Does Alignment with the IIRF Influence Capital Markets? Evidence from South Africa and the UK

Abstract

This study examines whether integrated reports that are more closely aligned with the International Integrated Reporting Framework (IIRF) are differently associated with firm value compared to those less aligned. Using panel estimated generalised least squares and other robust estimations, the analysis covers the Top 100 firms listed on South Africa’s Johannesburg Stock Exchange and the United Kingdom’s London Stock Exchange from 2011 to 2018. South Africa presents a mandatory integrated reporting (IR) setting, while the UK adopts a voluntary approach, offering a natural comparative context. An IR quality index was constructed to measure the degree of alignment with the IIRF, and market value of equity and Tobin’s Q are used as proxies for firm value. The results show no evidence of capital market differentiation in South Africa between more and less IIRF-aligned reports. In contrast, UK capital markets may differentiate, with less-aligned reports showing a significant negative association with firm value. These findings suggest that low-quality integrated reports may undermine firm value in voluntary IR settings.

1. Introduction

For much of the twentieth century, financial accounting alone was considered sufficient for stakeholder decision-making, as economic decisions focused mainly on firm performance (Delegkos et al., 2022). However, in recent years, there has been a wider call from regulators and professional bodies for firms to consolidate their financial and non-financial reports to enhance their investors’ decision-making (De Villiers & Dimes, 2023; Nwachukwu, 2022). This increased call to provide information beyond financial information, such as environmental, social, and governance (ESG) information, led to the emergence of a framework to regulate this new trend in reporting, namely the International Integrated Reporting Framework (IIRF) (Izzo et al., 2025).

Although the International Integrated Reporting Council (IIRC), which is the body promoting integrated reporting (IR), was formed in 2010 (IIRC, 2011), the true concept of IR predates this (Bek-Gaik & Surowiec, 2022). As early as the 2000s, some companies had already started issuing corporate reports that combined financial and non-financial information into a single report (Eccles et al., 2019). The occurrence of major global events, such as financial crises, corporate collapses, and environmental concerns like climate change, further exacerbated the need for IR (De Villiers et al., 2017). Such events led to calls for reporting that include both financial and non-financial aspects affecting the firm as a whole—namely, IR (Shanmugam et al., 2025).

As with any form of reporting, a new approach such as IR requires a clear guideline or framework that outlines the process to be followed in preparing these reports. As such, the IIRC released its first framework in 2013, which served as a guide for IR (IIRC, 2013). The purpose of the IIRF (2013) is to improve the quality of information available to investors to enable them to make effective business decisions; enable concise reporting by consolidating various reports into a single report; enforce integrated thinking, which enables value creation; and increase accountability over the management of an organisation’s capitals. As the interest in IR grew, IR also started attracting attention from academics and practitioners worldwide (Gaia et al., 2025). This led various scholars to start examining the role of IR in creating, maintaining, or eroding value for the firm’s stakeholders over time (Izzo et al., 2025). This is because the IIRF (2021) states that the primary purpose of an integrated report is to “explain to providers of financial capital how an organization creates, preserves or erodes value over time”. Prior studies have argued that this translates to the quality of IR, rather than the quantity of information disclosed in an integrated report (Nada & Győri, 2023). Contrary to its intended purpose, IR has been described as “one of the most disruptive innovations in corporate reporting” (Gibassier et al., 2018). In line with the IR adoption and implementation challenges reported in the literature, many organisations have struggled to adapt to the IIRF’s demands, particularly the shift towards integrated thinking and a new approach to reporting (Flower, 2015). As Strong (2015) pointed out at the inception of IR, this type of reporting has followed a controversial path and is likely to remain a complex and widely debated issue for the foreseeable future.

Despite a reported global increase in the number of firms producing integrated reports (Zúñiga et al., 2025), the academic literature on IR remains notably fragmented and underdeveloped (Fayad et al., 2024). While IR was introduced with the promise of enhancing decision-making—particularly for financial capital providers—its actual utility remains contested and empirically inconsistent (Kamotho et al., 2022). Several studies cast doubt on whether IR fulfils the objectives outlined by the IIRC, particularly its intended usefulness to capital providers (Dumay et al., 2017; Rensburg & Botha, 2014). Although Steyn (2014) argued early on for IR’s decision-usefulness, this position lacks robust, consistent empirical backing across various contexts. There is still limited understanding of how IR quality influences firm valuation across different regulatory settings (Permatasari & Tjahjadi, 2024). Research focusing specifically on IR quality—measured by alignment with the IIRF—is scarce (Zhou et al., 2017; Marrone, 2020). At the same time, evidence casts doubts about IR’s relevance in practice: Slack and Tsalavoutas (2018) found UK equity participants often prefer traditional reports over integrated ones, while McNally et al. (2017) showed many South African reports are only superficially integrated, limiting their usefulness. This lack of clarity restricts a full appreciation of IR’s value and leaves investors, regulators, and firms without clear guidance on the benefits of improving IR quality. Additionally, the issue of integrated report alignment to the IIRF is complex because Pistoni et al. (2018) also report that certain firms declare the use of IIRF in preparing their integrated reports, but a deeper analysis reveals that it is a mere declaration rather than alignment to the IIRF.

Against this backdrop, this study makes three contributions. First, it is one of the few studies that looks at reporting quality, measured by alignment with the IIRF, rather than just adoption or disclosure volume. Second, the study offers a distinctive cross-country perspective on how institutional contexts influence the capital market relevance of IR quality by contrasting South Africa, an obligatory IR setup, with the UK, a voluntary setting. Third, by connecting IIRF alignment to theories of legitimacy, information asymmetry, and signalling, we enhance research on IR. This places IR quality within well-established disclosure theories and emphasises its theoretical significance for comprehending how capital markets value corporate information.

The results show that capital markets in South Africa do not differentiate between integrated reports with high or low alignment to the IIRF. In contrast, UK capital markets do make this distinction, with less aligned reports negatively and significantly associated with firm value. Since this study compares two countries with different IR regimes, the findings hold relevance for regulators, especially where IR is voluntary. It is also possible that alignment variation is greater among South African firms, where IR is mandatory, than among UK firms, where adoption is voluntary. As for firms, the evidence suggests that in voluntary settings such as the UK, higher IR alignment can enhance firm value by signalling quality, whereas in mandatory settings such as South Africa, the market does not appear to reward higher IR quality. This guides firms to balance compliance with efforts to improve reporting. Overall, this study contributes to the literature by demonstrating that, depending on the regulatory context, IR quality—as measured by IIRF alignment—has varying capital market implications. By doing this, the study contributes to the current discussions on sustainability reporting, expands on disclosure theory, and offers advice to businesses, investors, and regulators.

