100 federal government website documents and concluded that, out of all the federal government documents surveyed, only 3% were written in plain English according to their StyleWriter editing program.3 In the context of capital markets, we are interested in understanding whether the readability of government disclosures matters to investors in the marketplace. Specifically, we examine how the readability of Accounting and Auditing Enforcement Releases (AAERs) announced by the U.S. Securities and Exchange Commission (SEC), influences market reactions to AAERs.4
Given its regulatory and enforcement responsibilities, the SEC does not have an incentive to strategically vary disclosure narratives. Hence, examining AAER reports enables us to isolate the effect of document readability on market reactions without the threat of strategic preparer bias in report readability, which likely confounds similar analyses of company disclosures (Hooghiemstra et al. 2017; Li, 2008).5 Further, AAERs are arguably a primary channel for the markets to understand the nature of the misconduct underlying the SEC’s enforcement actions (Feroz et al., 1991, Nourayi, 1994). In contrast, corporate disclosures are usually accompanied by various alternative information sources for investors to obtain supplemental information (Hwang and Kim, 2017, Lehavy et al., 2011).6 Hence, by focusing on AAERs, we avoid these alternative information sources and are better able to isolate the influence of other information channels and attribute the observed effects to AAER report readability.7
Prior research applies multiple theories to explain why investors’ investment decisions are influenced by the readability of disclosures. On the one hand, the incomplete revelation hypothesis asserts that the markets react less to information contained in complex disclosures because investors’ costs of extracting information increase when reading complex reports, thereby reducing their willingness to collect or process the information (Bloomfield, 2002, Grossman and Stiglitz, 1980, Lundholm, 1991). Given this perspective, we expect that markets will underreact to an AAER report that is more difficult to read.
On the other hand, the judgment and decision-making literature posits that people often rely on heuristics or rules of thumb to make decisions, especially when processing information demands a considerable amount of time and effort (Hirshleifer and Teoh, 2003, Libby et al., 2002, Tversky and Kahneman, 1974). Complex information, such as difficult-to-read AAERs, obstructs investors from comprehending the details of the event or gaining insights on its severity and thus creates uncertainty. Individuals are inclined to use the affect heuristic (i.e., a mental shortcut that allows people to make decisions and solve problems quickly and efficiently) and rapidly consult their affective feelings in making judgments and decisions, especially under uncertainty (Slovic et al., 2007). Investors generally believe that an AAER announcement represents a value-destroying event to the AAER firm (Christensen et al., 2010, Dechow et al., 1996, Feroz et al., 1991, Karpoff et al., 2017, McDowell, 2005, Nourayi, 1994). The psychology and neurobiology literatures suggest that individuals tend to develop a high degree of anxiety under adverse circumstances, particularly when the adversity is accompanied by uncertainty. As an instinctive response to anxiety and uncertainty, individuals often engage in defensive behavior to overcome an immediate threat (Blanchard and Blanchard, 1988, Fanselow, 1994, Graeff, 1994, Gray and McNaughton, 2000). Accordingly, investors may assume the worst-case scenario and bid down the market value of the AAER firm’s shares. We therefore expect markets to react more negatively to an AAER report that is written obscurely.
Using a sample of 222 AAERs from January 1, 1999 until December 31, 2016, we examine the effect of report readability that is measured by Bog Index based on the SEC’s Plain English Rule of 1998 (i.e., A Plain English Handbook).8 Our regression analysis controls for firm-specific characteristics as well as the complexity and severity of an AAER.9 We find evidence consistent with the view that investors rely on specific heuristics to make their decisions when AAER reports are more difficult to read. More specifically, we find that (1) less readable AAER reports are associated with more negative market reactions and (2) the relation between an AAER’s readability and market reactions is more pronounced when the investor base mainly consists of less sophisticated investors, when investors lack alternative channels for acquiring information on the AAER event (i.e., less visible firms), or when there is greater uncertainty in the markets (i.e., firms with higher stock return volatility before the AAER announcement). We employ various methods to control for the endogenous determination of AAER readability (i.e., selection bias). Our results are also robust to alternative model specifications.
Our study contributes to the literature in several important ways. First, while regulatory efforts to improve communication in public documents are increasing, studies on government report readability are rather scarce.10 Therefore, we extend this literature by examining the influence of government report readability on capital markets. Given regulatory concerns over plain English writing for government agencies, our investigation of AAER readability should be of interest to regulators who are concerned with writing clarity and reporting quality in government documents.
