Keywords

1 Introduction

A focal point of corporate financial reporting is the quality of financial statements, specifically, transparency in integrated reports (IR) (Leung et al., 2015). Financial reporting quality refers to the accuracy of a company’s disclosure relating to its position and operations to enable users of reports to make informed rational decisions (Cohen et al., 2004). A key link exists between a company’s financial reporting process, its directors, and the overall quality of financial statements.

Corporate scandals in South Africa such as VBS Mutual Bank and Steinhoff Ltd. have resulted in a credibility crisis in the accounting profession (Yasseen et al., 2020a). More importantly, this has raised concerns over the application of South African corporate governance principles (Rossouw & Styan, 2019). Narrative disclosures in financial reports have been seen as a mechanism to minimise the expectation gap between management and investors (Yasseen et al., 2020b). Narrative disclosures serve as a means of communication which only numerical information may not convey (Rutherford, 2003). IR consist of both regulatory items, such as International Financial Reporting Standards (IFRS) and the King Code IV (KING IV) required narratives which offer an understanding of a company’s financials, as well as non-regulatory or discretionary items. Discretionary narrative disclosures provide a vital way for executives to convey valuable information which investors can use for the assessment of the current and future performance of a company (Leung et al., 2015).

Merkl-Davies and Brennan (2007) found that disclosure strategies are used to provide additional useful information or as an impression management method. The impression management school views these narrative disclosures as a way for management to obfuscate and opportunistically disclose information to serve their interests (Courtis, 2004; Godfrey et al., 2003). One way in which impression management is achieved is through the use of a concealment strategy known as minimal narrative disclosure (MND) (Merkl-Davies & Brennan, 2007).

MND is a concealment strategy which results in the presentation of selective information in the IR (Merkl-Davies & Brennan, 2007). MND in IR is a common strategy employed by companies and additional research is required to determine the effect intentional obfuscation of information has on the perceptions associated with a company (Leung et al., 2015; Merkl-Davies & Brennan, 2007). The primary research question of this paper is: Does the management of JSE-listed companies use MND as part of their impression management strategy?

The main research question was answered using the following hypotheses adapted from the original research performed by Leung et al. (2015):

  • Hp1: Current performance of a company is associated with MND in IR.

  • Hp2: MND in IR is associated with the future performance of a company.

  • Hp3: The financial distress of a company is associated with MND in IR.

This paper is quantitative in nature and consists of a sample of JSE-listed companies. The period covered was the 2018 financial year. Impression management strategies include obfuscation of information, thematic manipulation, visual and structural manipulation, performance comparisons and attribution of organisational outcomes (Merkl-Davies & Brennan, 2007). This paper only focused on the obfuscation of information through MND as an impression management strategy.

2 Prior Literature and Theoretical Framework

Disclosures both financial and narrative which are contained in IR have recently been a topic of great interest in the accounting research space (Yasseen et al., 2017, 2019). Impression management is primarily focused on how a company is presented to others so it can be perceived in the most favourable manner (Leung et al., 2015; Merkl-Davies et al., 2011; Yasseen et al., 2017). Merkl-Davies et al. (2011) consider impression management to be a bias in which a positive perception is created by the manipulation of impressions to others. Impression management is derived from social psychology and focuses on how in a corporate context, the management of a company seeks to present the company in the best possible light (Hooghiemstra, 2000). Within a corporate reporting environment, Brennan & Merkl-Davies (2013) found that impression management encompasses the creation of an impression by a company to appeal to the stakeholders to whom they report. If effective, this impression weakens the quality of IR and stakeholders such as shareholders may allocate capital incorrectly (Brennan & Merkl-Davies, 2013).

Impression management given its ability to affect stakeholders can be viewed as being entrenched as part of agency theory (Leung et al., 2015). Agency theory essentially deals with the divergence of interests of the principal (shareholders) and agent (management) (Jenson & Meckling, 1976). As the management of a company may have competing interests with shareholders as a result of performance incentives, they may engage in impression management by choosing what to disclose and what to omit from company disclosures (Leung et al., 2015). Different impression management strategies may be adopted for different stakeholders (Marcus & Goodman, 1991). Financial information is often communicated using written narratives which are occasionally described as unquantified information (Brennan & Merkl-Davies, 2013). The purpose of accounting narratives in IR is to complement and provide additional context to quantified accounting information. Since the majority of accounting narratives in IR are not subject to an audit process, there is a heightened risk that impression management may be utilised to a greater extent (Brennan & Merkl-Davies, 2013).