The rest of the study proceeds as follows. Section 2 provides background information on IR, reviews literature on IR’s usefulness and empirical studies linking integrated reports’ alignment with the IIRF to firm value, and also discusses the theoretical framework and hypotheses. Section 3 outlines the research methodology, including the sample, IR alignment index, and regression analyses. Section 4 presents the results, followed by the discussion in Section 5. Section 6 concludes the study.

2. Literature Review

2.1. Integrated Reporting

The IIRC defines an integrated report as a “concise communication about how an organisation’s strategy, governance, performance and prospects lead to the creation of value over the short, medium and long term” (IIRC, 2013). At the core of IR is integrated thinking, which facilitates the connectivity of information to enhance management reporting and decision-making (IIRC, 2013). Therefore, IR goes beyond just consolidating financial and non-financial information, it provides comprehensive integrated disclosures that enable value creation (Pistoni et al., 2018). This is where the concept of value creation stems from (Zúñiga et al., 2020). Regarding the purpose of IR, the IIRC (2013) states that it is intended “to explain to the providers of financial capital how an organisation will create value in the future, and it contains financial and other information.” These providers of financial capital are identified in the literature as shareholders and investors (Wahl et al., 2020; Tlili et al., 2019; Barth et al., 2017). The IIRC (2013) also outlines that firms create value for these capital providers through the use of various forms of capital: financial, manufactured, intellectual, human, social and relationship, and natural capitals.

In terms of structure, the IIRF (2013) outlines seven guiding principles that preparers of integrated reports must adhere to: strategic focus and future orientation, connectivity of information, stakeholder relationships, materiality, conciseness, reliability and completeness, and consistency and comparability. Additionally, the IIRF (2013) outlines eight content elements to be included in an integrated report: organisational overview and external environment, governance, business model, risks and opportunities, strategy and resource allocation, performance, outlook, and basis of presentation. The Framework emphasises that the information presented should support connectivity among these elements. In 2021, the IIRC issued a revised framework which is applicable to all reporting periods starting on or after 1 January 2022 (IIRC, 2021). In this regard, it should be noted that this study used the earlier version of the IIRF, which was published in 2013, throughout the study, as the IIRF (2013) applies to the study’s sample firm-years (2011 to 2018).

The adoption of IR varies across different countries due to regulatory differences (Shanmugam et al., 2025). In this regard, South African firms have long been recognised as pioneers in corporate reporting and governance reforms (Erin & Adegboye, 2022). It is thus not surprising that South African firms were amongst the first to adopt IR, as early as from 2011 through King III’s “apply or explain” principle (IoDSA, 2013). It is also for this reason that most studies on IR tend to focus on South African data (for example, Lee & Yeo, 2016; Zhou et al., 2017; Barth et al., 2017; Zúñiga et al., 2020).

2.2. Usefulness of IR

The usefulness and relevance of information in IR is often evaluated through the lens of the Conceptual Framework, which defines relevance in terms of its capacity to influence user decisions (IASB, 2018). However, the methodological diversity in studies—ranging from case studies and surveys to interviews—has led to mixed findings. For instance, Slack and Tsalavoutas (2018) reported that IR had limited decision-usefulness for capital market participants in the UK, many of whom viewed it as redundant in relation to traditional annual reports. This scepticism is compounded by evidence that many South African integrated reports lack meaningful integration of financial and non-financial information, reducing their practical value (McNally et al., 2017).

Critiques of IR’s conceptual and practical shortcomings are not new. Flower (2020) attributes the failure of IR to regulatory capture, where the accounting profession has marginalised sustainability considerations. Similarly, Bridges et al. (2022) underscore that sustainability was noticeably downplayed in the IIRF (2013), raising concerns about the framework’s ability to reflect non-financial performance adequately. Ribeiro et al. (2024) reinforce this critique, showing that Brazilian investors regard IR as largely redundant, offering little value beyond conventional disclosures. Hossain et al. (2023) argue further that IR is often used instrumentally by firms to legitimise reporting practices rather than to substantively improve transparency or utility.

A persistent gap between preparers and users’ perceptions also undermines IR’s effectiveness. Naynar et al. (2018) identified a misalignment in understanding IR’s purpose between report preparers and stakeholders, echoing earlier findings from the IIRC’s (2017) consultation process. The widespread confusion over whether IR should serve all stakeholders or focus primarily on financial capital providers—despite the IIRF’s explicit stance—raises critical concerns about the coherence of the reporting process. This ambiguity further calls into question whether IR can be expected to meet user needs effectively. Supporting this concern, Abhayawansa et al. (2019) found that many analysts and investors were unaware of IR or the IIRF, even in firms that already issued integrated reports, indicating a fundamental disconnect between the intended users and the reporting framework. This raises the question: If the preparers of IR do not understand who the main audience is, how can it be useful to those users?

Additionally, disclosure quality continues to be a central concern. Nguyen et al. (2022), examining Vietnamese firms, found that only 43% of required IR information was disclosed, pointing to substantial gaps in compliance with IIRF guidelines. Although Nada and Győri (2023) report improvements in IR quality over time within the European Union, these findings are based on an index approach which has not yielded consistent results across various contexts. Moreover, Asadi et al. (2024) found that only in mandatory IR settings—particularly when combining IIRF and SASB frameworks—was there a positive association with firm value. This contrasts with voluntary environments, where no such effect was observed. Finally, Rossignoli et al. (2022) challenge the assertion that IR improves market efficiency, finding no significant improvement in analyst forecast accuracy in weaker regulatory settings.

In sum, while IR continues to attract scholarly attention, the evidence base supporting its effectiveness, particularly in fulfilling its primary objective of informing capital providers, remains inconclusive and, at times, contradictory. These limitations underscore the need for more rigorous, context-sensitive, and user-focused empirical work to determine whether IR can genuinely deliver on its promises.

2.3. Integrated Reporting Alignment with IIRF and Firm Value

The IIRC’s failure to establish a rigorous, enforceable framework for assessing IR quality has resulted in significant inconsistencies across disclosures. This lack of standardisation has not only undermined comparability but also cast doubt on IR’s purported ability to enhance value creation (Nada & Győri, 2023). The work of Furtuna and Uykulu (2025), examining Turkish firms, exemplifies this issue. Their findings—that only a minority of firms meaningfully align their reports with the IIRF content elements—underscore a gap between the framework’s aspirations and its practical adoption.

