Annual report readability and firms’ investment decisions

Dau, Nam Huong; Nguyen, Duy Van; Diem, Hai Thi Thanh Article Annual report readability and firms’ investment decisions Cogent Economics & Finance Provided in Cooperation with: Taylor & Francis Group Suggested Citation: Dau, Nam Huong; Nguyen, Duy Van; Diem, Hai Thi Thanh (2024) : Annual report readability and firms’ investment decisions, Cogent Economics & Finance, ISSN 2332-2039, Taylor & Francis, Abingdon, Vol. 12, Iss. 1, pp. 1-10, https://doi.org/10.1080/23322039.2023.2296230 This Version is available at: https://hdl.handle.net/10419/321397 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Cogent Economics & Finance ISSN: 2332-2039 (Online) Journal homepage: www.tandfonline.com/journals/oaef20 Annual report readability and firms’ investment decisions Nam Huong Dau, Duy Van Nguyen & Hai Thi Thanh Diem To cite this article: Nam Huong Dau, Duy Van Nguyen & Hai Thi Thanh Diem (2024) Annual report readability and firms’ investment decisions, Cogent Economics & Finance, 12:1, 2296230, DOI: 10.1080/23322039.2023.2296230 To link to this article: https://doi.org/10.1080/23322039.2023.2296230 © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. Published online: 15 Jan 2024. Submit your article to this journal Article views: 3415 View related articles View Crossmark data Citing articles: 3 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=oaef20 FINANCIAL ECONOMICS | RESEARCH ARTICLE Annual report readability and firms’investment decisions Nam Huong Dau a , Duy Van Nguyen b and Hai Thi Thanh Diem c a Center for International Knowledge Sharing, Ho Chi Minh National Academy of Politics, Hanoi, Vietnam; b Faculty of Business and Economics, Phenikaa University, Hanoi, Vietnam; c FPT University, Greenwich Vietnam, Hanoi, Vietnam ABSTRACT An easy-to-read report may carry positive information for decision making and conversely. In that spirit, this paper investigates the relationship between the readability of companies’annual reports, defined as the easiness to read, understand, and extract information from the reports, and investment decisions of Singapore companies. Empirical results with an DGMM analysis on 251 domestic companies listed on the Singapore Stock Exchange (SGX) show a positive relationship between the readability of annual reports this year and investments next year. As such, reports’readability can serve as a signal significant to predict companies’future investments. Our findings are consistent with signaling theory and contribute significantly to the literature for empirical investigations on the relationship of annual reports’readability and firms’ investment decisions. ARTICLE HISTORY Received 10 August 2023 Revised 24 October 2023 Accepted 13 December 2023 KEYWORDS Readability; investment decision; investment volume; annual report; Singapore stock exchange REVIEWING EDITOR David McMillan, University of Stirling, UK SUBJECTS Finance; Corporate Finance; Investment & Securities JEL A1; G1; M40 1. Introduction Annual reports, which are legally required to be published by all listed companies, are a means of communication of companies’leadership to capital market participants such as investors, creditors, and other stakeholders (Ertugrul et al., 2017); they are a critical source of information for the latter. The related role of the reports’readability, defined as the ease of understanding given information based on a report (Barnett & Leoffler, 2016), is underlined by various research findings (Cazier & Pfeiffer, 2016,2017; Huddart et al., 2007; Lim et al., 2018; Loughran & Mcdonald, 2014; Yu & Miller, 2010). The readability of the annual report is found to have a significant impact on the effective communication of information to stakeholders (Loughran & Mcdonald, 2014). There are also evidences that investors and/or stakeholders rely on the information in annual reports to make decisions (buy or sell, invest or not, lend or control lending) (Cazier & Pfeiffer, 2016,2017; Huddart et al., 2007; Lim et al., 2018). Detailed reports with low degree of readability may imply less information and confusion for readers; this can limit readers’judgment and evaluation, and hence decision-making ability (Li, 2008; Lim et al., 2018; You & Zhang, 2008; Yu & Miller, 2010). According to signaling theory and related research, through annual reports, a company’s owners and managers may disclose some signals of business strategy (Lim et al., 2018; You & Zhang, 2008). According to studies related to reading comprehension, readability affects the ability to understand published/disclosed information, which in turn impacts readers’judgments (Kintsch & van Dijk, 1978; Masson & Waldron, 1994; Rennekamp, 2012). Intuitively, the