# Leveling the playing field for less-sophisticated non-professional investors: Does plain English matter?

Eileen Taylor (Poole College of Management, North Carolina State University, Raleigh, North Carolina, USA)

ISSN: 2514-4774

Publication date: 13 October 2017

## Abstract

### Purpose

The purpose of this paper is to explore how non-professional investors (NPIs) with varying levels of financial sophistication interpret and perceive corporate disclosures and management credibility, specifically risk factors, when those disclosures are presented in readable and less-readable formats.

### Design/methodology/approach

The paper uses an online experiment to test hypotheses related to the effects of financial sophistication (measured) and readability (manipulated) on NPIs’ equity valuations and perceptions of management credibility (competence and trustworthiness).

### Findings

Increased readability appears to counteract less-sophisticated NPIs’ conservatism in equity valuations, such that they are not statistically significantly different from more-sophisticated NPIs’ equity valuations. Further, less-sophisticated NPIs judge management as less competent when disclosures are less readable, while more-sophisticated NPIs judge management as more competent when disclosures are less readable.

### Research limitations/implications

The paper has important implications for the SEC’s regulations related to plain English requirements for risk factor and other corporate disclosures. Financial sophistication varies among NPIs, and readability appears to influence these individuals in different ways.

### Practical implications

The SEC’s Concept Release (April 13, 2016) acknowledges the need to update and improve risk factor disclosure regulations. This study provides evidence that contributes to those decisions.

### Originality/value

The paper extends the research on processing fluency, by examining readability of disclosures with a consistent tone (negative). The NPIs surveyed are directly representative of the population of interest for risk factor disclosure regulations.

## Keywords

#### Citation

Taylor, E. and Riley, J. (2017), "Leveling the playing field for less-sophisticated non-professional investors: Does plain English matter?", Journal of Capital Markets Studies, Vol. 1 No. 1, pp. 36-57. https://doi.org/10.1108/JCMS-10-2017-004

### Publisher

:

Emerald Publishing Limited

## Introduction

With the increase in online trading and wider stock market access for non-professional investors (NPIs), in the USA, and especially globally, it is of interest to examine how current reporting standards and practices may differentially influence subsets of this group. In the USA, the SEC has offered guidance on ways to improve disclosures by making them more timely and understandable; i.e., investors must be able to read and understand corporate disclosures in order to make informed decisions and manage their risk (Linsley and Shrives, 2005). Research finds that NPI underperformance in the market largely arises from their inability to understand financial information (e.g. Barber and Odean, 2000; Odean, 1999). Coram (2010) further proposed that a lack of task-specific knowledge by NPIs results in conservatism in equity valuation and investment behavior.

In 1998, the SEC helped to increase the understandability of financial information by passing the Plain English Rule 421(d), accompanied by the “Plain English Handbook” (Securities and Exchange Commission, 1998), which provides guidance to registrants in writing certain disclosures using plain English principles. Such principles include short sentences; definite, concrete, and everyday words; active voice; tables or bullet point lists; no legal jargon and highly technical business terms; and no multiple negatives (Securities and Exchange Commission, 1998).

The rationale for increasing readability through plain English seems clear and logical. Plain English disclosures should improve readability by presenting information akin to individuals’ normal processing style (Securities and Exchange Commission, 1998). This “processing fluency” represents the ease with which individuals process external information (Winkielman et al., 2003). Such disciplines as psychology, communications, and marketing have studied fluency (for a review see Alter and Oppenheimer, 2009), and consistently find that fluency influences judgments over and above content (Schwarz et al., 1991). Disclosure information presented in plain English should be more readable and more fluent, and thus be easier for NPIs to process.

Research indicates, however, that financial information has become too disfluent and complex for NPIs to process (Miller, 2010; Li, 2008; Courtis, 1995, 1998, 2004; Linsley and Lawrence, 2007; Jones and Shoemaker, 1994). Additionally, registrants may use lower readability to hide information or confuse investors. Some studies find evidence that managers use less-readable disclosures to obfuscate negative and declining performance in communications to investors such as the MD&A (Courtis, 2004; Subramanian et al., 1993; Li, 2008; Rutherford, 2003; Jones and Shoemaker, 1994). Other managers may go so far as to decrease readability to deceive investors into believing that performance is better than the numbers show (Lee et al., 2013).

While it seems logical that plain English disclosures would be more useful to investors than less-readable or other complex language disclosures, this is far from proven. Questions remain about whether plain English is always better for communicating information; particularly in communicating corporate information to investors, researchers have questioned the effects of requiring plain English on investors’ understanding, perceptions, and decision making (Cui, 2016; Tan et al., 2014, 2015; Rennekamp, 2012). Overall, these studies suggest that we need more research to understand the varied effects of readability on understanding, perceptions, and decision making, and that the relationship between readability and investor behavior is more complex than previously assumed.

This paper explores the effects of readability in the context of risk disclosures, which became part of the required SEC registrants’ disclosures since 2005. This regulation requires registrants to disclose risks “[…] that may adversely affect the issuer’s business, operations, industry, or financial position, or its future financial performance” (Securities and Exchange Commission, 2004, 2005). Because the SEC intended this disclosure to help all investors, including NPIs, it falls under the above plain English principles. This study is especially timely in light of the SEC’s Concept Release on April 15, 2016 that seeks public comment on suggested changes to business and financial disclosures required by regulation S-K, and presents the results of “a comprehensive evaluation of the Commission’s disclosure requirements recommended in the staff’s report on Review of Disclosure Requirements in Regulation S-K, which was mandated by Section 108 of the Jumpstart Our Business Startups Act (‘JOBS’ Act)” (Securities and Exchange Commission, 2016, p. 9).

Using survey responses from 359 NPIs, we examine how readability and task-specific knowledge (i.e. financial sophistication) affect judgments about equity valuation, ease of understanding, and management credibility (i.e. competence and trustworthiness). Overall, findings indicate that investors with low financial sophistication (measured using a composite index similar to Tan et al., 2014) consistently provided lower valuation estimates than those with high financial sophistication, regardless of readability. However, all investors are influenced by readability and provide higher valuation estimates when readability is high. No evidence finds plain English significantly increases less-sophisticated NPIs’ understanding over more-sophisticated NPIs. Although understanding does increase for all investors under plain English, it is more pronounced in the more-sophisticated NPI group. Financial sophistication and readability interact to influence perceptions of management credibility. Low-readability results in decreased perceptions of credibility among less financially sophisticated investors and increased perceptions among more financially sophisticated investors. This finding implies that less financially sophisticated investors may be skeptical of managers who use difficult-to-read language, but financially sophisticated investors may actually reward managers who use complex language.

Importantly, we find an apparent similarity of decisions and perceptions between less-sophisticated NPIs presented with plain English disclosures and high sophisticated NPIs presented with less-readable disclosures. On most measures, both groups exhibited nearly identical behaviors, implying that increasing the readability of disclosures does help compensate for a lack of task-specific knowledge, and “levels the playing field” between less- and more-sophisticated NPIs. Of course, corporations will not produce disclosures in two different formats, and since plain English did appear to impact more-sophisticated NPIs as well, more research is needed to determine whether increasing readability is beneficial to the NPI population overall.