While proponents claim that IR can reduce information asymmetry and improve firm performance indicators such as cost of capital and liquidity (Zúñiga et al., 2025), these assertions often rest on selective and context-dependent evidence. For instance, early studies like Lee and Yeo (2016) and Barth et al. (2017) present a positive link between IR quality and firm performance, but these are situated within South Africa—a unique context where IR has been mandatory since 2011 under King III (IoDSA, 2013). The argument that mandatory compliance leads to higher-quality reports (Muttakin et al., 2020) is plausible, yet it also exposes a structural flaw: IR’s effectiveness appears contingent on compulsion, not voluntary uptake. This fragility becomes more evident when examining contradictory findings within the same jurisdiction. Boujelben et al. (2024), using South African data, found no significant overall link between IR quality and capital markets—although firms with the highest-quality reports did show some association. However, this nuance fails to rescue the broader claim that IR universally enhances market performance. The implication here is clear: only a select subset of firms—perhaps those already predisposed to transparency—derive benefit from IR, calling into question its broader utility.

Outside South Africa, the empirical record is similarly fragmented. Studies focusing on Europe (e.g., Hurghiş et al., 2025) and Asia (e.g., Sun et al., 2022; Nakajima & Inaba, 2022) suggest marginal benefits such as reduced forecast error or improved firm value—but again, only under specific conditions, such as robust disclosure of the IIRF’s six capitals (financial, manufactured, human, social, intellectual, and natural capital). These findings risk overstating IR’s utility, especially given their reliance on self-selected samples of voluntary adopters. Even the claimed reduction in financing costs in Turkey (Pirgaip & Rizvić, 2023) lacks broader validation. By contrast, a growing body of research undermines the case for IR in voluntary contexts. Cooray et al. (2020) report no significant effect on firm value in Sri Lanka, while Wahl et al. (2020) even document a negative relationship. Leukhardt et al. (2022) further reinforce the notion that IR quality has little bearing on firm value absent regulatory pressure. Hsiao et al. (2022) extend this critique, showing no discernible impact of IIRF adoption on market perceptions, again in voluntary contexts. These studies collectively suggest that IR’s ability to influence capital markets and firm value is highly context-specific and often overstated. Without mandatory enforcement or incentives, IR may simply function as symbolic compliance rather than a meaningful strategic tool.

2.4. Theory and Hypothesis Development

Several theories have been adopted in the literature to examine the relationship between IR alignment to the IIRF and firm value. Agency theory suggests that this relationship is influenced by the dynamics which exist between the company and its shareholders, which if not in alignment, results in information asymmetry (De Villiers & Maroun, 2018). IR or any form of corporate reporting has the ability to reduce this information asymmetry by providing sufficient and relevant disclosures, which ultimately increase transparency and potentially increase firm value (Hamdouni, 2025). Therefore, if firms provide integrated reports, which are prepared as per the IIRF, using the guiding principles and containing all the content elements as guided by the IIRF, then IR is likely to reduce the agency costs between the firm and the providers of financial capital. This notion is supported by Zhou et al. (2017), who found that integrated reports of JSE-listed firms that aligned with the IIRF tended to reduce analyst forecast error. Furthermore, Zhou et al. (2017) reported that increased alignment with the IIRF was associated with reduced cost of capital. Additionally, firms with integrated reports aligned to the IIRF have been shown to reduce information asymmetry (Martinez, 2016). However, since the IIRF is principle-based, the level of alignment can vary across firms, which limits comparability, and does not always guarantee value relevance (Zúñiga et al., 2020).

There is also evidence that high quality integrated reports send a positive signal to users of integrated reports, which ultimately improves firm reputation (Amirrudin et al., 2021). This stems from the signalling theory, which has been extensively used in the literature to describe behaviour when two different parties do not have access to the same information (Connelly et al., 2011). In simple terms, one party (the signaller) communicates information about a firm, and the other party (the receiver who has limited or no information) interprets what that information signals about the quality of the firm (M. A. Omran & El-Galfy, 2014). This signal is either information, which was not known previously, or incremental information (Yasar et al., 2020). In capital markets research, signalling theory has been used to explain investors’ hesitancy in investing in a firm because of the information asymmetry that exists between managers of firms and investors (Maama & Marimuthu, 2021). The argument is that disclosures signal the superior performance of the firm, which in turn increases firm value (Hsiao & Kelly, 2018). This is because corporate disclosures are regarded as signalling mechanisms employed by firms to showcase their superior performance (Shehata, 2014). It can thus be argued that IR aligned to the IIRF is a signalling mechanism for such firms.

Considering that the IIRF is principle-based, the current study argues that firms issuing integrated reports may not align their reports with the IIRF as suggested by the IIRC, or there may be a variation in the level of alignment. Because of this variation, firms are free to decide what they disclose, so firms may use their discretion in terms of what to disclose in their integrated reports, and this variation between integrated reports may be considerable (Bakker et al., 2020). Furthermore, the signalling theory suggests that firm disclosures serve as signals of firm quality, helping to reduce the risk of adverse selection (Oktorina et al., 2022). Based on this, integrated reports that are more aligned with the IIRF are expected to benefit the reporting firm by lowering information asymmetry. This is because Permatasari and Tjahjadi (2024) argue that IR can also be used strategically to influence competitor perceptions. As such, integrated reports with higher alignment to the IIRF are expected to show a positive association with firm value, while those with lower alignment may show no association or even a negative one. Based on the theoretical framework and prior literature, the hypotheses are stated as follows:

H₁: 

In South Africa, integrated reports that are more aligned with the IIRF are evaluated differently by capital markets compared to those that are less aligned.

H₂: 

In the United Kingdom, integrated reports that are more aligned with the IIRF are evaluated differently by capital markets compared to those that are less aligned.

3. Research Methods

3.1. Research Philosophy and Method

This study is embedded in a positivist research paradigm, which assumes a singular, objective reality that can be measured independently of the researcher. The aim was to identify and test associations between variables using empirical and statistical methods. The positivist paradigm was deemed appropriate because the research followed a structured process: identifying theory, developing hypotheses, measuring variables, performing statistical analysis, and interpreting results in relation to the theoretical framework. Accordingly, a quantitative research method was adopted.