clearer and easier the information from the ß2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. CONTACT Hai Thi Thanh Diem [email protected],[email protected] FPT University, Hanoi 12116, Vietnam COGENT ECONOMICS & FINANCE 2024, VOL. 12, NO. 1, 2296230 https://doi.org/10.1080/23322039.2023.2296230 report is to read and comprehend, the better it is for investors as well as other stakeholders to understand the firm performance and make more accurate decisions (Lim et al., 2018; Rennekamp, 2012; Shah & Oppenheimer, 2007). In this paper, we conduct empirical research, with data of Singapore financial market, to investigate the relationship between annual report readability and firm investment decisions. In this research we define the readability of a report as the easiness to read the report; a low readability level means the report is difficult to read, and conversely, a higher level of readability means that the report is more readable, i.e. easier to read and understand. Specifically, this paper seeks to answer the questions what the relationship between a firm’s annual report readability and its investments is, and further if the readability of the report is a signal, which can help investors and/or stakeholders predict the company’s future investment viability—hence, it is a channel that readers (investors, stakeholders) can rely on to react promptly and make decisions, including investment ones. Singapore is a country with a developed economy and an advanced financial market; also, it can be considered a financial center of the Asia Pacific region (Chow & Pei, 2018). The Global Financial Centers Index updated to 2021 ranks Singapore as the fifth most influential financial center in the world, after New York, London, Shanghai, and Hong Kong (Wardle & Mainelli, 2021). The Singapore Stock Exchange (SGX) is the largest internationalized exchange in Asia with more than 40% of companies listed on the platform originating outside of Singapore; it has about 800 companies and is also the largest Real Estate Investment Trust (REIT) after Japan (Chow & Pei, 2018). Therefore, the listed companies’annual reports are deemed to meet high standards of completeness and transparency of a developed financial market (Au, Thompson, & Yeung 2006). Consequently, corporate reporting serves as a significant and meaningful indicator for stakeholders. Conducting an empirical analysis, using the Flesh Kincaid Grade (FKG) and Simple Measure of Gobbledygook (SMOG) as indicators for readability, where lower the values of the indicators indicating a higher level of readability of a document, and employing Difference Generalized Method of Moments (DGMM) model to address endogeneity and data of domestic companies listed on the Singapore Stock Exchange (SGX), we find that report readability carries signals significant for predicting firms’investments; more specifically, when reports are considered easy to read, companies invest more and conversely; as such investors can rely on the readability of the report as a predictor for investment viability. Conducting a quantile regression, we detect a positive relationship between the reports’readability and companies’investments with medium to high investment volumes (insignificant in a company with a low investment volume), thereby confirming the above finding. To our best knowledge, this paper, with the research questions posed and empirical analysis with data from SGX, fills in the gap of literature for empirical investigations on the relationship of report readability and investment decisions of enterprises. This paper can be considered a novel study on annual report readability and investment decisions. 2. Literature review 2.1. Readability of reports and investment decisions The signaling theory Spence (1978) proposes explanation of the behavior of two parties who differ in ability to receive and transmit information. A party as an informant (the party making a report) must choose, whether or not, to provide information that is complete and understandable (Washburn, 2017) and this can be done either intentionally or unintentionally. On the one hand, recipients of information from the report including shareholders, investors, and other stakeholders, choose how to interpret the signals they receive (Washburn, 2017). The signal given by the manager will be the information for investors and stakeholders to make their decisions. For listed companies, the publication of annual reports is a way to reduce information asymmetry in the market (Asare & Wright, 2012). However, these disclosures can positively or negatively impact information users, including shareholders, investors, depending on if the signals they carry are positive or negative (Connelly