Second, our focus on AAER reports is unique in examining the effect of disclosure readability. Prior studies focus on corporate disclosures and there is a general concern that disclosures made by firms are susceptible to preparer bias, which could cloud the effect of report readability. For example, observed negative market reactions to bad news disclosed in a less readable way could be explained by the possibility that investors can “see through” managerial intent (i.e., obfuscation) in reporting and consider low readability to be a warning sign, thereby sanctioning the firm for their lack of transparency (Merkl-Davies and Brennan, 2007, Tan et al., 2014). Alternatively, investors could view greater linguistic complexity in bad news disclosures as evidence that the preparer is attempting to provide incremental information for the complex news. Therefore, all else being equal, market reactions could be less negative (Bushee et al., 2017). In either case, it is challenging to attribute observed market reactions to report readability as opposed to preparer bias. By focusing on an AAER setting in which report preparer (i.e., the SEC) does not have obvious incentives to bias disclosure narratives, we are better able to tease out the impact of report readability.11
Relatedly, in our setting, the nature of the news is less ambiguous. Rennekamp (2012, p. 1326) contends that “[i]t is important to understand how investors react to bad news disclosures that are low in readability given prior claims that managers strategically obfuscate negative news.” However, in a real-world scenario, it is challenging to collect a “clean” sample of bad news, because report preparers have incentives to influence the formation of investors’ perceptions and their judgment on the nature of the news (Laksmana et al., 2012, Hooghiemstra et al., 2017). Indeed, prior research finds that corporate disclosure preparers may manage reporting tone strategically to portray their firm’s economic prospects in the best possible light (Huang et al., 2014). Investors may even react positively to bad news when it is glossed over and prospects are puffed up. In contrast, AAER announcements clearly deliver adverse news to investors (Christensen et al., 2010, Dechow et al., 1996, Feroz et al., 1991, Karpoff et al., 2017, McDowell, 2005, Nourayi, 1994). Thus, we are better able to infer whether investors react more (or less) negatively to AAER readability when the nature of AAERs is definitive. Overall, our exploration of AAER report readability is in line with recent efforts in textual analysis studies that employ novel settings to address challenges facing the literature (Bozanic et al., 2017, Hwang and Kim, 2017, Lo et al., 2017).
Section snippets
Related research
The underlying rationale for the relation between market reactions and disclosure readability is well-rooted in the literature on human psychological and cognitive processes. This line of research suggests that investors are influenced by firms’ reporting strategy because their decision making is subject to bounded rationality (e.g., Kahneman and Tversky, 1979; Schlenker, 1980, Tetlock and Manstead, 1985). Specifically, the incomplete revelation hypothesis posits that investors are subject to
Data and sample selection
Our investigation period is from January 1, 1999 until December 31, 2016. We manually collect regulatory enforcement actions on accounting and auditing matters from the SEC’s AAER database,14 and we complement this data collection process with the AAER data set from Berkeley’s Center for Financial Reporting and Management (CFRM). The AAER database consists of enforcement actions concerning civil lawsuits brought by the SEC in federal
Descriptive statistics
Fig. 2 describes the development of AAER readability over time. According to StyleWriter’s plain English standard, any document with a BOG score above 60 is rated to be of poor readability. The Bog Index of AAERs is above 60 in all the years during our investigation period, except for 1999,25 suggesting that AAERs are generally difficult to read. Surprisingly, the Bog Index of AAERs
Shareholder sophistication
Hypotheses 1a and 1b assume that investors are, on average, unable to fully process the information contained in AAERs that are written in a less readable manner. However, our assumption could vary with shareholder sophistication. Less sophisticated investors, constrained by information processing capacity, are more susceptible to the way in which information is presented, as the use of heuristics tends to dominate less sophisticated investors’ judgment and decision making process. In contrast,
Endogeneity
Clearly, it is hard to claim that AAER readability is a random treatment, as both observable and unobservable factors could determine the way in which an AAER is written and influence market reaction to the AAER. Although we add a host of control variables in our main regressions to capture the observable factors, there are potentially unobservable factors that could explain our results. To mitigate the selection on unobservables (Lennox et al., 2012), we rely on a two-stage Heckman selection
Conclusions
The U.S. government has made continuous efforts to improve the readability of government reports. However, prior evidence suggests that regulatory disclosures and legal documents still suffer low readability (Adelberg, 1982, Daily et al., 2010, Smith and Richardson, 1999). Using a sample of AAERs from 1999 until 2016, we find evidence consistent with the notion that government report readability matters—that is, poor readability at the disclosure of bad news gives rise to greater investor
Acknowledgements
We thank Marco Trombetta and two anonymous referees for their most constructive advice. We also appreciate the valuable comments and suggestions provided by Mary Barth, Patricia Dechow, Miles Gietzmann, Ian Gow, Reggy Hooghiemstra, and Ole-Kristian Hope. We are grateful to Bill McDonald for his insights on the plain English measurements. We thank Nick Wright (Editor Software) for sharing information on the U.S. government agency style audit based on StyleWriter’s statistics and assistance in
References (92)
- et al.