3.2. Sample and Data Sources

The study covered two samples for the period 2011 to 2018. The period 2011 to 2018 was selected for this study as it commenced during the foundational years of IR, both in South Africa and the United Kingdom. South African firms were mandated to prepare IR in 2011, through King III, on an apply or explain basis (IoDSA, 2013), and the UK also started seeing a rise in a number of firms voluntary adopting IR (Lopes & Coelho, 2018). The data period ends in 2018, eight years after the IIRC promulgated IR, hence providing a sufficient period to conduct empirical analysis using the original IIRF (2013). The study applied the IIRF (2013) since this was the framework applicable during the sample period. Following other similar studies in the literature, data from 2019 onwards was excluded to avoid the volatile impact of the COVID-19 pandemic (Senani et al., 2024).

The South African sample included the Top 100 JSE-listed firms by market capitalisation. Financial data were obtained from IRESS, while integrated reports were manually collected from company websites. Disclosure was guided by IIRF (2013). The sampled firms belonged to the Basic Materials, Financials, Consumer Goods, and Consumer Services sectors. After excluding firms with missing financial data, shorter listing periods, or unavailable integrated reports, the final South African sample included 521 firm-year observations. Likewise, the UK sample also consisted of the Top 100 LSE-listed firms. Integrated reports were sourced from company websites, and financial data from Refinitiv. Firms were retained only if data were complete and publicly available for all years. After exclusions, the final UK sample consisted of 525 firm-year observations. Similarly to other cross-country IR studies in the literature, financial data were converted to ZAR using year-end exchange rates (Wu & Zhou, 2022). Leading sectors included Financials, Consumer Services, Industrials, and Consumer Goods. The final pooled sample included 1046 firm-year observations.

3.3. Measurement of Integrated Reporting Alignment (IRSCORE)

This study adopted a validated IR scoring index from Khatlisi and Mashamba (2025). The IR score (IRSCORE) was based on the IIRF (2013). The IRSCORE incorporated both the guiding principles and content elements of the IIRF (2013). Similarly to other studies in the literature such as Manes-Rossi et al. (2021) and Kılıç and Kuzey (2018), the study adopted a non-weighted IRSCORE, which assigned a score of 1 if the firm’s integrated report contained the disclosure element as described by IIRF or 0 otherwise. The IR scoring index is as shown in Appendix A. The IRSCORE included 35 disclosure items—20 covering the IIRF’s seven guiding principles and 15 addressing the eight content elements. These items were drawn from prior empirical studies and structured around specific questions formulated from the framework (Soriya & Rastogi, 2023). All information which was analysed was attributed the same weighting in order to avoid subjectivity that can occur when weights are unequal (Manes-Rossi et al., 2021). Therefore, the final IRSCORE was expressed as a composite index, where each item carried the same weight. Therefore, the IRSCORE was computed as follows:

IRSCORE = (Guiding Principles × 0.50) + (Content Elements × 0.5)/100

The validity and reliability tests of the IRSCORE were carefully considered. Firstly, the IRSCORE was developed from literature. We carefully considered similar IR studies in the literature which utilise disclosure checklists, including the Ernst and Young IR Awards (Ernst and Young, 2019; Ahmed Haji & Anifowose, 2017). We also based our index on the IIRF (2013) to ensure that it assessed all elements as intended by the IIRC (Radwan & Wang, 2024). Secondly, a pilot test using a sample of 20 integrated reports was performed, and any discrepancies between the two coders carefully considered and resolved (Manes-Rossi et al., 2021). Thirdly, to ensure the reliability of the index, inter-item correlations were calculated to ensure that the items on the scale assessed the same content (Pistoni et al., 2018). The values of inter-item correlations for guiding principles and content elements were within the recommended range of 0.15–0.50 (Clark & Watson, 2016).

3.4. Empirical Specification, Variables and Proxies

• Model I—Impact of IR alignment on market value of equity

This study adopted the levels approach to measure firm value, consistent with its objective of examining the impact of IR alignment on firm value. The Ohlson (1995) model is a widely utilised modelling approach in such studies where market value of equity, (MVE) is expressed as a function of book value of equity, earnings, and non-accounting information, (Permatasari & Narsa, 2022). All accounting values were scaled by the number of shares to address scaling issues (Radwan & Wang, 2024). The key independent variable was IRSCORE, capturing the extent to which integrated reports aligned with the IIRF. Similarly to prior studies in the literature utilising the Ohlson model (Permatasari & Narsa, 2022), the regression equation is formulated as follows:

MVE_it = β₀ + β₁BVE_it + β₂EARN_it + β₃HighIRSCORE_it + β₄LNASSETS_it + β₅ROE_it + β₆LEV_it + β₇INDSENS_it + β₈LOSS_it + YR_it + IND_it + Ɛ_it

(1)

MVE_it = β₀ + β₁BVE_it + β₂EARN_it + β₃LowIRSCORE_it + β₄LNASSETS_it + β₅ROE_it + β₆LEV_it + β₇INDSENS_it + β₈LOSS_it + YR_it + IND_it + Ɛ_it

(2)

• Model II—Impact of IR alignment on Tobin’s Q

The study also utilised the Tobin’s Q model to examine the impact of IR alignment on firm value, similar to studies in the literature (Senani et al., 2024). Tobin’s Q is determined by the sum of market value plus the liabilities of the firm divided by total assets (Nirino et al., 2022). Similarly to prior studies in the literature utilising the Tobin’s Q model (Lee & Yeo, 2016), the regression equation is expressed as follows:

TOBINQ_it = β₀ + β₁HighIRSCORE_it + β₂LEV_it + β₃ROA_it + β₄CAPEXR_it + β₅LNASSETS_it + β₆SALESG_it + β₇DIV_it + β₈INDSENS_it + β₉LOSS_it + YR_it + IND_it + Ɛ_it

(3)

TOBINQ_it = β₀ + β₁LowIRSCORE_it + β₂LEV_it + β₃ROA_it + β₄CAPEXR_it + β₅LNASSETS_it + β₆SALESG_it + β₇DIV_it + β₈INDSENS_it + β₉LOSS_it + YR_it + IND_it + Ɛ_it

(4)

The variables included in both models are described in

Table 1. Definition of Variables.