et al., 2011). Previous studies have primarily employed signal theory to explain the relationship between annual report readability and financial performance (Eugene-Baker & Kare, 1992; Dalwai et al., 2021) or annual report readability and 2 N. H. DAU ET AL. earnings management (Li, 2008; Lo et al., 2017). Therefore, research utilizing signal theory to elucidate the relationship between annual report readability and investment decisions remains limited. Related theoretical and empirical results also show that readability and related aspects can influence the perception and decisions of a reader (Lim et al., 2018; Rennekamp, 2012; Shah & Oppenheimer, 2007). In making judgments, fluent, or easy to process, information is weighed more heavily than disfluent information (Shah & Oppenheimer, 2007). Related research also finds that more readable disclosures, which facilitate processing fluency, can affect investors’valuation judgments. Intuitively, processing fluency from a more readable report can serve as a subconscious hint and reinforce investors’beliefs whether they should rely on the report; as more specific finding, small investors react more strongly to more readable reports, with more positive changes in valuation judgments when news is good and conversely (Rennekamp, 2012). Consistently, if it is easy for a recipient reads and understands information in a report, it can be considered a positive signal helpful for the reader’s decision making (Connelly et al., 2011). Complex information requires investors and other stakeholders to make more conscious efforts. This undermines recipients’understanding and ability to assess a company’s prospects based on the information and hence may dampen their decision-making capacity (Lee, 2012; Lim et al., 2018). More directly related to annual report readability, there are findings that easy-to-read reports help readers make timely decisions (Libby, Bloomfield, & Nelson, 2001; Grossman, 1980); and hard-to-read reports distract or confuse the readers (Courtis, 1998; Lim et al., 2018; Rutherford, 2003). Intuitively, the information from annual reports helps investors and stakeholders with detailed, transparent information, thus strengthening their understanding of a company’s potential and position (Lee, 2012; Lim et al., 2018). Some studies also show that the readability of reports is related to firm performance (Eugene Baker & Kare, 1992; Biddle et al., 2009; Courtis, 1995,1998; Hassan et al., 2018; Lee & Tweedie, 1975; Smith et al., 2006); for example, companies with higher annual report readability have higher profits and lower agency costs (Hassan et al., 2018; Smith et al., 2006); companies with annual reports that are easier to read have positive earnings which are more persistent (Li, 2008). Readability and tone ambiguity of a firm’s financial disclosures are also proved to be related to managerial information hoarding; and less readable and more ambiguous annual reports are associated with an increased cost of external financing (Ertugrul et al., 2017). Some other studies suggest that reports’readability can affect the tightening of enterprises’borrowing (Huddart et al., 2007). 2.2. Measuring annual report readability For the sake of clarity, first let’s make it clear that the readability of a report is understood as the easiness or difficult to read, understand and extract information from the report. There are different indicators of report readability as proposed by research in computational linguistics, such as Flesch Kincaid Grade (FKG) and Simple Measure of Gobbledygook (SMOG) Index; each indicator has a different calculation method. As will be clarified below, more accurately, FKG and SMOG are in fact interpreted directly as measures of non-readability of reports, in the sense that they represent how difficult a text is to read; more specifically, the higher the values of the indicators, the lower the readability of a report, and conversely (The FKG, SMOG is larger, the annual report is more difficult to read). The authors describe and use both indicators in this research. The FKG index as an indicator for report readability (Li, 2008; Solnyshkina et al., 2017; Worrall et al., 2020), also known as the Kincaid index, is calculated using the following formula: FKG ¼0:398 words sentences  þ11:8syllables words  −15:59 FKG indicates the number of years of education generally required to understand a report. Hence, the higher the FKG score for a report, the more difficult it is to read the report. As such, we can also interpret FKG as a measure of non-readability of reports, where a higher FKG score of a report implies a higher non-readability level the report assumes. The SMOG (Simple Measure of Gobbledygook) index was introduced by Mc Laughlin (1969) to assess readability. It is a two-variable formula, as followed: COGENT ECONOMICS & FINANCE 3 SMOG ¼1:0430 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 30 complex words sentences  sþ3:1291 where complex words are words with 3 or more syllables (Mc Laughlin, 1969). The index is an estimate of the education level a reader needs to ensure a thorough comprehension of a text, for example SMOG Grades 13-16 indicate the need for college education (Mc Laughlin, 1969). Like the FKG index, SMOG of a report can also be interpreted more directly as a measure of non-readability - the reading difficulty - of the report. 