Similarly, to the studies by Rennekamp (2012) and Cui (2016), we conclude that readability is one part of a complex formula for investor judgments, and that plain English requirements are not a panacea for improving corporate disclosure usefulness. Readability differences have inconsistent effects across a variety of disclosures studied in prior research, and across the NPI population. Focusing on plain English may divert attention from other challenges to the SEC’s goal to provide useful information to the investing public. The results contribute to the growing body of research on the style and presentation of disclosures, and the subsequent effect on investors’ behavior. Unlike the study by Miller (2010), which examined the entire 10-K, this study isolates risk disclosures, thereby limiting the potential influence of other information and information overload. This study adds to Rennekamp’s (2012) findings that are based on a single press release regarding earnings, by exploring the effects of sophistication on non-financial disclosure judgments. It also complements Tan et al.’s (2014) findings by holding language sentiment constant. The findings indicate that readability affects NPIs regardless of management’s choice of positive or neutral language. Further, this study contributes to this literature because our subject pool comprises members of NPIs’ education and support association. Thus, the conclusions are based upon individual respondents who are the target of the SEC’s initiatives on plain English disclosures.

These results should be of interest to public company registrants, the SEC, and other investor protection agencies (e.g. Financial Industry Regulatory Authority), NPIs and their supporting organizations (e.g. BetterInvesting), and standard-setters. Further, because US investors have access to many resources, such as analyst reports, financial media, etc., it may be even more important for regulatory bodies in countries other than the USA to be aware of our findings. Overall, this area will benefit from additional research, and theory development and testing, to understand and predict how written language influences investment decisions.

## Background and hypotheses

In concert with the substantial increase of NPIs in the market (Coram, 2010), much research has examined their use of financial information to make investment decisions. Such research has indicated that the internet-fueled increase in access to valuable financial information has not seen improved investment behavior. NPIs systematically misinterpret the financial information available (Odean, 1999), such that they underperform the market by 3.7 percent annually (Barber and Odean, 2000). This research indicates that a lack of financial knowledge and an inability to understand financial information contribute to NPIs’ investment underperformance. In keeping with prior literature, we use the term sophistication to refer to the level of investor’s task-specific knowledge, namely, their familiarity with investing and financial analysis.

Coram (2010) proposed that a lack of sophistication results in conservatism in equity valuation and investment behavior because the likelihood of unintended effects (i.e. conservatism) of cognitive impairment is negatively related to such knowledge (Smith and Kida, 1991). When compared to professional investors who are sophisticated, NPIs make more conservative judgments when reading positive non-financial disclosures (Coram, 2010).

Thus, sophistication impacts NPIs’ investment behavior vs professional investors; it may also influence behavior within the NPI group. NPIs are not a homogenous group (Pinsker and Wheeler, 2009), and some NPIs are likely to be more financially sophisticated than are others. Sophistication is a contributing factor to investment behavior because it enables an investor to understand and appropriately use the value-relevant information (Victoravich, 2010; Frederickson and Miller, 2004). However, because it is not possible, nor fair to exclude NPIs from the capital markets based on their lack of sophistication, it is important to determine if any interventions can compensate for this shortcoming. Readability may be a solution.

We also find negative effects of complexity and readability on investor decisions that the archival literature documented. For example, You and Zhang (2009) found investors under-react to 10-K information, which is greater for firms with more complex reports. Miller (2010) found that more complex 10-Ks result in lower trading activity by, and less consensus among, small investors, concluding that complex disclosures may be too costly for NPIs to process.

Plain English disclosures should improve readability by presenting information consistent with individuals’ normal processing style (Securities and Exchange Commission, 1998). This “processing fluency” represents the ease with which individuals can process external information (Winkielman et al., 2003). Disciplines such as psychology, communications, and marketing have studied fluency (for a review see e.g. Alter and Oppenheimer, 2009), which is shown to impact judgments beyond content (Schwarz et al., 1991). Disclosure information presented in plain English should be more readable, or more fluent, and thus should be easier for less-sophisticated NPIs to process.

However, despite regulators, legislators, and professional groups’ contentions on plain English’s positive effects, research evidence is inconsistent. Some studies suggest that plain English results in greater comprehension of bank contracts (Campbell, 1999), insurance policies (Blazzard and Hasenauer, 1998), and mutual fund prospectuses (Johnson, 2004). Others, however, caution that plain English initiatives may not always result in the intended outcome. Bernstein (2006) reported that some earlier SEC attempts to make 10-Ks more readable to average investors caused companies to simply eliminate disclosures. Jacoby et al. (1982) found that the Federal Trade Commission’s attempts to improve communications by using plain English actually resulted in greater confusion in a wide range of audiences.

Recent research in accounting raises even more questions about the effects of plain English disclosures. Rennekamp (2012) found small investors susceptible to overreaction, responding more positively to good news and more negatively to bad news when press releases are more readable. This finding is based on a homogenous pool of NPIs, and is therefore unclear if the effect holds across all NPIs (i.e. those with both high and low sophistication).

Recent readability and sophistication studies find both variables affecting investor decisions and behavior. Tan et al. (2014) showed that readability effects may reverse when considered jointly with other variables such as positive or negative language sentiment. Tan et al. (2015) found that inconsistencies in benchmarks affect the response of investors to strategically readable management disclosures. Cui (2016) also examined inconsistencies, and found that sophistication and readability interact.

While these studies examine readability and NPI’s sophistication levels in the context of some inconsistent message and/or tone, and use either MBA students (Cui, 2016; Tan et al., 2014, 2015) or Amazon MTurk participants, the present study differs. We examine readability and sophistication in a consistent message (presentation of risk factors, a negative disclosure), and the participants are all active non-professional US nationwide investors who have earned varying educational degrees ranging from high school to PhDs and MDs. Thus, this adds generalizability to Tan et al.’s (2014, 2015) and Cui’s (2016) findings. Tan et al. (2014) noted that their sample may be unrepresentative of the continuum of investor sophistication, and suggested future research on this topic. These points do not criticize Tan et al. (2014, 2015) or Cui (2016), but rather point to divergence to indicate the complementary aspects of this work.

Based on prior research, a more conservative bias should prevail for less-sophisticated NPIs, if they cannot fully process a risk disclosure. Increasing the readability of disclosures should increase understanding and reduce conservatism. H1a tests whether less-sophisticated NPIs are more conservative than more-sophisticated NPIs. H1b tests the effect of readability on conservatism. If readability of risk factors reduces conservatism (increases equity valuations) for less-sophisticated NPIs, then the SEC’s focus on readability is justified. Formally:

H1a.

Less-sophisticated NPIs will estimate lower equity valuations than more-sophisticated NPIs.

H1b.

Less-sophisticated NPIs will estimate lower equity valuations when a disclosure is less readable than when it is more readable.

If support suggests that less-sophisticated investors are influenced by readability, before whole-heartedly recommending plain English across the board, we must also explore readability’s effects on more-sophisticated investors. Does readability similarly affect both high- and low-sophistication NPIs? High-sophistication NPIs should understand disclosures regardless of readability, which should result in little difference in their equity valuations. When considered with the above hypotheses, this would indicate an ordinal interaction: the more-sophisticated investors’ valuations forming a straight line between readability conditions and the less-sophisticated forming a diagonal line increasing from low to high readability (approaching, but remaining below the more-sophisticated investors). However, an important factor affecting the use of disclosures is the perception of message or source credibility (Pinsker and Wheeler, 2009; Mercer, 2004). NPIs’ perceptions of credibility may be influenced by assurance (Pinsker and Wheeler, 2009), labeling (Koonce et al., 2005), and other factors (Mercer, 2004), including readability. Lower source credibility may negatively influence equity valuations, introducing complexity into the judgment.