The argument made by Merkl-Davies and Brennan (2007) is that selectivity entails ‘omitting’ or ‘including’ specific details of the information. This, in turn, is an element of an impression management strategy which is undertaken by some companies. Merkl-Davies and Brennan (2007) further explain that the studies conducted to date on strategies of ‘concealment’ in discretionary narrative disclosures concentrate on examining the content of these disclosures in which selectivity is used in presenting information. MND is one way in which a concealment strategy in connection with selectivity is used in impression management (Merkl-Davies & Brennan, 2007). MND is the action of omitting information on a company’s weaknesses or prospects in respect of its current or future performance, as well as financial distress if any in voluntary disclosures (Leung et al., 2015). The reasons why MND is employed by companies include but are not limited to management safeguarding their own interests and standing (Leung et al., 2015). In order to sustain its position of power, management can choose not to disclose voluntary yet vital information on both performance and possible opportunities from stakeholders by using their discretion (Conway et al., 2015). Besides self-interest, managers of a company can be led to suppress corporate information purposefully by employing the tactic of minimal disclosure (Cho et al., 2010; Leung et al., 2015).

There have been mixed findings in the studies relating to the link between concealment strategies and company performance. While Subramanian et al. (1993), Courtis (2004), Leung et al. (2015) and Li (2008) have found an inverse relationship between the use of concealment strategies and company financial performance, Courtis (1995), Smith and Taffler (1992), Clatworthy and Jones (2001) and Rutherford (2003) have not concurred and found that no such relationship exists. The conclusion drawn by Sydserff and Weetman (2002) is that management is fair in exhibiting narrative information, irrespective of the financial performance of the company. These findings on the association between current performance and the level of disclosure made by a company formed the basis for the development of Hp1.

By analysing narrative disclosures provided within IR; Tennyson et al. (1990) compared a corresponding sample of financially distressed and non-financially distressed companies and found that a non-financially distressed company is more likely to disclose its internal functions, development, and expansion aspects, but that this strategic information is often withheld by companies in financial distress. Boo and Simnett (2002) similarly found that companies in financial distress will omit or reduce the disclosure of management’s future projections in their IR because of distress and a poor future outlook. Despite the findings from the above studies, Leung et al. (2015) and Miller (2002) argue that financially distressed companies face greater market pressure regarding information on their performance and future outlook given their need for financial assistance from financing institutions. As a result, they would tend to disclose more information in order to avoid the risk of managerial reputation and litigation costs associated with non-disclosure of information but, may engage in MND if there is no need for external funding (Leung et al. 2015; Miller 2002). These findings on the association between financial distress and the level of disclosure made by a company formed the basis of the development of Hp3.

When considering the association between MND conduct and future company performance requirements, MND depends on whether poor performance is temporary or will continue in the future (Leung et al., 2015). Leung et al. (2015) found that where negative performance was persistent, companies were more likely to engage in MND compared to companies that reported positive results in future periods. Barring the findings from Leung et al. 2015, there has not been further studies which consider future performance and MND. The findings on the association between future performance and the level of disclosure made by a company as discussed above formed the basis for the development of Hp2.

The prevalence of MND within IR is thus a common occurrence but concealment of information in IR is not well comprehended (Leung et al., 2015). The pivotal reason non-disclosure in IR requires further understanding is that intentional non-disclosure of material information which is relevant may lead to stakeholders making incorrect decisions based on the information or lack thereof. Preceding impression management studies within the South African context, are focused on agency theory or attribution theory perspectives in Chairmen Statements such as the study carried out by Yasseen et al. (2019) and not on MND in a South African context. This paper explored the voluntary disclosure literature, using disclosure indices and the impression management literature within the South African context. Additionally, this paper focused on the strategy of impression management as a deliberate concealment tool utilised by companies by following the methodology of Leung et al. (2015) conducted in Hong Kong to prove the hypotheses developed and answer the research question. By identifying the disclosures companies omit and their effects on company performance, regulators are provided with a foundation to develop corporate reporting guidelines and possible legislation.

3 Methodology

The research involved the identification of MND companies. MND behaviour was identified by quantifying the level of omission of voluntary narrative disclosure within IR. The voluntary disclosure by companies in their IR was contrasted based on anticipated and actual voluntary disclosure. To contrast the anticipated and expected voluntary disclosures made, a standard level of required narrative disclosures was required on the company’s operating, financial, strategic, and overall performance which shareholders would use to assess the company to make an informed decision when investing (Leung et al., 2015).