Variable	Proxy/Description	Type	Variable
Firm Value	Tobin’s Q calculated as the total assets minus book value of equity plus market value of equity, all divided by total assets	Dependent	Tobin’s Q
Firm Value	Market Value of Equity at reporting date scaled by the number of shares at the end of the year	Dependent	MVE
High Integrated reporting scoring index	IRSCOREs equal to or above the median	Independent	HighIRSCORE
Low Integrated reporting scoring index	IRSCOREs below the median	Independent	LowIRSCORE
Book value of equity	Book value of equity at reporting date scaled by the number of shares at the end of the year	Independent	BVE
Earnings	Net profit for the year scaled by the number of shares at the end of the year	Independent	EARN
Size of the firm	Logarithm of Total Assets	Control	LNASSETS
Profitability	Return on assets calculated as net profit for the year divided by total book value of assets	Control	ROA
Profitability	Return on equity calculated as net profit for the year divided by book value of equity	Control	ROE
Leverage	Debt Ratio	Control	LEV
Loss	Binary: 1 = loss-making firm; 0 otherwise	Control	LOSS
Industry Sensitivity	Binary: 1 = sensitive industries, 0 otherwise	Control	INDSENS
Sales Growth	Total sales of the current year minus total sales of the previous year, divided by sales of the previous year	Control	SALESG
Capital expenditure ratio	Total capital expenditure divided by total book value of assets	Control	CAPEXR
Dividend	Binary variable equal to 1, if firm declared or paid a dividend during the year; 0 otherwise.	Control	DIV

below.

Both models included the following control variables, based on prior literature:

• Firm profitability, measured by return on equity (ROE), and return on assets (ROA) was included in the study in each model. More profitable firms are typically better positioned to invest in high-quality reporting and have stronger incentives to do so. In addition, profitability enhances investor appeal (Sobhan & Mia, 2024; Salvi et al., 2022). Based on this, a positive relationship is expected between profitability and firm value.
• In contrast, firms that report losses may lack both the financial capacity and motivation to produce high-quality integrated reports. These firms may also have limited incentives to signal value to the market (Muttakin et al., 2020). Loss (LOSS) is defined as a binary variable equal to one if a firm reported a loss in a given year. A negative association is expected between reporting a loss and firm value.
• Leverage (LEV), captured by the debt ratio, was also considered. Highly leveraged firms are more closely monitored by creditors and are often expected to disclose more to reduce agency conflicts. This increased demand for transparency creates an incentive for enhanced disclosure (Bansal, 2025). As a result, leverage is expected to have a positive association with firm value.
• Industry sensitivity (INDSENS) is measured using a binary variable equal to one for firms in environmentally sensitive industries such as mining, oil, or tobacco. These firms typically face greater reputational risk and external scrutiny, which drives them to engage in more voluntary disclosure (Appiagyei & Donkor, 2024). Therefore, industry sensitivity is expected to be positively associated with firm value, reflecting the role of disclosure in building legitimacy and managing stakeholder expectations.
• Sales Growth (SALESG) captured by sales growth ratio was also considered as stronger growth in sales reflects improved firm financial performance. This can send a favourable message to capital markets, potentially impacting the firm’s overall value (Salvi et al., 2022).
• Capital expenditure (CAPEXR) was also considered. Firms with greater opportunities for capital growth may provide more information through integrated reports to attract investors thereby lowering agency costs (Obeng et al., 2021). Therefore, a positive association is expected between capital expenditure and firm value.
• Dividends (DIV) was also considered. A dividend declaration or payment is an indication of a firm’s growth, which ultimately reduces agency costs (Asadi et al., 2024). Therefore a positive association is expected between a dividend declaration or payment expenditure and firm value.

This study followed a structured three-step procedure to evaluate whether the capital markets differentiate between firms with higher and lower levels of IR alignment.

Step 1: For each country sample (South Africa and the UK), firms were split into two subgroups based on their IRSCORE.

• The HighIRSCORE group included firms with IRSCOREs equal to or above the median.
• The LowIRSCORE group included firms with scores below the median.
  This classification follows prior literature, where a higher IRSCORE reflects stronger alignment with the IIRF and better disclosure quality (Oktorina et al., 2022; Zhou et al., 2017).

Step 2: Separate regression analyses were conducted (as discussed above) for each subgroup to test the relationship between IRSCORE and firm value, using both Market Value of Equity and Tobin’s Q as outcome variables.

Step 3: A Z-statistic test (Paternoster et al., 1998) was used to compare the regression coefficients from the HighIRSCORE and LowIRSCORE groups. This test assessed whether the difference in the strength of association was statistically significant between the two subgroups.

3.5. Estimation and Diagnostic Tests

Consistent with the economic approach of prior studies employing similar regression analyses (Tobin’s Q and Ohlson models), the study employed the panel estimated generalised least squares (EGLS) estimator for the analysis, with robust standard errors applied to control for heteroscedasticity (Mervelskemper & Streit, 2017). Year and industry fixed effects were also incorporated to control for time-specific and sector-specific influences (Landau et al., 2020). To test for serial correlations between errors in the regression models, the Durbin-Watson test was calculated for all regression models in this study in line with prior studies (Seltman, 2012). Prior studies suggest that Durbin-Watson values between 1.5 and 2.5 indicate the absence of serial correlation in the models (Buallay et al., 2020; Marcia et al., 2015). However, Field (2013) asserts that, as a conservative rule of thumb, only Durbin-Watson statistic values below 1 and greater than 3 are a serious cause for concern—based on Field (2013), values within the range of 1 to 3 were thus regarded as acceptable in this study.

Furthermore, the period seemingly unrelated regression (SUR) estimates, which corrected for heteroskedasticity and general correlation of observations within a cross-section were applied. We also applied White’s (1980) diagonal standard errors and covariance, which is a robust standard error estimation method (Reitmaier et al., 2024). The purpose of this exercise was to check that the significance values had not been influenced by heteroskedasticity (Mervelskemper & Streit, 2017). Following prior studies in the literature such as Cortesi and Vena (2019), this exercise was performed for all regression equations performed in the study.

4. Results

4.1. Descriptive Statistics

Descriptive statistics for the two countries’ samples are shown in

Table 2. Descriptive statistics: South African sample (Model I and Model II).