3. Method 3.1. Research hypothesis and model To address the research questions and decipher the relationship between the readability of firms’annual reports and firms’investments, the paper formulates the following research hypothesis. Hypothesis: A higher readability of a company’s annual reports predicts a higher future investment level of the company, and conversely. To test the hypothesis and address the research question posed, we consider the following the econometric model: Investmentit ¼aiþcNonReadabilityit−1þdControl Variablesit−1þeit (1) where, dis a coefficient vector, aiis time-invariant unobserved variable (firm fixed effect), and eit,is error term. The variables are described in detail in Table 1. In this study, investment is represented by the CAPEX index (Capital Expenditure). The FKG and SMOG in this study represent the non-readability of the annual reports; specifically, the lower the FKG and SMOG values, the more readable a report is and conversely, as elaborated above. As such, an expected negative sign associated with each of these variables is interpreted as a positive relationship between the readability and the dependent variable (investments). The leverage, revenue growth, and firm size are the control variables in this research. Leverage is measured by the ratio of debt to total assets. The higher the leverage of a company, the greater the risks associated with interest expenses if the company is not as efficient as expected (Nguyen, Nguyen et al., 2021). Therefore, with a larger leverage ratio, a company would tend to reduce its investment volume to control potential issues related to interest expenses (e.g. an increase in the borrowing costs, a business risk dampening the company’s revenue or an unexpected cost undermining the company’s repayment plan etc.). The logarithm of total assets is the proxy for the firm size in this study. The size of a company may contain information about the company’s development strategy (Nguyen et al., 2020; Nguyen, Ho et al., 2021). Finally, the revenue growth is also included to control the impact of report readability on the investment volume. A higher revenue growth rate of a company may indicate a higher development cycle for the company (Nguyen, Nguyen et al., 2021). This, in general, would help companies to be more confident with their investment decisions. Table 1. The variable definition. Variable name Content Expected Dependent variables (Investment) LnCAPEX ¼Ln (Capital Expenditure) Independent variables (Non-Readability) a FKG Annual report readability (Flesch Kincaid Grade) – SMOG Annual report readability (Simple Measure of Gobbledygook) – Control variables SIZE ¼Ln (Total assets) þ LEV ¼Liability/total assets – GROWTH ¼(Revenue t -revenue t-1 )/revenue t-1 þ aFKG and SMOG are indicators of readability; by definition of the readability, we make clear above for this research, FKG and SMOG are directly measure of non-readability, as the higher the values of these indicators, the more difficult it is to read and understand a document. 4 N. H. DAU ET AL. 3.2. Data Singapore is one of the world’s five largest financial markets. Listed companies on the Singapore Exchange (SGX) are required to provide more stringent information, leading to more reliable data collection. Data is collected for the sample of domestic companies (of Singapore) listed on the Singapore Stock Exchange (SGX) from 2016 to 2021. Out of 443 companies with data collected from the SGX, the research excludes financial companies and companies with incomplete reporting data, namely those with 2 consecutive years without annual reports. After data cleaning and filtering, there are 251 companies remained, and they are all non-financial. The summary statistics of data on readability are described in detail in Table 2. The results show that the FKG index takes values between 0.1 (min) and 10.4 (max); the average value of the FKG index is 3.56. It means that, on average, the level of non-readability/difficulty to read of the documents in the sample is rather low, i.e. the documents are rather easy to