Do sophisticated investors interpret earnings conference call tone differently than investors at large? Evidence from short sales
J. Corp. Finance
(2015) Discussion of “Annual report readability, current earnings, and earnings persistence”
J. Account. Econ.
(2008)- et al.
A plain English measure of financial reporting readability
J. Account. Econ.
(2017) - et al.
Measuring security price performance
J. Financ. Econ.
(1980) - et al.
Using daily stock returns: the case of event studies
J. Financ. Econ.
(1985) - et al.
Frontal midline theta reflects anxiety and cognitive control: meta-analytic evidence
J. Physiol.-Paris
(2015) Startle reactivity and anxiety disorders: aversive conditioning, context, and neurobiology
Biol. Psychiatry
(2002)- et al.
Is the SEC captured? Evidence from comment-letter reviews
J. Account. Econ.
(2017) - et al.
Limited attention, information disclosure, and financial reporting
J. Account. Econ.
(2003) - et al.
It pays to write well
J. Financ. Econ.
(2017)
Capital markets research in accounting
J. Account. Econ.
Measuring long-horizon security price performance
J. Financ. Econ.
Econometrics of event studies
Compensation discussion and analysis (CD&A): readability and management obfuscation
J. Account. Public Policy
On the use of instrumental variables in accounting research
J. Account. Econ.
Individual investors and financial disclosure
J. Account. Econ.
Annual report readability, current earnings, and earnings persistence
J. Account. Econ.
Experimental research in financial accounting
Account., Organ. Soc.
Earnings management and annual report readability
J. Account. Econ.
A role for the human dorsal anterior cingulate cortex in fear expression
Biol. Psychiatry
Stock price responses to the SEC's enforcement actions
J. Account. Public Policy
The adaptive threat bias in anxiety: Amygdala–dorsomedial prefrontal cortex coupling and aversive amplification
NeuroImage
The affect heuristic
Eur. J. Oper. Res.
Role of the bed nucleus of the stria terminalis versus the amygdala in fear, stress, and anxiety
Eur. J. Pharmacol.
An empirical evaluation of the communication of authoritative pronouncements in accounting
Account. Finance
Threat-related attentional bias in anxious and nonanxious individuals: a meta-analytic study
Psychol. Bull.
The balance sheet as an earnings management constraint
Account. Rev.
Judgment in Managerial Decision Making
Ethoexperimental approaches to the biology of emotion
Annu. Rev. Psychol.
The role of dissemination in market liquidity: evidence from firms' use of Twitter™
Account. Rev.
The “incomplete revelation hypothesis” and financial reporting
Account. Horizons
The impact of narrative disclosure readability on bond ratings and the cost of debt
Rev. Acc. Stud.
Soft information in loan agreements
J. Account., Audit. Finance
Summary annual reports: length, readability and content
Account. Finance
Linguistic complexity in firm disclosures: obfuscation or information?
J. Account. Res.
Domestic and foreign earnings, stock return variability, and the impact of investor sophistication
J. Account. Res.
The Econometrics of Financial Markets
Market efficiency and investor reactions to SEC fraud investigations
J. Forensic Invest. Account.
Readability of tax court opinions
Phasic vs sustained fear in rats and humans: role of the extended amygdala in fear vs anxiety
Neuropsychopharmacology
Predicting material accounting misstatements
Contemp. Account. Res.
Causes and consequences of earnings manipulation: an analysis of firms subject to enforcement actions by the SEC
Contemp. Account. Res.
Efficient capital markets: II
J. Finance
The adjustment of stock prices to new information
Int. Econ. Rev.
Neural organization of the defensive behavior system responsible for fear
Psychonomic Bull. Rev.
Cited by (6)
Market reactions to voluntary disclosures of a strategic alliance agreement: evidence from China
2023, Revista Espanola de Financiacion y ContabilidadIs Neuroaccounting Taking a Place on the Stage? A Review of the Influence of Neuroscience on Accounting Research
2022, European Accounting ReviewReadability of Accounting Disclosures: A Comprehensive Review and Research Agenda*
2021, Accounting Perspectives