	N	Mean	Median	Std. Deviation	Minimum	Maximum
MVE	521	107,355	35,612	263,995	6924	2,466,698
BVE	521	54,845	16,371	140,328	−1020	1,118,475
EARN	521	7822	2131	34,954	−90,433	638,942
SALESG	521	0.148	0.092	0.793	−1.677	17.319
ROE	521	0.195	0.152	0.494	−1.911	9.523
LEV	521	2.978	1.123	5.527	−44.355	36.731
ROA	521	0.073	0.060	0.101	−0.332	1.125
CAPEXR	521	0.071	−0.049	3.060	−2.380	69.656
TOBIN’SQ	521	2.100	1.399	2.411	−0.225	40.151
LNASSETS	521	7.693	7.600	0.677	5.880	9.550
LOSS	521	0.083	0.000	0.275	0.000	1.000
INDSENS	521	0.253	0.000	0.435	0.000	1.000
DIV	521	0.987	1.000	0.115	0.000	1.000
HighIRSCORE	521	0.503	1.000	0.500	0.000	1.000
LowIRSCORE	521	0.497	0.000	0.500	0.000	1.000
IRSCORE	521	0.803	0.830	0.126	0.359	0.982

Note: The descriptive statistics of MVE, BVE and EARN are denoted in ZAR million. All variables are as defined in Table 1.

(South Africa) and

Table 3. Descriptive statistics: United Kingdom’s sample (Model I and Model II).

	N	Mean	Median	Std. Deviation	Minimum	Maximum
MVE	525	229,976	118,483	267,037	25,985	1,673,516
BVE	525	155,974	53,226	343,411	−19,248	2,909,137
EARN	525	17,544	7213	51,018	−131,666	828,767
SALESG	525	0.084	0.038	0.522	−0.901	10.758
ROE	525	0.223	0.168	0.553	−7.706	3.626
LEV	525	5.148	1.131	19.927	−95.756	226.919
ROA	525	0.073	0.060	0.082	−0.232	0.814
CAPEXR	525	−0.109	−0.043	0.697	−11.484	0.590
TOBINQ	525	1.831	1.454	1.432	−2.687	13.934
LNASSETS	525	8.316	8.160	0.710	6.634	10.408
LOSS	525	0.05	0.00	0.221	0	1
INDSENS	525	0.19	0.00	0.392	0	1
DIV	525	0.99	1.00	0.115	0	1
HighIRSCORE	525	0.50	1.00	0.500	0	1
LowIRSCORE	525	0.50	0.00	0.500	0	1
IRSCORE	525	0.802	0.805	0.085	0.021	0.961

Note: The descriptive statistics of MVE, BVE and EARN are denoted in ZAR million. All variables are as defined in Table 1.

(United Kingdom). According to the findings, UK businesses are typically far bigger than South African businesses. For instance, South African companies’ mean market value of equity (MVE) was R107,355 million, while UK companies’ MVE was R229,976 million. The size disparity is also reflected in the medians (R35,612 million vs. R118,483 million). Similar disparities can be seen in profits (EARN) and book value of equity (BVE), with UK companies reporting significantly higher averages. In contrast, South Africa and the UK had nearly similar average IR scores (IRSCOREs), with South Africa scoring 0.803 and the UK scoring 0.802. But compared to the UK (0.021), the South African minimum score (0.359) was significantly greater. This difference reflects South Africa’s mandatory IR regime during the study period, which set a higher disclosure baseline. Meanwhile, outliers were identified in MVE, BVE, and EARN. To mitigate their impact, these variables were scaled by the number of shares, and winsorised at the 5% and 95% levels. This adjustment aligns with prior disclosure studies (Tlili et al., 2019; M. S. Omran et al., 2021).

4.2. Multicollinearity Test

Pearson correlation coefficients for the South African models are shown in

Table 4. Correlation coefficients: South African sample (Model I and Model II).

(a): Correlation Coefficients: South African Sample (Model I)

MVE

BVE

EARN

IRSCORE

HighIRSCORE

LowIRSCORE

ROE

LEV

LNASSETS

INDSENS

LOSS

MVE

1

0.686 **

0.570 **

−0.104

−0.037

0.037

0.121 **

−0.009

0.478 **

0.074

−0.064

BVE

0.686 **

1

0.402 **

−0.156 **

−0.109 *

0.109 *

−0.041

−0.021

0.569 **

0.156 **

0.031

EARN

0.570 **

0.402 **

1

−0.027

0.000

0.000

0.356 **

0.016

0.315 **

0.117 **

−0.149 **

IRSCORE

−0.104

−0.156 **

−0.027

1

0.756 **

−0.756 **

0.065

0.033

0.028

0.050

−0.025

HighIRSCORE

−0.037

−0.109 *

0.000

0.756 **

1

−1.000 **

0.064

0.055

0.079

0.041

−0.023

LowIRSCORE

0.037

0.109 *

0.000

−0.756 **

−1.000 **

1

−0.064

−0.055

−0.079

−0.041

0.023

ROE

0.121 **

−0.041

0.356 **

0.065

0.064

−0.064

1

−0.240 **

−0.039

−0.067

−0.097 *

LEV

−0.009

−0.021

0.016

0.033

0.055

−0.055

−0.240 **

1

0.379 **

−0.0169 **

−0.083

LNASSETS

0.478 **

0.569 **

0.315 **

0.028

0.079

−0.079

−0.039

0.379 **

1

0.022

0.037

INDSENS

0.074

0.156 **

0.117 **

0.050

0.041

−0.041

−0.067

−0.169 **

0.022

1

0.210 **

LOSS

−0.064

0.031

−0.149 **

−0.025

−0.023

0.023

−0.097 *

−0.083

0.037

0.210 **

1

(b) Correlation Coefficients: South African Sample (Model II)

TOBIN’SQ

IRSCORE

High IRSCORE

Low IRSCORE

LEV

ROA

CAPEXR

LNASSETS

SALESG

DIV

INDSENS

LOSS

TOBIN’SQ

1

0.100 *

0.066

−0.066

−0.133 **

0.265 **

−0.049

−0.393 **

−0.004

0.061

−0.098 *

−0.052

IRSCORE

0.100 *

1

0.756 **

−0.756 **

0.033

−0.034

−0.041

0.028

−0.039

0.175 **

0.050

−0.025

LowIRSCORE

−0.066

−0.756 **

−1.000 **

1

−0.055

0.038

0.045

−0.079

0.055

−0.017

−0.041

0.023

LEV

−0.133 **

0.033

0.055

−0.055

1

−0.165 **

−0.012

0.379 **

−0.006

0.044

−0.169 **

−0.083

ROA

0.265 **

−0.034

−0.038

0.038

−0.165 **

1

−0.001

−0.269 **

0.021

0.075

−0.083

−0.441 **

CAPEXR

−0.049

−0.041

−0.045

0.045

−0.012

−0.001

1

−0.020

−0.006

0.004

−0.027

−0.012

LNASSETS

−0.393 **

0.028

0.079

−0.079

0.379 **

−0.269 **

−0.020

1

−0.070

−0.050

0.022

0.037

SALESG

−0.004

−0.039

−0.055

0.055

−0.006

0.021

−0.006

−0.070

1

0.040

−0.044

−0.049

DIVIDEND

0.061

0.175 **

0.017

−0.017

0.044

0.075

0.004

−0.050

0.040

1

−0.047

−0.147 **

INDUSTRY

−0.098 *

0.050

0.041

−0.041

−0.169 **

−0.083

−0.027

0.022

−0.044

−0.047

1

0.210 **

LOSS

−0.052

−0.025

−0.023

0.023

−0.083

−0.441 **

−0.012

0.037

−0.049

−0.147 **

0.210 **

1

All variables are as defined in Table 1. * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).

a,b, whereas results for the UK models are shown in

Table 5. Correlation Analysis: United Kingdom sample (Model I and Model II).