On this relationship, Rennekamp (2012) found a positive effect of readability on management credibility indirectly through increased processing fluency. But again this result relies on studying NPIs as a homogenous group. Tan et al. (2014) reported that more-sophisticated NPIs may view less-readable disclosures as management’s attempt to obfuscate information, and that they see through management’s strategic intentions when a stimulus is difficult to read by punishing management with lower equity valuations. Supplemental analysis indicates that this effect arises partially from lower perceptions of credibility when positive language is difficult to read. This conforms to psychology research that perceived bias from a source reduces the message’s credibility (Birnbaum and Stegner, 1979). The management obfuscation hypothesis proposes that managers have an incentive to make annual reports difficult to read to impair investors from incorporating adverse information into stock prices (Bloomfield, 2002). Thus, more-sophisticated NPIs may be more skeptical of complex wording regardless of whether the message is positive or adverse. They may then react by lowering their opinion of managers who use difficult-to-read language, thus resulting in a lower equity valuation. The effect of readability on equity valuations for this more-sophisticated group of NPIs is proposed in H2 below:

H2.

More-sophisticated NPIs will estimate lower equity valuations when a disclosure is less readable than when it is more readable.

While H1a and H2 predict lower equity valuations for all NPIs using less-readable disclosures, we propose there are different reasons at the different investor sophistication levels. To test directly these proposals, we examine whether the hypothesized effects arise from less-sophisticated NPIs’ increased understanding of plain English disclosures or more-sophisticated NPIs’ reduced perceptions of management credibility using less-readable disclosures. Prior research implies that it may be both. Plain English should compensate for the lack of task-specific knowledge (sophistication) expected to contribute to less-sophisticated NPIs’ conservatism[1], leading to H3:

H3.

Less-sophisticated NPIs will perceive lower understanding when disclosure readability is lower than when readability is higher.

At the same time, low readability should decrease perceptions of management credibility, but only for more-sophisticated NPIs. Stated formally:

H4.

More-sophisticated NPIs will perceive lower management credibility when disclosure readability is lower than when readability is higher.

## Method

### Participants

Study participants consisted of 359 NPIs. Participants completed an online survey distributed by BetterInvesting Inc., an organization that provides educational webinars, web-based mutual funds and stock tools, in-person learning events, and similar activities for NPIs. The sample contains 207 (58 percent) males and 152 (42 percent) females. Average years investing is 23.38 years, average trades per year is 10.23, and average age is 65.03. Although this average age appears high, it is common for research samples of NPIs (Coram, 2010; Elliott et al., 2008). Of the 44,000 people who received the e-mail survey invitation, 1,279 people started the survey, for an initial response rate of 3 percent. We eliminated incomplete attempts, resulting in 359 complete, usable responses. Full demographic information appears in Table I.

### Design

A 2×2 (disclosure readability by investor sophistication level) between-participants design tests the hypotheses. Participants are randomly assigned to either more- or less-readable treatments. Investor sophistication level is measured through demographic information. Tan et al. (2014) created a sophistication index using the number of accounting and finance courses, and the frequency of reading annual reports and earnings releases to determine delineation of sophistication of participants. Following similar logic, the sophistication index in the present study uses the presence of an accounting or finance degree and the frequency of buying and selling stock over the last year.

Participants received identical financial statements and background information for a fictional company (see Appendix 1), followed by the disclosure that is the subject of the readability manipulation (see Appendix 2). The experiment investigates the readability and sophistication issue in the context of one specific type of required disclosure: risk factors. Regulation S-K, implemented by the SEC in 2005, requires that companies list and discuss “[…] the most significant factors that may adversely affect the issuer’s business, operations, industry, financial position, or its future financial performance” within their 10-K filing (Securities and Exchange Commission, 2004, 2005). The regulation represents the latest in the SEC’s ongoing goal of providing full and fair risk disclosure: a focus that began in earnest in 1994 after massive derivative losses surprised investors of companies such as Proctor & Gamble (Linsmeier and Pearson, 1997). We examine risk disclosure readability because the SEC specifically required companies to use plain English for these disclosures. Related research uses earnings’ press releases (e.g. Tan et al., 2014, 2015; Rennekamp, 2012) that have no plain English requirements, and additionally, are quantitative. Thus, this study complements this research by examining a qualitative disclosure that the SEC currently mandates be written in plain English.

To develop the disclosure, we reviewed risk factors from the 10-Ks of a sample of 90 Fortune 500 companies. A review of item 1a disclosures (2004-2008) shows that companies report a range of risks (Blaskovich et al., 2012). Some are simple and common, while others are complex and company specific. Examples of common risks include lost revenues from competition, regulation, and unforeseen natural disasters. Less commonly mentioned risks include inability to produce accurate financial reports due to control weaknesses, forced repayment due to debt covenant violations, and risk of declining financial performance due to changes in accounting rules. The frequency of other risks falls somewhere between such as risk of litigation due to computer hacking or security lapses.

To simulate reality, disclosures in three diverse areas were created: sales, data security, and control over financial reporting[2]. Investor’ reactions to risk are partially based on familiarity (Song and Schwarz, 2009), warranting a diverse set of factors along this continuum. The first risk factor describes a common risk of sales revenue declines, due to a competitive environment and potentially negative effects of changing consumer preferences. The second risk factor discusses the moderately common risk of data security breaches, related to the increasing practice of collecting and storing sensitive client data. The third risk factor addresses the uncommon risk that an internal control weakness may compromise the reliability of the financial reports. We varied risk presentation order to test and control for potential order effects.

To manipulate readability, we chose the text of a risk factor disclosure from an actual 10-K, which served as the low-readability condition[3]. The SEC’s “Plain English Handbook” (Securities and Exchange Commission, 1998) served as a guide for revision of each risk factor into a high-readability condition, using readability scores from MS Word.

An important consideration is the difference in length that is inherent in a plain English disclosure, which by definition, calls for short sentences and active voice. Since length is related to disfluency (Miller, 2010), prior studies hold length constant across readability (Rennekamp, 2012). However, length was allowed to occur naturally in the plain English disclosures. If firms adhere to the SEC mandate, current disclosures would most likely be shorter and more concise. Thus, these disclosures more closely represent what investors would see in a plain English disclosure. Although the impact of report length is a consideration in investor behavior, we must note that Miller’s findings are based on the length of a full 10-K. This study’s experimental materials are a small subset of information, and the effect of the difference in overall length between the two readability conditions is small. To investigate the effect of length, participants were asked to report their perception of the length of the disclosures. Although the perception of length between the groups differs significantly, it does not correlate significantly with their equity valuation or their perceptions of credibility (not tabulated). Accordingly, length differences are not believed to affect the hypothesized results.

Two questions assessed the language manipulation’s effectiveness. Differences between both readability conditions are significant in the expected direction for the following questions: the risk factors were easy to read (t=−5.040, p<0.000, two-tailed), and written in plain English (t=−5.206, p<0.000, two-tailed). These results hold when the questions are examined for investor sophistication level, as both less- and more-sophisticated participants successfully indicate more-readable vs less-readable language. Accordingly, the conclusion is that the readability manipulation was successful.