This paper used the guidelines applied by the Working Committee of the Annual Report Award Competition of Singapore as the basis for determining the standard level of voluntary narrative disclosure as was done in the paper by Leung et al. 2015. The working committee developed their guidelines based on voluntary disclosure which regulators, professional bodies and industry experts view as exemplary. The guidelines were reviewed for items which are unique to Singapore and may reasonably not be expected for a JSE-listed company to disclose or address. No items were identified which needed to be removed as the guidelines are not country specific.

The comprehensive disclosure guidelines are presented in Annexure 1. A company received a score of 1 if it disclosed information on the specific guideline in its IR and a score of 0 if it did not disclose any information on the guideline in its IR. MND companies were identified as companies with a relatively low overall disclosure score based on the 76 guidelines for disclosure. A company was classified as an MND company if its total disclosure score was less than 41.8. Companies with a disclosure score of 41.8 or more were considered non-MND companies. This differed from the classification parameters set by Leung et al. (2015) for the following reasons:

  • The application of King IV is a requirement of JSE-listed companies, requiring a greater number of disclosures in the IR.

  • When the original paper was performed it was the first year in which IFRS was required in Hong Kong (2005), whereas currently within the South African context, IFRS is a long-standing requirement for companies listed on the JSE.

  • The International Integrated Reporting Council has increased the scope of non-financial disclosures in the IR.

3.1 Research Instrument and Data Analysis

After a company was identified as an MND company, to explore Hp1-Hp3, two multivariate regression analyses were performed. This involved controlling for the costs associated with disclosure, demand for external financing and market competition.

3.2 Hypotheses: Hp1 and Hp3

Regression model 1 adapted from Leung et al. (2015) was used to assess if there was an association between an MND company and its current performance as well as financial distress level as detailed below.

Regression model 1.

$$ \mathrm{MNDCi}=\mathrm{ai}+\mathrm{b}1\mathrm{ROAi}+\mathrm{b}2\mathrm{TOBINQi}+\mathrm{b}3\mathrm{DISTRESSi}+\mathrm{b}4\mathrm{NEWEQUITYi}+\mathrm{b}5\mathrm{DEBTi}+\mathrm{b}6\mathrm{LIQUIDi}+\mathrm{b}7\mathrm{HHIi}+\mathrm{b}8\mathrm{SIZEi}+\mathrm{b}9\mathrm{BIG}4\mathrm{i}+\mathrm{b}10\mathrm{LISTi}+\mathrm{b}11\mathrm{SEGi}+\mathrm{b}12\mathrm{GROWTHi}+\mathrm{b}13\mathrm{LCEOOWNi}+\mathrm{b}14\mathrm{DUALITYi}+\mathrm{b}15\mathrm{PINDi}+\mathrm{INDDUMMIES}+\mathrm{ei} $$

Where:

  • MNDC = a dummy variable for MND companies that equals 1 if the company’s total actual disclosure score was less than 41.8 of the maximum possible scores of the full disclosure checklist and 0 otherwise.

  • ROA and TOBINQ were two measures of current performance.

  • ROA = returns on assets, measured as the ratio of net income (before tax and interest) over total assets.

  • TOBINQ = Tobin’s Q, measured as the ratio of the sum of the market value of equity and total debts over total assets.

  • DISTRESS = financial distress measured by Altman’s Z-score (Altman & La Fleur, 1981). Altman’s Z-score was used as a tool used to measure the risk of insolvency that a company faces through the following calculation:

$$ \mathrm{Z}=1.2\mathrm{X}1+1.4\mathrm{X}2+3.3\mathrm{X}3+0.6\mathrm{X}4+1.0\mathrm{X}5, $$

Where:

  • X1 was the ratio of working capital to total assets.

  • X2 was the ratio of retained earnings to total assets.

  • X3 was the ratio of earnings before interest and taxes to total assets.

  • X4 was the ratio of the market value of equity to the book value of total liabilities.

  • X5 was the ratio of sales to total assets.

Altman’s Z-score is generally interpreted using the following scale (Fig. 1):

Fig. 1
An illustration of Altman’s Z-score model depicts the following range. The Distress Zone is from 0 to 1.8. The Grey Zone is from 1.8 to 3.0. The Safe Zone is from 3.0 to 4.0.

Altman’s Z-score model scale (Image sourced from: https://corporatefinanceinstitute.com/resources/knowledge/credit/altmans-z-score-model/. The Altman’s Z-scores Grey zone has several different interpretations of the Grey Zone (example 1.1 > Grey Zone>2.6). The scale presented in Fig. 1 is used in this Paper)

Full size image

  • NEWEQUITY = a dummy variable that equalled 1 if the company issued more than 5% of common equity during the year.