(a): Correlation Analysis: United Kingdom Sample (Model I)

MVE

BVE

EARN

IRSCORE

HighIRSCORE

LowIRSCORE

ROE

LEV

LNASSETS

INDSENS

LOSS

MVE

1

0.673 **

0.472 **

−0.048

0.024

−0.024

−0.019

0.036

0.621 **

0.061

0.048

BVE

0.673 **

1

0.480 **

−0.139 **

−0.107 *

0.107 *

−0.091 *

0.016

0.606 **

0.181 **

0.134 **

EARN

0.472 **

0.480 **

1

−0.120 **

−0.085

0.085

0.094 *

−0.022

0.271 **

0.126 **

−0.117 **

IRSCORE

−0.048

−0.139 **

−0.120 **

1

0.705 **

−0.705 **

−0.104 *

−0.077

0.023

0.022

0.031

HighIRSCORE

0.024

−0.107 *

−0.085

0.705 **

1

−1.000 **

−0.058

−0.130 **

0.013

0.059

−0.011

LowIRSCORE

−0.024

0.107 *

0.085

−0.705 **

−1.000 **

1

0.058

0.130 **

−0.013

−0.059

0.011

ROE

−0.019

−0.091 *

0.094 *

−0.104 *

−0.058

0.058

1

0.168 **

−0.160 **

−0.040

−0.124 **

LEV

0.036

0.016

−0.022

−0.077

−0.130 **

0.130 **

0.168 **

1

0.397 **

−0.088 *

0.017

LNASSETS

0.621 **

0.606 **

0.271 **

0.023

0.013

−0.013

−0.160 **

0.397 **

1

0.043

0.189 **

INDSENS

0.061

0.181 **

0.126 **

0.022

0.059

−0.059

−0.040

−0.088 *

0.043

1

0.020

LOSS

0.048

0.134 **

−0.117 **

0.031

−0.011

0.011

−0.124 **

0.017

0.189 **

0.020

1

(b): Correlation Analysis: United Kingdom Sample (Model II)

TOBINQ

IRSCORE

HighIRSCORE

LowIRSCORE

LEV

ROA

CAPEXR

LNASSETS

SALESG

DIV

INDSENS

LOSS

TOBINQ

1

−0.069

−0.044

0.044

−0.092 *

0.476 **

−0.533 **

−0.492 **

0.001

0.059

−0.138 **

−0.115 **

IRSCORE

−0.069

1

0.705 **

−0.705 **

−0.077

−0.093 *

0.042

0.023

0.007

−0.028

0.022

0.031

HighIRSCORE

−0.044

0.705 **

1

−1.000 **

−0.130 **

−0.057

0.068

0.013

0.027

0.018

0.059

−0.011

LowIRSCORE

0.044

−0.705 **

−1.000 **

1

0.130 **

0.057

−0.068

−0.013

−0.027

−0.018

−0.059

0.011

LEV

−0.092 *

−0.077

−0.130 **

0.130 **

1

−0.149 **

0.029

0.397 **

0.008

−0.006

−0.088 *

0.017

ROA

0.476 **

−0.093 *

−0.057

0.057

−0.149 **

1

−0.223 **

−0.463 **

0.027

0.017

−0.032

−0.256 **

CAPEXR

−0.533 **

0.042

0.068

−0.068

0.029

−0.223 **

1

0.186 **

0.011

−0.014

0.027

0.019

LNASSETS

−0.492 **

0.023

0.013

−0.013

0.397 **

−0.463 **

0.186 **

1

−0.079

−0.027

0.043

0.189 **

SALESG

0.001

0.007

0.027

−0.027

0.008

0.027

0.011

−0.079

1

0.010

−0.008

−0.002

DIV

0.059

−0.028

0.018

−0.018

−0.006

0.017

−0.014

−0.027

0.010

1

0.014

−0.048

INDSENS

−0.138 **

0.022

0.059

−0.059

−0.088 *

−0.032

0.027

0.043

−0.008

0.014

1

0.020

LOSS

−0.115 **

0.031

−0.011

0.011

0.017

−0.256 **

0.019

0.189 **

−0.002

−0.048

0.020

1

All variables are as defined in Table 1. * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).

a,b. No correlation in either sample was higher than 0.90, indicating that multicollinearity was not an issue (Asuero et al., 2006). Strong positive correlations between MVE, BVE, and EARN were found, as anticipated, and these correlations were in line with the valuation model proposed by Ohlson (1995). Although its HighIRSCORE and LowIRSCORE splits were treated as distinct subgroup studies, IRSCORE showed a moderate correlation with them. Other correlations, like the one between profitability and leverage, fell into the low to moderate range. Additionally, variance inflation factors (VIFs) were computed to verify that multicollinearity was not present. The credibility of the regression results was further supported by the fact that all untabulated VIFs were below 10 (Thompson et al., 2017; M. S. Omran et al., 2021).

4.3. Regression Results

4.3.1. Regression Results—South Africa

• Model I: Market value of equity as a proxy for firm value

Regression results for South African companies with market value of equity (MVE) as the dependent variable are shown in

Table 6. Panel EGLS regression results: market value of equity and the level of integrated reporting in line with the IIRF: HighIRSCORE (South Africa).