After reviewing the information, participants provided their expected equity valuation and perceptions of management credibility (discussed in more detail in the following section). Finally, standard demographic data were collected.

## Results

### Sample

Following the index procedure discussed previously, the sophistication scores of the sampled NPIs range from −2.56 to 5.32. Investors were split into sophistication groups based on the median, resulting in 169 (190) less-sophisticated (more-sophisticated) participants. The mean sophistication scores for more and less sophistication are −0.66 and 0.86, respectively. Less (more) sophisticated investors buy and sell stock an average of 7.97 (12.56) times per year. No significant differences arose between the two groups for mean age or years investing. Less- vs more-sophisticated investors’ mean age is 65.77 and 64.28 years, and mean years investing is 23.35 and 24.01 years, respectively[4].

### Equity valuations

The first two hypotheses test whether the equity valuations of NPIs differ based on their sophistication level; and whether readability influences equity valuations for less-sophisticated investors. These hypotheses were tested by examining the stock price estimates of less-vs more-sophisticated participants who were provided with more-readable vs less-readable risk disclosures. Descriptive statistics appear in Panel A of Table II.

Our two-way ANOVA helped us to determine the effects of sophistication and readability on equity valuations. Results from Panel B of Table II, indicate that the interaction is not significant (F=0.75, p=0.39). A statistically significant main effect difference in stock price estimates arose for all investors because of sophistication level (F=3.89, p=0.02), and for all investors because of readability (F=5.28, p=0.05). As shown in Figure 1, less-sophisticated investors estimate lower stock prices than more-sophisticated investors, and all investors estimate lower equity values when disclosures are less readable. Because of unequal sample sizes, the main effects of sophistication and readability using the estimated marginal means are used for interpretation. The marginal means for equity value are $25.46 for less-sophisticated NPIs and$25.92 for more-sophisticated NPIs, a statistically significant difference of $−0.47 (p=0.05). This supports H1a, which predicted that less-sophisticated investors would estimate lower equity prices than more-sophisticated investors. Turning to readability, the marginal means for equity value are$25.42 for less-readable disclosures and $25.96 for more-readable disclosures, a statistically significant difference of$−0.54 (p=0.02). H1b and H2 predict lower equity prices when disclosures appear in less-readable vs more-readable language.

These results support the overall expectation that a lack of task-specific knowledge contributes to less-sophisticated NPIs’ conservatism. It also supports the proposal that disclosure readability influences NPIs’ judgments, the SEC’s main target for plain English requirements. Of critical note in the results is the similarity between less-sophisticated NPIs using plain English disclosures and more-sophisticated NPIs using less-readable disclosures. A t-test performed between these two groups finds no significant difference between their mean equity valuations ($25.83 vs$25.75, respectively, t=0.27, p=0.79, not tabulated). This suggests that an increase in readability does appear to compensate for task-specific knowledge in less-sophisticated investors to the point of creating a “level playing field” with more-sophisticated investors. Considering that the current state of readability in financial information is quite low, this provides support for the SECs continuing effort to encourage plain English.

### Understanding and management credibility

The final two hypotheses investigate whether the difference in equity valuations relates to increased understanding (for less-sophisticated NPIs) and decreased management credibility perceptions (for more-sophisticated NPIs). Specifically, less-sophisticated NPIs are expected to report greater understanding when readability is high vs low. More-sophisticated investors are expected to be more skeptical of management when readability is low vs high, as they are more likely to recognize management obfuscation. To measure understanding, we asked participants how difficult it was for them to understand the disclosures. Descriptive statistics of the dependent variable for ease of understanding by treatment condition are presented in Table III, Panel A, and ANOVA results in Panel B. The interaction between sophistication and readability on understanding is not significant (F=0.56, p=0.45). A significant main effect of readability on all investors (F=7.56, p=0.01) is found, indicating that NPIs viewed the plain English disclosure as easier to understand, regardless of sophistication level. H3 is not supported, as increased readability results in increased ease of understanding for NPIs overall, but not solely for less-sophisticated NPIs. When each sophistication level is examined separately, the more-sophisticated NPIs perceive a difference in understanding from readability. t-Tests on the two levels indicate that while not statistically significant for less-sophisticated NPIs (t-statistic=1.55, p=0.12), the difference in ease of understanding is significant for more-sophisticated NPIs (t-statistic=2.26, p=0.03). The more-sophisticated NPIs drive the overall significant effect of readability on ease of understanding. This unexpected result is considered in the “Discussion” section of the paper (Figure 2).

A six-item questionnaire assessed perceptions of credibility, which we adapted (presented in Appendix 3) from the studies by McCroskey (1966) and Leathers (1992), and validated in the study by Mercer (2005). Following Mercer, a composite measure of management credibility is calculated, based on trustworthiness and competence, which we evaluated individually below. Descriptive statistics of the composite measure of credibility by treatment condition appear in Table IV, Panel A, and ANOVA results appear in Panel B. A significant interaction arises between sophistication and readability on management credibility (F=4.76, p=0.03). Simple main effects analysis indicates a statistically significant difference in management credibility between less-sophisticated NPIs in the two readability conditions. These results appear in Panel D of Table IV. When readability is low, less-sophisticated NPIs’ perceptions of management credibility are significantly lower than when using more-readable disclosures (3.88 vs 4.05, p=0.04). However, more-sophisticated NPIs’ perceptions of management are not significantly affected by readability (4.04 low readability vs 3.97 high readability, p=0.32). Also, when readability is low, less-sophisticated NPIs’ perceptions of management credibility are significantly lower than more-sophisticated NPIs (3.88 vs 4.04, p=0.05). Taken together, these results imply that it is less-sophisticated NPIs who are more skeptical of management who use low-readability disclosures, not more-sophisticated NPIs. Thus, H4 is not supported, and rather, is the opposite of expected for sophistication level (Figure 3).

Per Mercer (2005), the credibility scale contains two factors: competence (the first three questions) and trustworthiness (the last three questions). We examine both components separately. Tables V and VI present descriptive statistics (Panel A) and ANOVA results (Panel B) for respective measures of competence and trustworthiness. While no significant differences in results on the trustworthiness measure are found, results do support an interactive effect of readability and education on competence. As presented in Panel B of Table V, the interaction is significant (p=0.04). Simple effects tests (Panel C) indicate a significant difference in mean competence perceptions by less- vs more-sophisticated NPIs when readability is low (3.68 vs 3.92, p=0.04). Additionally, the lowest perception of management competence is found when less-sophisticated NPIs experience low-readability disclosures (mean=3.68) and the highest perception is noted when less-sophisticated NPIs experience high-readability disclosures (mean=3.97). This difference is significant (p=0.01). When readability is high, less-sophisticated investors will perceive higher competence of management than when readability is low, while more-sophisticated investors will perceive lower competence of management when readability is high than when low. These results may result from expectations gaps between more- and less-sophisticated investors when it comes to management competence.

### Plain English as a compensating factor

The primary results above indicate the similarity between the equity valuations of less-sophisticated NPIs using plain English disclosures and more-sophisticated NPIs using less-readable disclosures. Increased readability may compensate for less-sophisticated investors’ lack of task-specific knowledge to help create a “level playing field” with more-sophisticated investors. With the exception of ease of understanding, the same results are found for the other variables examined. The means of less-sophisticated NPIs using high-readability disclosures and more-sophisticated NPIs using low-readability disclosures are nearly identical on overall management credibility (4.05 vs 4.04, respectively), management competence (3.97 vs 3.92, respectively), and management trustworthiness (4.17 vs 4.16, respectively). These results mirror the findings presented earlier on equity valuations, bolstering the proposition that plain English pulls less-sophisticated NPIs to a similar standing as more-sophisticated NPIs.