read. Similarly, the mean SMOG index of 4.34 also indicates that the annual reports are at a relatively easy level to read and comprehend. The details of comparison between annual report readability indicators over years in Figure 1. The Table 3 describes the variables about firms’characteristics, control variables of the research model. With 251 companies included in the analysis, descriptive statistics of research variables show that the mean of CAPEX is $60 million; i.e. on average, companies tend to invest more instead of withdrawing their investments or selling assets. In addition, the mean of LEV (Liability/total assets) is 0.496, the mean of GROWTH is 0.057 (5.7%) and the mean of SIZE for the whole period is 19.84. Summary statistics of the variables are presented in detail in Table 3. 3.3. Data analysis This study uses panel data of 251 companies collected from the Singapore Stock Exchange (SGX) for the period 2016 to 2021. Conventional Fixed effect model (FEM) and Random effect model (REM) are used to investigate the predictability of annual report readability for companies’investments. The FEM is a further development of Ordinary least squares (OLS) to address the unobserved time-invariant heterogeneities across the individuals; REM estimates both the within-individual and between-individual variances, allowing for handling unobserved heterogeneity and also more generalizable results. There is no relationship between the residuals and the model’s independent variables in the REM. However, both Fixed Effects Model (FEM) and Random Effects Model (REM) have some limitations when it comes to addressing endogeneity, which can lead to less reliable estimation results. Therefore, in this case, the Table 2. Descriptive variables –the indicators (FKG and SMOG) of annual report readability. Variable Obs Mean Std. Dev Min Max FKG 1,530 3.56 2.67 0.10 10.40 SMOG 1,530 4.34 2.28 3.2 10.8 Figure 1. The Annual report readability. COGENT ECONOMICS & FINANCE 5 Difference Generalized Method of Moments (DGMM) model is employed. In addition, note that in similar setups, endogenous phenomena are often encountered. Furthermore, the DGMM model employs differencing to eliminate endogeneity issues without focusing on identifying strictly exogenous variables. Therefore, it can be said that DGMM is a straightforward method suitable for research data with a small T and a large N. This model is used to address the endogeneity, by adding the lagged variable of the dependent variable to the regression model and taking the first difference. (see footnote for a detailed explanation of the mechanism). 1 We also conduct quantile regressions at 25%, 50%, 75%, and 95% to examine the effect of report readability on investment decisions in firms with different levels of investment. 4. Results As mentioned, the regression analysis with DGMM is used in this study to deal with endogeneity; however, the two models FEM and REM are also performed to compare with the DGMM. Table 4 summarizes the regressions results of alternative models: (1) and (2) are FEM models, (3) and (4) are REM models, Table 3. Descriptive variables –firms’characteristics –control variables of the model. Variable Mean Std. Dev Min Max CAPEX 60 300 0 5000 LEV 0.496 0.408 0.022 8.875 GROWTH 0.057 0.562 −1.000 8.962 SIZE 19.846 1.737 11.249 24.796 Observation ¼1,530 Table 4. The result of regressions –FEM, REM an DGMM models: Estimation of the relationship between measures of readability (FKG, SMOG) and firms’investments. (1) (2) (3) (4) (5) (6) Variables Fixed effect model Fixed effect model Random effect model Random effect model DGMM DGMM LnCAPEX t-1 0.657 0.652 (0.0790) (0.0784) FKG t-1 −0.0281 −0.0275 −0.0405 (0.0106) (0.0106) (0.0115) SMOG t-1 −0.0239 −0.0264 −0.0448 (0.0121) (0.0121) (0.0125) LEV t-1 −0.566 −0.562 −0.405 −0.400 −3.199 −3.113 (0.172) (0.173) (0.161) (0.161) (1.192) (1.195) GROWTH t-1 0.292 0.291 0.287 0.287 0.221 0.229 (0.0529) (0.0530) (0.0528) (0.0528) (0.0676) (0.0674) SIZE t-1 0.454 0.440 0.790 0.787 0.412 0.413 (0.134) (0.134) (0.0632) (0.0633) (0.0933) (0.0927) Constant 6.334 6.603 −0.456 −0.382 −1.380 −1.298 (2.662) (2.663) (1.261) (1.263) (0.653) (0.640) Observations 1,229 1,229 1,229 1,229 1,223 1,223 Number of firms 251 251 251 251 251 251 Hausman test 0.000 AR(1) 0.000 0.000 AR(2) 0.803 0.858 Hansen test 0.377 0.388 Standard errors in parentheses. p<0.01,  p<0.05, p<0.1. 1 The feature of the DGMM model is to add the lagged variable of the dependent variable to the regression model. Specifically, the initial equation in DGMM model is as follows:. Yit ¼b0þti ðÞ þb1Yit−1þb2Xit þeit (2). Eq (2) is transformed into first-difference form to suppress potential fixed effects assumed in panel data. DYit ¼b1DYit−1þb2DXit þDeit (3). Where:. tit ¼miþeitDtit ¼miþeit ðÞ −miþeit−1 ðÞ ¼Deit Taking the difference would help eliminate the endogeneity problem in the model. 