  • DEBT = the debt ratio of long-term debts to total assets.

  • LIQUID = the liquidity ratio of the sum of cash and short-term investment to total assets.

  • HHI = the Herfindahl Hirschman Index, which is a widely used measure of market competition, calculated as the sum of the squared market share based on net sales in the 2-digit SIC industry.

Other control variables were selected based on previous studies on voluntary disclosure (Ahmed & Courtis, 1999; Lang & Lundholm, 1993) and include;

BIG4 = dummy variable that equalled 1 if a company was audited by a Big-4 auditor, and 0 otherwise. LIST = a dummy variable that equalled 1 if a company was cross listed on an overseas exchange, and 0 otherwise.

SEG, which is the number of business and geographic segments to proxy for business complexity.

GROWTH = which is company growth in total assets.

The researchers included in the model the percentage of CEO ownership (CEOOWN), CEO duality (DUALITY = 1 if the CEO was also Chairman of the company) and the proportion of independent non-executive directors (PIND). Likewise, the researchers included the 2-digit SIC industry dummies (IND_DUMMIES) to control for industry effects on the likelihood of nondisclosure in IR.

A negative coefficient of ROA and TOBINQ suggested that companies with poor performance are more likely to engage in concealment of voluntary narrative information (Leung et al., 2015) and assisted the researchers in answering research Hp1.

A positive coefficient of DISTRESS suggested that a company with a higher insolvency risk would tend to obfuscate investors through minimal narrative disclosures in the IR (Leung et al., 2015). This assisted the researchers in answering Hp3. The results of the control variables also provided valuable insight into the other factors that could influence minimal narrative disclosures in companies despite not answering the hypotheses (Leung et al., 2015).

To measure a company’s current performance, data on a company’s return on assets (ROA) and Tobin’s Q (the ratio of the sum of the market value of equity and total debts over total assets) were gathered. To measure the financial distress of a company, Altman’s Z-score was used.

The t-statistic significance and a confidence level of 95% were used to assess the statistical significance of the variables.

The mean, median, standard deviations and minimum and maximum values were observed for each variable in regression model 1 and for the different categories in the disclosure body namely:

  • Actual disclosure score for the “company overview, objectives and strategy” category.

  • Actual disclosure score for the “operating review” category.

  • Actual disclosure score for the “financial review” category.

  • Actual disclosure score for the “general” category.

These descriptive statistics allowed the researchers to observe whether voluntary disclosure regarding specific categories is high or low and provide insight to users regarding specific topics or information companies choose to be silent about or avoid disclosing.

3.3 Hypothesis: Hp2

Regression model 2 adapted from Leung et al. (2015) was used to assess the association between an MND company and its future performance.

Regression Model 2

To test whether the future performance of non-disclosure companies improves or deteriorates (Hp2), the researcher estimated the following change model:

$$ \Delta \mathrm{PERFi}=\mathrm{ai}+\mathrm{b}1\mathrm{MNDCi}+\mathrm{b}2\Delta \mathrm{Ti}+\mathrm{b}3\Delta \mathrm{DISTRESSi}+\mathrm{b}4\Delta \mathrm{EWEQUITYi}+\mathrm{b}5\Delta \mathrm{DEBTi}+\mathrm{b}6\Delta \mathrm{LIQUIDi}+\mathrm{INDDUMMIES}+\mathrm{ei} $$

Where:

All the change variables are calculated as the differences between the 2019 and 2018 financial years.

  • ΔPERF = the changes in ROA (ΔROA).

  • MNDC = the dummy variable classifying MND and non-MND companies.

  • ΔTA = the changes in total assets over the two years.

  • ΔDISTRESS = the changes in Altman’s Z-score.

  • ΔEWEQUITY = the changes in the issue of new equity over the two years.

  • ΔDEBT = the changes in leverage.

  • ΔLIQUID = the changes in liquidity over the two years. IND_DUMMIES controls the variation in industry performance in the subsequent year.

To assess the future performance of company data relating to changes in ROA, total assets (TA), Altman’s Z-score, equity issued, and liquidity were gathered.

A significant statistical (measured by using a t-statistic significance and a confidence level of 95%) negative coefficient on MNDC for changes in ROA suggested that, compared with the rest of the sample companies, those engaging in concealment through MND in IR would report deteriorating performance in the following year. This suggested that the current poor performance was not transitory. This assisted the researchers in answering Hp2.