## Discussion and conclusion

Research indicates that financial reports have become too complex for NPIs to grasp (Miller, 2010), requiring at least a graduate degree to understand. A lack of task-specific knowledge contributes to less-sophisticated investors’ conservatism (Coram, 2010). The SEC, which wants to level the playing field for NPIs, has responded by requiring that risk disclosures in the 10-K be written using plain English principles. Some research questions the seemingly intuitive belief that plain English communications are “better,” and can create a level playing field for NPIs, particularly among varying levels of NPI sophistication. This study provides experimental evidence on this issue. This study hypothesizes that less- and more-sophisticated NPIs’ equity assessments, ease of understanding, and perceptions of management credibility differ when risk disclosures are presented in plain English vs less-readable language.

The significant main effect for both sophistication and readability indicates that less-sophisticated NPIs provide lower equity valuations, and that low-readability disclosures result in lower equity valuations. But, a noteworthy result is that less-sophisticated NPIs report significantly higher valuation estimates when disclosures are more readable, to the extent that they make judgments similar to those of more-sophisticated NPIs with less-readable disclosures. This implies that sophistication level is, in fact, an important consideration in the readability of financial reports, and that plain English affects less-sophisticated investors in such a way as to level the playing field – a primary objective of the SEC’s mandate. Three of the four perception measures follow the same pattern. The means of less-sophisticated NPIs using high-readability disclosures and more-sophisticated NPIs using low-readability disclosures are nearly identical on composite management credibility, management competence, and management trustworthiness. Taken together, these results provide further evidence that plain English disclosures compensate for a lack of task-specific knowledge, and increase credibility perceptions of management among less-sophisticated NPIs.

Readability also affects the perceived fluency of disclosures. Surprisingly, this effect is significant for more-sophisticated NPIs only, which is opposite of our expectation. Perhaps less-sophisticated investors do not consciously perceive a difference in understandability afforded by plain English disclosures, or they are less likely to admit a lack of understanding. The lowest perception of fluency was noted by the more-sophisticated NPIs with low-readability disclosures. This may be because these investors are experienced in reading financial information and are thus more likely to recognize the complexity of current financial reporting. They may be more cognizant of the growing chorus of complaints regarding the lack of readability present in financial reporting, and identify “legalese” as such. These explanations are speculative, and encourage additional research to identify the factors that contributed to this result.

In another unexpected result, less-sophisticated NPIs perceive lower management credibility for less-readable disclosures. The opposite effect occurs for more-sophisticated NPIs. Prior research finds more-sophisticated NPIs should be more sensitive to management obfuscation and to punish management for providing low-readability disclosures. More-sophisticated NPIs, who expect low-readability disclosures, may question the sincerity and authenticity of plain English disclosures. Some recent research suggests that difficult-to-process stimuli can actually increase perceived source credibility because the recipient ascribes greater intelligence and favorability to someone who can use big words and complex linguistics (e.g. Alter et al., 2007). If more-sophisticated NPIs are accustomed to seeing disclosures written in “legalese,” they may question the intelligence of the source and reduce their perceived credibility. This supports Tan et al.’s (2014) suggestion that there are boundaries to expected effects on NPI behavior. They found that NPIs with greater sophistication may punish management for strategically disfluent communications, but evidence presented in the present paper indicates that this may not hold if there is not a perceived strategic intent to obfuscate.

For regulators, the question becomes whether the readability of financial disclosures improves investor judgments, and for which investors. Within the recent SEC concept release, “One commenter suggested that current disclosure is too complicated for the everyday person to read and that it should be less duplicative and more straightforward. Another commenter noted the diversity of the investor community and that the Commission’s mandate is to protect all investors” (Securities and Exchange Commission, 2016, p. 48). These comments indicate the importance of acknowledging investors with varying sophistication levels, and the concept release call for comment on the very issue of whom the disclosures should serve. The SEC has long pushed plain English reporting, to provide a more level playing field for NPIs. Our study indicates that it does help level the playing field for the least-sophisticated NPIs, but it also influences the more-sophisticated NPIs. Because reports will be either plain English or not, the implications and consequences must be examined in detail. More research should help us understand and predict the effects and limitations of disclosure presentation style. Is the effect on perceptions of credibility a positive outcome? Does increased readability reduce conservatism for less-sophisticated investors, and is that outcome truly beneficial to the investor in the long term, or does such conservatism act as a protection for overconfidence? Finally, should the SEC worry about this type of investor at all, or assume that efficient markets will effectively constrain and guide them? The SEC calls attention to this in the concept release by stating, “To the extent some investors rely on market prices to efficiently incorporate all public information, rather than relying on disclosures directly, it could be argued that disclosures should be tailored to those users most likely to actively follow a registrant, transact in the registrant’s securities and set the market price” (Securities and Exchange Commission, 2016, p. 58). Future research should examine these questions to guide regulators’ attempts to provide useful information and improve capital market efficiency.

As for all experimental research, these findings face several limitations. The experimental materials are limited, and NPIs would have access to much more information for making equity valuation decisions. We also make no claims of the accuracy of equity assessments or management credibility, and thus offer no normative solutions. Future research that includes additional disclosures and financial information offers the potential to further examine the effect of information overload and readability on NPIs, in conjunction. Lastly, the respondents likely expended less effort on a hypothetical case than they would in an actual investment situation. A controlled experiment involving compensation for investment choices may increase participants’ engagement. Despite these limitations, the results, obtained from a relatively large sample of active NPIs, provide important information for regulators as well as opportunities for future research.

## Figures

#### Figure 1

Equity valuation by readability and sophistication level

#### Figure 2

Perceptions of understanding by readability and sophistication level

#### Figure 3

Management credibility by readability and sophistication level

## Table I

Demographic data

 n Mean (SD) Range Number of years investing 358 23.38 (13.09)0-62 years Average no. of trades per year 359 10.23 (14.25)0-120 Average age 359 65.03 (10.06)23-90 Gender=58% Male (207/359); 42% Female (152/359)Certified public accountant or certified financial analyst=4% (16/359) Frequency % Highest degree earned High school 5 1.4 Some college/associate’s degree 42 11.7 Bachelor’s degree 141 39.3 Graduate degree 171 47.6 Total 359 100 Field of highest degree Accounting and finance 36 10.0 Other business 64 17.8 Healthcare 50 13.9 Engineering 39 10.9 Education 36 10.0 Physical sciences 31 8.6 Other college degree 98 27.3 High school diploma 5 1.5 Total 359 100

## Table II

Non-professional investors’ equity valuation judgments

 Panel A: descriptive statistics: mean (standard deviation); range Readability Sophistication Low (n=169) High (n=190) Overall (n=359) Less financially sophisticated (n=182) $25.08 (3.09)$11.25-$29.00n=84$25.82 (1.85)$18.00-$30.00n=98 $25.48 (2.52) More financially sophisticated (n=177)$25.75 (2.02)$18.63-$31.00n=85 $26.09 (1.80)$20.00-$30.00n=92$25.93 (1.91) Overall (n=359) $25.42 (2.62)$25.96 (1.83) $25.70 (2.25) Panel B: ANOVA tests of between-subjects effects Type III SS df MSE F p-Value Financial Sophistication 19.35 1 19.35 3.89 0.02 Readability 26.25 1 26.25 5.28 0.05 Sophistication × readability 3.72 1 3.72 0.75 0.39 Error 1,766.25 355 4.98 Panel C: pairwise comparisons based on estimated marginal means 95% confidence interval for difference Means Difference ($) SE Lower bound Upper bound H1 – sophistication: less vs more $25.46 vs$25.92 0.47 0.24 −0.929 −0.001 H2 – readability: low vs high $25.42 vs$25.96 0.54 0.24 0.078 1.006

Note: DV=What is your estimate of the most likely stock price for the year ended December 31?