6 N. H. DAU ET AL. and (5) and (6) are DGMM models; in each pair of models, either FKG or SMOG is the used as the variable proxied for the report readability. The Hausman test with p-value ¼0.000 shows FEM is more suitable than REM. Therefore, correlation of residuals and independent variables occurs. With p-value of AR(1) <0.05 and AR(2) >0.05, the DGMM shows the autocorrelation is corrected. The regression results show a consistency between DGMM and FEM or REM models. FKG t-1 and SMOG t-1 are both significant signals to predict investment volumes (b FKG ¼−0.0405 and significant at 1%; b SMOG ¼−0.0448 and significant at 1%). The results also show that LEV t-1 has a negative effect on investment (b LEV ¼−3.199 and significant at 1%); GROWTH t-1 has a positive effect on investment (b GROWTH >0 and significant at 1%); SIZE t-1 has a positive impact on investment (b>0 and significant). The FEM, REM, and DGMM regression results are presented in detail in Table 4. The results indicate that the more difficult the annual reports are to read, the lower the investment volumes. In other words, companies will make the decision to invest more when previous annual reports are considered easy to read. With the model setup with lags in explanatory variables, the empirical result can be interpreted that the reports’readability is a signal significant to predict future investments of firms. It can be showed that the annual report’s readability, whether it is easy or difficult to read, may serve as a signal for a company’s investments in the following year. Therefore, the signaling theory provides a good explanation for the relationship between readability and investment decisions. By providing a readable report, a company demonstrates transparent and easily understandable communication of information. Consequently, stakeholders are more likely to perceive positive developments within the company, as positive information tends to be conveyed more clearly than negative information. Therefore, the support of stakeholders for the company’s decisions in the following year is likely to be higher. As a result, the company will find it easier to increase its investments in the next year (Cazier & Pfeiffer, 2016,2017; Lim et al., 2018). Intuitively, reports that are hard to read indicate that the information given is not easy to decipher and it is difficult to connect to the necessary information (Huddart et al., 2007; Lim et al., 2018); the ambiguity of the wording can make investors and stakeholders confused and uncertain about the parameters in the report. This is a negative signal holding back the company from making decision to invest more in the future (Cazier & Pfeiffer, 2016,2017;Huddartetal.,2007;Limetal.,2018). In addition, leverage has a negative impact on investments, indicating that a higher debt ratio makes managers more limited in investment. With the risks coming from payables in general and loans in particular, there are pressures on managers to consider their investments to bring expected firm performance (Ertugrul et al., 2017). The revenue growth has no impact on investments. This result indicates that an increase in revenue is not a cause for companies’investment decisions. Finally, total asset growth has a positive impact on investments. This result suggests that an increase in total assets results in a higher investment level. It can be seen that high or low investment decisions can be differently affected by the readability or difficulty of reading the report. We also use quantile regressions to elaborate our evaluation of the predictability of readability for investment decisions in cohorts of companies with different investment volumes. With quantile regression analysis, the 25% quantiles; 50%; 75% and 95% of the dependent variable (investment decision) are considered in this study. The results show that the reports’readability is a sound predictor of investment decisions in companies with medium and high investment levels. However, taking a closer look, report readability is insignificant as a predictor of the investment volume for companies with a low investment level. Specifically, companies with investment volume in quantile 1 (25%) have no relationship between report’s readability and investment volume (see Table 5). Table 5. Quantile regression. (1) (2) (3) (4) CAPEX q25 q50 q75 q95 FKG t-1 −0.0130 −0.0247 −0.0362 −0.0483 (0.0135) (0.00972) (0.0133) (0.0210) SMOG t-1 −0.0112 −0.0223 −0.0345 −0.0464 (0.0142) (0.0105) (0.0146) (0.0229) Standard errors in parentheses. p<0.01, p<0.05, p<0.1. COGENT ECONOMICS & FINANCE 7