3.4 Data Collection

The IR of each company was obtained from each of listed company’s website. The financial data for the variables to be used in the regression were obtained from the JSE, IRESS South Africa Financial reporting database or by manual collection where applicable. The statistical results derived from these methods were completed with the help of a statistician.

3.5 Population and Sample

The population was all companies listed on the JSE. The IR which were investigated related to the 2018 financial year and only narrative disclosures were examined. As part of determining the effect of MND in IR of the 2018 financial year and its association with future performance, the 2019 performance of each sampled company was considered as part of this paper. A purposive sample was used in this research as it was time and cost-effective. This sampling technique was appropriate as the researchers selected the sample size, based on the entities with the highest market capitalisation. Companies with the highest market capitalisation usually have the largest number of shareholders and users of IR. The tactic of impression management through MND will be most apparent if applied by these companies. The IR of the top 100 JSE-listed companies were obtained. However, the sample size was 79 companies (n = 79) as it excluded companies in the financial sector as these companies are subject to distinct regulatory disclosure procedures and companies for which reliable market-related data could not be obtained.

4 Findings

4.1 Disclosure Index Score Results and Key Informational Characteristics of Non-disclosure

The results obtained and represented in Fig. 2 show that just under half of the companies sampled were classified as MND companies as their disclosure score were less than 41.8 (55% of the total). This illustrates that the MND companies did not communicate with stakeholders on 45% or greater of potential narrative disclosure items in IR. These findings suggest that some companies from the selected sample are selective in the information disclosed within IR. This further suggests that the discourse of voluntary items in South Africa is still fairly low, looking at the JSE-listing requirements to prepare IR and that disclosure of voluntary narratives while positive is still developing in South Africa.

Fig. 2
A pie chart depicts the following data. Non-M N D company, 51%. M N D company, 49%.

Proportion of MND companies and non-MND companies

Full size image

Table 1 provides the descriptive statistics for the variables in this paper. The mean overall disclosure scores were 11.278 out of 20 of the possible disclosure items for “Company overview”, “objectives and strategies”; 16.785 out of 29 for “Operating review”; 8.278 out of 16 for “Financial review” and 5.101 out of 11 for “General”. These results suggest that companies tend to disclose more on the overall direction and objectives of the company than on the operations and financial aspects in voluntary disclosure. This may point to the possible use of MND on items which in the short term could impact investor sentiment. The mean total disclosure score for MND companies was 41.4 out of the maximum possible score of 76 representing 54.47% of the disclosure checklist. This indicated that companies may view some disclosures as more important than others. These results confirm prior findings made by Conway et al. (2015), Merkl-Davies and Brennan (2007) and the original study by Leung et al. (2015) that management of companies are selective in their choice of disclosure.

Table 1 Descriptive statistics of the variables
Full size table

4.2 Hp1 and Hp3

4.2.1 Hp1

Regression model 1 allowed the researcher to conclude on Hp1 and Hp3. The regression analysis presented in Table 2 revealed that there is no association between a company’s current performance and its MND score. In terms of both variables used as a measure for current performance, namely, ROA and TOBINQ, the results are not statistically significant (at both the 0.05 and 0.1 level). There is no evidence, in terms of Hp1, to support the statement that companies with poor current performance are more likely to engage in the concealment of voluntary narrative disclosures in IR. This finding differed from the original study carried out by Leung et al. (2015), which found that there is an inverse association between a company’s performance and MND score. The findings, however, are in line with prior research on concealment strategies carried out by Courtis (1995), Smith and Taffler (1992), Clatworthy and Jones (2001) and Rutherford (2003) which found no such association exists.

Table 2 Logistic Regression results
Full size table

4.2.2 Hp3

The DISTRESS variable in Table 2 which was used as a measure of financial distress is statistically significant at the 0.05 level. There indicated that there is an association between a company’s financial distress levels and its MND score. However, the results for financial distress showed a significantly negative coefficient (−0.664) which indicated that companies with a higher insolvency risk are inclined to disclose more information in their IR. This supports the statement by Miller (2002) that whilst the financial distress of a company is associated with MND in IR, companies choose to disclose more information using narratives when facing increasing levels of financial distress. This finding however differs from that of Leung et al. (2015), which found a positive correlation between MND score and the level of financial distress.