## Table III

Non-professional investors’ perceptions of understanding

 Panel A: descriptive statistics: mean (standard deviation); range Readability Financial sophistication Low (n=169) High (n=190) Overall (n=359) Less financially sophisticated (n=182) 5.17 (1.26)2-7 n=84 5.44 (1.11)2.7n=98 5.31 (1.18) More financially sophisticated (n=177) 4.88 (1.58)1-7n=85 5.36 (1.18)2-7n=92 5.13 (1.40) Overall (n=359) 5.02 (1.43) 5.40 (1.14) 5.22 (1.30) Panel B: ANOVA tests of between-subjects effects Type III SS df MSE F p-Value Financial sophistication 2.97 1 2.97 1.79 0.18 Readability 12.52 1 12.52 7.56 0.01 Sophistication × readability 0.93 1 0.93 0.56 0.45 Error 355 Panel C: t-Tests of ease of understanding Mean difference of low vs high readability t-Statistic p-Value Less sophisticated 0.27 1.55 0.13 More sophisticated 0.48 2.26 0.03

Note: DV=disclosure was easy to understand (1=strongly disagree, 7=strongly agree)

## Table IV

Non-professional investors’ perceptions of management credibility

 Panel A: descriptive statistics: mean (standard deviation); range Readability Financial sophistication Low (n=169) High (n=190) Overall (n=359) Less financially sophisticated (n=182) 3.88 (0.53)2.0-5.2n=84 4.05 (0.48)2.3-5.3n=98 3.97 (0.51) More financially sophisticated (n=177) 4.04 (0.50)2.8-5.5n=85 3.97 (0.57)2.5-5.2n=92 4.0 (0.54) Overall (n=359) 3.96 (0.52) 4.01 (0.53) 3.98 (0.52) Panel B: ANOVA tests of between-subjects effects Type III SS df MSE F p-Value Financial sophistication 1.1 1 1.17 0.43 0.51 Readability 0.16 1 0.16 0.61 0.44 Sophistication × readability 1.29 1 1.29 4.76 0.03 Error 96.13 355 0.27 Panel C: simple main effects tests df MSE F p-Value Readability within low sophistication 1 1.20 4.44 0.04 Readability within high sophistication 1 0.26 0.97 0.32 Financial sophistication within low readability 1 1.03 3.81 0.05 Financial sophistication within high readability 1 0.34 1.26 0.27

Note: DV=composite measure of credibility

## Table V

Non-professional investors’ perceptions of management competence

 Panel A: descriptive statistics: mean (standard deviation); range Readability Financial sophistication Low (n=169) High (n=190) Overall (n=359) Less financial sophistication (n=182) 3.68 (0.76)1.0-5.3n=84 3.97 (0.69)2.3-6.3n=98 3.84 (0.74) More financial sophistication (n=177) 3.92 (0.71)1.7-5.3n=85 3.88 (0.78)2.3-5.7n=92 3.90 (0.74) Overall (n=359) 3.80 (0.74) 3.92 (0.74) 3.86 (0.74) Panel B: ANOVA tests of between-subjects effects Type III SS df MSE F p-Value Financial sophistication 0.45 1 0.45 0.83 0.36 Readability 1.36 1 1.36 2.51 0.11 Sophistication × readability 2.29 1 2.29 4.24 0.04 Error 192.28 355 0.54 Panel C: simple main effects tests df MSE F p-Value Financial sophistication within low readability 1 −2.26 4.17 0.04 Financial sophistication within high readability 1 0.38 0.70 0.40 Readability within low sophistication 1 −0.28 6.71 0.01 Readability within high sophistication 1 −0.04 0.11 0.74

Note: DV=competence of management

## Table VI

Non-professional investors’ perceptions of management trustworthiness

 Panel A: descriptive statistics: mean (standard deviation); range Readability Financial sophistication Low (n=169) High (n=190) Overall Less financial sophistication (n=182) 4.08 (0.54)3.0-5.7n=84 4.17 (0.66)2.0-6.3n=98 4.13 (0.61) More financial sophistication (n=177) 4.16 (0.52)2.7-5.7n=85 4.07 (0.62)1.0-6.0n=92 4.12 (0.57) Overall (n=359) 4.12 (0.539) 4.12 (0.64) 4.12 (0.59) Panel B: ANOVA tests of between-subjects effects Type III SS df MSE F p-Value Financial sophistication 0.00 1 0.004 0.01 0.92 Readability 0.00 1 0.002 0.00 0.94 Financial sophistication × readability 0.69 1 0.69 1.97 0.16 Error 123.90 355 0.35

Note: DV=trustworthiness of management

## Table AI

Genyx income statement

Statement of operations
Historical-year ended December 31
In thousands, except share and per share data 2009 2008 2007 2006 2005
Sales $95,204$85,170 $60,640$46,246 $36,122 SG&A 77,005 70,239 52,362 44,230 32,000 Income (loss) from operations 15,499 14,931 8,278 2,016 2,848 Net income (loss) per common share Basic$0.97 $0.96$0.60 $0.15$0.23
Weighted average common shares outstanding
Basic 14,169,140 13,833,590 12,181,221 11,853,732 11,032,449

## Table AII

Genyx balance sheet

Consolidated balance sheet
Historical as of December 31
In thousands 2009 2008 2007 2006 2005
Cash $16,278$15,751 $12,425$8,650 $7,239 Accounts receivable 24,650 16,378 11,772 9,127 5,576 Total current assets 46,629 32,771 24,848 18,337 13,417 Non-current assets 16,557 16,205 15,336 13,588 10,237 Total assets 63,186 48,976 40,184 31,925 23,654 Current liabilities 16,224 15,300 12,708 9,628 7,113 Non-current liabilities 919 820 203 183 204 Total liabilities 17,143 16,120 12,911 9,811 7,317 Total shareholders’ equity 46,043 32,856 27,273 22,114 16,337 ## Table AIII Genyx statement of cash flows Statement of cash flows Historical-year ended December 31 In thousands 2009 2008 2007 2006 2005 Operating activities$804 $7,295$4,986 $−296$−842
Investing activities −1,764 −5,030 −4,230 450 267
Financing activities 1,487 1,061 3,019 1,257 9

## Notes

1.

More-sophisticated NPI’s understanding should not be sensitive to readability.

2.

Of the 90 companies reviewed, 86 included a factor regarding the competitive environment, 56 included a factor on sensitive client data, and only 15 included a factor on internal control weaknesses.

3.

Our own review of risk disclosures from a random sample of 100 companies from the Fortune 500 companies for the period 2005-2008 finds that they are primarily written in what is often called legalese, and lack the readability intended by the SEC mandate. We calculated the Flesch Reading Ease (Flesch, 1948) score for the risk disclosure section of this random sample. The average grade level necessary to comprehend the 10-Ks is 17.0, which is representative of post-graduate study.

4.

The differences in mean age and mean years investing are not statistically significant.

## Appendix 1. Information provided to all conditions

Please review the information for Genyx, a fictitious US company. Assume that you do not own stock in the company yet, but are considering whether to invest in it.

Genyx information appears below in the following order:

• Short summary about the company – provided by management.

• Three basic audited financial statements – provided by management.

• Three company risk factors – provided by management.

• Analyst estimates of earnings per share (EPS) and stock price (publicly available information).

Note: Genyx’s CPA firms have provided unqualified (clean) opinions on the financial statements and on internal control. The opinion letters are not included in these materials to save space.

### Summary

Genyx Incorporated was founded in 2003 to provide human resource and payroll services to technology companies. The company grew successfully and went public in 2005 – it is currently traded on the New York Stock Exchange and the share price since going public has ranged from $14.00-$28.18.

Table AI

Table AII

Table AIII

## Appendix 2. Risk factor disclosure in low vs high readability

### Risk factor 1 – less readable

We must continue to differentiate our services from those of our competitors; we operate in an evolving industry that requires us to anticipate changes in client preferences and deliver services that demonstrate value to our clients. We operate in a highly competitive environment and in an industry that is subject to significant ongoing changes from market pressures brought about by client demands, as well as business consolidations, strategic alliances, legislative reform, and marketing practices. These factors require us to differentiate our services by anticipating changes in client preferences and delivering services that demonstrate value to our clients. Failure to anticipate changes in client preferences and deliver services that demonstrate value to our clients can affect our ability to retain or grow our client base, which can adversely affect our results of operations.

### Risk factor 1 – more readable (plain English)

Our sales and profits depend on our competitors and clients’ behavior. We operate in a constantly changing industry where many different events occur that might hurt our sales and profits. We need to be aware of market trends and respond quickly to our competitors and clients’ behaviors. If we do not keep up with the competition, we could lose clients, sales, and profits.

### Risk factor 2 – less readable

We would be adversely affected if we fail to adequately protect client payroll and other sensitive information. We maintain large amounts of client payroll and other sensitive data about our clients in the ordinary course of our business. Our business therefore depends substantially on our clients’ willingness to entrust us with their human resource related and other sensitive information. Events that negatively affect that trust, including failing to maintain appropriate safeguards to keep sensitive information secure, whether as a result of our action or inaction or that of one of our vendors, could expose us to litigation and other proceedings, fines and/or penalties, any of which could adversely affect our business, operating results or financial condition.

### Risk factor 2 – more readable (plain English)

As part of our daily operations, we receive and retain large amounts of personal information about our clients and their employees. Our clients provide this information to us because they believe we will keep it confidential. If we do not keep their information confidential, we will lose their trust and our reputation might suffer. Additionally, clients could sue us and we could be fined and assessed penalties.

### Risk factor 3 – less readable

The Company’s internal controls may not be sufficient to ensure timely and reliable financial information. As reported under Item 9a of this Form 10-K, the Company’s management completed its assessment of the effectiveness of the Company’s internal control over financial reporting as of December 31, 2009 and based on that assessment, concluded that the Company maintained effective internal control over financial reporting as of December 31, 2009. The Company’s auditor has issued an attestation report on management’s assessment that expresses unqualified opinions on management’s assessment and on the effectiveness of the Company’s internal control over financial reporting. Still, however, the Company’s growth continues to place stress on its internal controls, and there can be no assurance that the Company’s control procedures will continue to be adequate. The effectiveness of the Company’s controls and procedures may be limited by a variety of risks, including, among other things, faulty human judgment, simple errors, omissions and mistakes, collusion of two or more people or inappropriate management override of procedures. If the Company fails to have effective internal controls and procedures for financial reporting in place, it could be unable to provide timely and reliable financial information.

### Risk factor 3 – more readable (plain English)

In order to produce timely and reliable financial reports, we need effective internal controls. We completed a formal assessment of our internal control system and found it to be effective. Our auditors agree. If our existing controls weaken, we may be unable to prepare timely and reliable financial information. This could happen because of our growth, human errors and mistakes, collusion, or management override.

## Appendix 3. Management Credibility Questionnaire (competence and trustworthiness)

1. I believe that Genyx management is very competent at providing risk factor disclosures.

2. I believe that Genyx management has little knowledge of the factors involved in providing useful risk factor disclosures.

3. I believe that few people are as qualified as Genyx management to provide useful risk factor disclosures about Genyx.

4. I believe that Genyx management is very trustworthy.

5. I believe that Genyx management is very honest.

6. I believe that Genyx management may not be truthful in their risk factor disclosures.

Scale=1 (strongly disagree) to 7 (strongly agreed).

## References

Alter, A. and Oppenheimer, D. (2009), “Uniting the tribes of fluency to form a metacognitive nation”, Personality and Social Psychology Review, Vol. 13 No. 3, pp. 219-235.

Alter, A., Oppenheimer, D., Epley, N. and Eyre, R. (2007), “Overcoming intuition: metacognitive difficulty activates analytic reasoning”, Journal of Experimental Psychology, Vol. 136 No. 4, pp. 569-576.

Barber, B.M. and Odean, T. (2000), “Trading is hazardous to your wealth: the common stock investment performance of individual investors”, Journal of Finance, Vol. 55 No. 2, pp. 773-806.

Bernstein, D. (2006), “Do investors gain from new SEC disclosure?”, International Financial Law Review, Vol. 25 No. 6, pp. 32-34.

Birnbaum, M. and Stegner, S. (1979), “Source credibility in social judgment: bias, expertise, and the judge’s point of view”, Journal of Personality and Social Psychology, Vol. 37 No. 1, pp. 48-74.

Blaskovich, J., Davis, C. and Taylor, E. (2012), “Enterprise risks, rewards, and regulation”, Journal of Applied Business Research, Vol. 28 No. 4, pp. 563-579.

Blazzard, N. and Hasenauer, J. (1998), “This is our plea for plain English”, National Underwriter, Vol. 102 No. 39, pp. 16-19.

Bloomfield, R. (2002), “The ‘incomplete revelation hypothesis’ and financial reporting”, Accounting Horizons, Vol. 16 No. 3, pp. 233-243.

Campbell, N. (1999), “How New Zealand consumers respond to plain English”, Journal of Business Communication, Vol. 36 No. 4, pp. 335-361.

Coram, P. (2010), “The effect of investor sophistication on the influence of nonfinancial performance indicators on investors’ judgments”, Accounting and Finance, Vol. 50 No. 2, pp. 263-280.

Courtis, J. (1995), “Readability of annual reports: western versus Asian evidence”, Accounting, Auditing and Accountability Journal, Vol. 16 No. 64, pp. 4-17.

Courtis, J. (1998), “Annual report readability variability: tests of the obfuscation hypothesis”, Accounting, Auditing and Accountability Journal, Vol. 11 No. 4, pp. 459-471.

Courtis, J. (2004), “Corporate report obfuscation: artefact or phenomenon?”, British Accounting Review, Vol. 36 No. 3, pp. 291-312.

Cui, X. (2016), “Calisthenics with words: the effect of readability and investor sophistication on investors’ performance judgment”, International Journal of Financial Studies, Vol. 4 No. 1, pp. 1-14.