The results on the control variables in Table 2 further illustrate that the SIZE variable is statistically significant at the 0.05 level and has a positive coefficient (1.179). Whilst Leung et al. (2015) found an inverse relationship between size and MND score, in a South African context the results indicate that smaller companies disclose more information. A possible reason for this could be the need for smaller companies to obtain financing in the future in order to grow the company (Miller, 2002).

The results of the liquidity control variable (LIQUID) in Table 2 were significant at the 0.1 level and had a significant positive coefficient (3.387). This illustrated that, despite a company’s liquidity increasing and the company requiring less external funding, companies still choose to disclose more information by means of voluntary narratives and not decrease or conceal information in IR. This finding differed from that of Leung et al. (2015) which found that there is an inverse relationship between liquidity and a company’s MND score. This may be due to voluntary disclosures having a beneficial effect on both liquidity and company value and that these effects are plausibly causal, providing a justification to companies for voluntarily disclosing more information than is mandated (Balakrishnan et al., 2014).

The results also show that all other control variables such as CEO ownership and other corporate governance variables are not associated with minimal disclosure behaviour as these variables are not statistically significant at the 0.05 and 0.1 levels as was found in the original study by Leung et al. (2015).

Overall, the results in respect of financial distress support Hp3, that there is an association between a company’s MND score and level of financial distress however, this is an inverse association, resulting in more information disclosed as financial distress increases. These findings are contrary to that of Leung et al. (2015) which found a positive association between a company’s MND score and level of financial distress.

4.2.3 Hp2

Regression model 2 was used to investigate Hp2. The results from regression model 2 are reported in Table 3 and reflects the analysis of whether MND companies experience performance deterioration or improvement in the subsequent year. All the change variables were calculated as the differences between the 2018 and 2019 financial years. In the original paper by Leung et al. (2015), companies which reported deteriorating performance in the following year signalled that current poor performance was long-lasting.

Table 3 Regression model 2 results for changes in the future performance of MND companies
Full size table

Contrary to the findings of the study performed by Leung et al. (2015), the results in Table 3 reveal that there is no statistical significance for the MND company variable at the 0.05 level (or 0.1 level). This illustrated that there is no evidence to conclude that MND in IR is correlated in any way with the future performance of an MND company (whether future performance improves or deteriorates) within the context of this research.

The results also reveal that the ΔDISTRESS variable is statistically significant at the 99% confidence level and has a positive coefficient. This, expectedly, results in the deterioration of an MND company’s future performance as the company experiences more financial distress (measured using the Altman’s-Z score as expressed within the Methodology section). All other change variables within the results are shown not to be statistically significant and do not affect the company’s future performance (ΔPERF).

Overall, in respect of Hp2, the results of regression model 2 do not support the statement that there is an association between a company’s MND score and future performance. This finding differed from that of Leung et al., 2015, which found that where negative performance was consistent and continued in a future period, there was an association between a company’s MND score and future performance.

5 Conclusion

The purpose of this paper was to determine if the top 100 JSE-listed companies utilised impression management strategies, specifically MND in their IR and its related impact on their performance. This was explored through the use of disclosure guidelines against which the narrative disclosures made in IR by the sampled companies were analysed. The research further focused on the types of information management a company may choose to conceal and explored possible reasons for the non-disclosure, particularly as annual reporting frameworks call for greater transparency in reports.

The results from the collected data revealed that from the sample of 79 companies selected, 49% of these companies were classified as MND companies based on their disclosure score obtained from the disclosure guidelines. The paper also revealed that there is no association between a company’s current performance and its MND score. However, the results revealed that there is an inverse association between a company’s financial distress levels and the MND score of the company, as companies facing financial distress disclosed more information. No evidence was obtained to support that MND in IR relates to a company’s future performance.

The results obtained differed from the findings of Leung et al. (2015) which found that there is an association between a company’s MND score compared to current performance and future performance. Both papers found an association between financial distress levels and the MND score of a company however, Leung et al. (2015) found that companies facing financial distress would tend to disclose less information if there was no need for external funding whilst this study found the opposite to be true. Both papers concluded that the management of companies utilises impression management as a strategy by selectively choosing the type of information disclosed within IR. A limitation of this paper is that it consists of only a sample of JSE-listed companies and is not representative of all JSE-listed companies. Furthermore, this paper only covers the IR for the 2018 financial year and the focus was placed primarily on the obfuscation of information through MND as an impression management strategy. Future areas of research can include expanding on investigating impression management strategies in narrative disclosures within other countries and regions to expand on the body of knowledge. Additionally, within a South African context future research can focus on specifically exploring the reasons, such as incentives and disincentives, linked to a company being classified as an MND company.