Elliott, W., Hodge, F. and Jackson, K. (2008), “The association between nonprofessional investors’ information choices and their portfolio returns: the importance of investing experience”, Contemporary Accounting Research, Vol. 255 No. 2, pp. 473-498.

Flesch, R. (1948), “A new readability yardstick”, Journal of Applied Psychology, Vol. 32 No. 3, pp. 221-233.

Frederickson, J. and Miller, J. (2004), “The effects of pro forma earnings disclosures on analysts’ and nonprofessional investors’ equity valuation judgments”, Accounting Review, Vol. 79 No. 3, pp. 667-686.

Gunning, R. (1952), The Technique of Clear Writing, McGraw-Hill International Book Company, New York, NY.

Jacoby, J., Nelson, M. and Hoyer, W. (1982), “Corrective advertising and affirmative disclosure statements: their potential for confusing and misleading the consumer”, Journal of Marketing, Vol. 46 No. 1, pp. 61-72.

Johnson, D. (2004), “Has the Secrity and Exchange Commission’s Rule 421 made mutual fund prospectuses more accessible?”, Journal of Financial Regulation and Compliance, Vol. 12 No. 1, pp. 51-63.

Jones, M. and Shoemaker, P. (1994), “Accounting narratives: a review of empirical studies of content and readability”, Journal of Accounting Literature, Vol. 13 No. 1, pp. 142-184.

Koonce, L., Lipe, M. and McAnally, M. (2005), “Judging the risk of financial instruments: problems and potential remedies”, The Accounting Review, Vol. 80 No. 3, pp. 871-895.

Leathers, D. (1992), Successful Nonverbal Communications: Principles and Applications, Macmillan, New York, NY.

Lee, C.-C., Churyk, N. and Clinton, B.D. (2013), “Validating early fraud prediction using narrative disclosures”, Journal of Forensic and Investigative Accounting, Vol. 5 No. 1, pp. 35-57.

Li, F. (2008), “Annual report readability, current earnings and earnings persistence”, Journal of Accounting and Economics, Vol. 45 Nos 2/3, pp. 221-247.

Linsley, P. and Lawrence, M. (2007), “Risk reporting by the largest UK companies: readability and lack of obfuscation”, Accounting, Auditing and Accountability Journal, Vol. 20 No. 4, pp. 620-627.

Linsley, P. and Shrives, P. (2005), “Disclosure of risk information in the banking sector”, Journal of Financial Regulation and Compliance, Vol. 13 No. 1, pp. 205-214.

Linsmeier, T. and Pearson, N. (1997), “Quantitative disclosures of market risk in the SEC release”, Accounting Horizons, Vol. 11 No. 1, pp. 107-135.

McCroskey, J. (1966), “Scales for the measurement of ethos”, Speech Monographs, Vol. 33, pp. 65-72.

Mercer, M. (2004), “How do investors assess the credibility of management disclosures”, Accounting Horizons, Vol. 18 No. 3, pp. 185-196.

Mercer, M. (2005), “The fleeting effects of disclosure forthcomingness on management’s reporting credibility”, The Accounting Review, Vol. 80 No. 2, pp. 723-744.

Miller, B. (2010), “The effect of reporting complexity on small and large investor trading”, The Accounting Review, Vol. 85 No. 6, pp. 2107-2143.

Odean, T. (1999), “Do investors trade too much?”, American Economic Review, Vol. 89 No. 5, pp. 1279-1298.

Pinsker, R. and Wheeler, P. (2009), “The effects of expanded independent assurance on the use of firm-initiated disclosures by investors with limited business knowledge”, Journal of Information Systems, Vol. 23 No. 1, pp. 25-49.

Rennekamp, K. (2012), “Processing fluency and investors’ reactions to disclosure readability”, Journal of Accounting Research, Vol. 50 No. 5, pp. 1319-1354.

Rutherford, B. (2003), “Obfuscation, textual complexity, and the role of regulated narrative accounting disclosure in corporate governance”, Journal of Management and Corporate Governance, Vol. 7 No. 2, pp. 187-210.

Schwarz, N., Strack, F., Bless, H., Klumpp, G., Rittenauer-Schatka, H. and Simons, A. (1991), “Ease of retrieval as information: another look at the availability heuristic”, Journal of Personality and Social Psychology, Vol. 61 No. 2, pp. 195-202.

Securities and Exchange Commission (1998), A Plain English Handbook: How to Create Plain English Disclosure, SEC Office of Investor Education and Assistance, Washington, DC.

Securities and Exchange Commission (2004), “Final rule: securities offering reform (SEC Release No. 33-8501)”, Government Printing Office, Washington, DC, available at: www.sec.gov/rules/final/33-8501.pdf

Securities and Exchange Commission (2005), “Final rule: securities offering reform (SEC Release No. 33-8591)”, Government Printing Office, Washington, DC, available at: www.sec.gov/rules/final/33-8591.pdf

Securities and Exchange Commission (2007), “Speech by SEC Staff: feedback from individual investors on disclosure”, Office of Investor Education and Assistance, Malvern, PA, available at: www.sec.gov/news/speech/2007/spch011907ljs.htm

Securities and Exchange Commission (2016), “Concept Release: business and financial disclosure required by regulation S-K (SEC Release No. 33-10064; 34-77599)”, Government Printing Office, Washington, DC, available at: www.sec.gov/rules/concept/2016/33-10064.pdf

Smith, J. and Kida, T. (1991), “Heuristics and biases: expertise and task realism in auditing”, Psychological Bulletin, Vol. 109 No. 3, pp. 472-489.

Smith, R. and Taffler, R. (1995), “The incremental effect of narrative accounting information in corporate annual reports”, Journal of Business Finance and Accounting, Vol. 22 No. 8, pp. 1195-1210.

Song, H. and Schwarz, N. (2009), “If it’s difficult to pronounce, it must be risky”, Psychological Science, Vol. 20 No. 2, pp. 135-142.

Subramanian, R., Insley, R. and Blackwell, R. (1993), “Performance and readability: a comparison of annual reports of profitable and unprofitable companies”, The Journal of Business Communication, Vol. 30 No. 1, pp. 49-61.

Tan, H.-T., Wang, E. and Zhou, B. (2014), “When the use of positive language backfires: the joint effect of tone, readability, and investor sophistication on earnings judgments”, Journal of Accounting Research, Vol. 52 No. 1, pp. 273-302.

Tan, H.-T., Wang, E. and Zhou, B. (2015), “How does readability influence investors’ judgments? Consistency of benchmark performance matters”, The Accounting Review, Vol. 90 No. 1, pp. 371-393.

Victoravich, L. (2010), “Overly optimistic? Investor sophistication and the role of affective reactions to financial information in investors’ stock price judgments”, Journal of Behavioral Finance, Vol. 11 No. 1, pp. 1-10.

Winkielman, P., Schwarz, N., Fazendeiro, T. and Reber, R. (2003), “The hedonic marking of processing fluency: implications for evaluative judgment”, in Musch, J. and Klauer, K.C. (Eds), The Psychology of Evaluation: Affective Processes in Cognition and Emotion, Erlbaum, Mahway, NJ, pp. 189-217.

You, H. and Zhang, X.-J. (2009), “Financial reporting complexity and investor reaction to 10-K information”, Review of Accounting Studies, Vol. 14 No. 4, pp. 559-586.