Equation (1): MVEit = β0 + β1BVEit + β2EARNit + β3HighIRSCOREit + β4LNASSETSit + β5ROEit + β6LEVit + β7INDSENSit + β8LOSSit + YRit + INDit + Ɛit
Variable		Unstandardised Coefficient	Robust Standard Error	T-Statistic	p-Value
BVE		0.2500	0.1707	1.4647	0.1443
EARN		7.0762 ***	0.9950	7.1116	0.0000
HighIRSCORE		−60.9524	68.1741	−0.8940	0.3722
LNASSETS		22.7218 *	11.9550	1.9006	0.0585
ROE		41.9395	33.5406	1.2504	0.2123
LEV		−3.7050 **	1.6754	−2.2113	0.0279
INDSENS		−3.1098	11.7042	−0.2657	0.7907
LOSS		35.6196 ***	10.4708	3.4017	0.0008
N	262
Durbin-Watson statistic	1.3910
R-squared	0.6558
Adjusted R-squared	0.6348
F-statistic	31.2565 ***
Prob (F-statistic)	0.0000

***, ** and * denote significance at the 1%, 5% and 10% levels, respectively. Refer to Table 1: Variable definitions.

and

Table 7. Panel EGLS regression results: market value of equity and level of integrated reporting in line with the IIRF: LowIRSCORE (South Africa).

Equation (3): MVEit = β0 + β1BVEit + β2EARNit + β3LowIRSCOREit + β4LNASSETSit + β5ROEit + β6LEVit + β7INDSENSit + β8LOSSit + YRit + INDit + Ɛit
Variable		Unstandardised Coefficient	Robust Standard Error	T-Statistic	p-Value
BVE		0.4352 ***	0.1661	2.6199	0.0093
EARN		3.6876 ***	1.1606	3.1772	0.0017
LowIRSCORE		−15.0179	37.2228	−0.4034	0.6870
LNASSET		24.1801 **	12.0582	2.0052	0.0460
ROE		125.5725 **	50.7749	2.4731	0.0141
LEV		−0.9834	2.3155	−0.4247	0.6714
INDSENS		2.1848	23.6005	0.0925	0.9263
LOSS		32.3186 *	19.0788	1.6939	0.0916
N	259
Durbin-Watson statistic	1.1888
R-squared	0.5755
Adjusted R-squared	0.5493
F-statistic	21.9651 ***
Prob (F-statistic)	0.0000

***, ** and * denote significance at the 1%, 5% and 10% levels, respectively. Refer to Table 1: Variable definitions.

. The coefficient on HighIRSCORE for companies with high IRSCOREs (Table 6) was negative and not statistically significant (−60.95; p = 0.37). This implies that business value was not substantially impacted by greater alignment with the IIRF. On the other hand, leverage (LEV) and loss reporting (LOSS) had a substantial negative impact, whilst earnings (EARN) and company size (LNASSETS) had considerable positive benefits. These findings suggest that South African investors penalised companies with large debt or losses and placed a higher value on profitability and scale than on disclosure alignment. However, the outcomes for companies with low IRSCOREs (Table 7) were essentially the same. Additionally, the LowIRSCORE coefficient was not significant (−15.02; p = 0.69). However, along with earnings and business size, book value of equity (BVE) grew considerable and positive. Although it was only marginally significant, loss reporting remained favourable. There was no discernible difference between the high and low IRSCORE groups, according to the Paternoster Z-test (Z = 0.59). Therefore, when MVE was employed as the firm value metric, Hypothesis 1 was disproved. When combined, these results show that differences in disclosure quality had no discernible impact on market valuation in South Africa’s required IR environment. Rather, investors placed more weight on conventional financial metrics including business size, book equity, and earnings.

To assess whether integrated reports with high levels of alignment to the IIRF are evaluated differently by capital markets compared to those with low alignment, the study applied the Z-statistic test by Paternoster et al. (1998). The test yielded a Z-statistic of 0.5913, indicating no statistically significant difference in how capital markets evaluate high versus low IIRF-aligned reports. This result aligns with the findings in Table 6 and Table 7. Consequently, H₁ was rejected when firm value was measured using the market value of equity.

• Model II: Tobin’s Q as a proxy for firm value

The regression results for Equations (3) and (4) when Tobin’s Q was used as a proxy for firm value are presented in

Table 8. Panel EGLS regression results: Tobin’s Q and the level of integrated reporting in line with the IIRF: HighIRSCORE (South Africa).

Equation (3): TOBINQit = β0 + β1HighIRSCOREeit + β2LEVit + β3ROAit + β4CAPEXRit + β5LNASSETSit + β6SALESGit + β7DIVit + β8INDSENSit + β9LOSSit + YRit + INDit + Ɛit
Variable		Unstandardised Coefficient	Robust Standard Error	T-Statistic	p-Value
HighIRSCORE		1.3013	0.9463	1.3750	0.1704
LEV		0.0930 ***	0.0213	4.3597	0.0000
ROA		12.7197 ***	1.2078	10.5305	0.0000
CAPEXR		−0.6027	0.8070	−0.7469	0.4558
LNASSETS		−0.5553 ***	0.1742	−3.1869	0.0016
SALESG		−0.3026	0.2657	−1.1390	0.2558
DIV		0.4829 *	0.2780	1.7371	0.0836
INDSENS		−0.0069	0.1884	−0.0367	0.9707
LOSS		1.0919 ***	0.1938	5.6325	0.0000
N	262
Durbin-Watson statistic	1.2830
R-squared	0.7439
Adjusted R-squared	0.7272
F-statistic	44.4849 ***
Prob (F-statistic)	0.0000

***, and * denote significance at the 1% and 10% levels, respectively. Refer to Table 1: Variable definitions.

and

Table 9. Panel EGLS regression results: Tobin’s Q and level of integrated reporting in line with the IIRF: LowIRSCORE (South Africa).

Equation (4): TOBINQit = β0 + β1LowIRSCOREit + β2LEVit + β3ROAit + β4CAPEXRit + β5LNASSETSit + β6SALESGit + β7DIVit + β8INDSENSit + β9LOSSit + YRit + INDit + Ɛit
Variable		Unstandardised Coefficient	Robust Standard Error	T-Statistic	p-Value
LowIRSCORE		0.2106	0.5504	−0.0447	0.7198
LEV		0.0596 ***	0.0191	3.1182	0.0020
ROA		7.4504 ***	1.1132	6.6927	0.0000
CAPEXR		0.6325	1.0583	0.5976	0.5507
LNASSETS		−0.5728 ***	0.0992	−5.7741	0.0000
SALESG		−0.1270	0.3371	−0.3767	0.7067
DIV		0.6008	0.4620	1.3003	0.1947
INDSENS		−0.2123	0.2828	−0.7509	0.4534
LOSS		0.6228 ***	0.2171	2.8680	0.0045
N	259
Durbin-Watson statistic	1.2030
R-squared	0.5870
Adjusted R-squared	0.5597
F-statistic	21.5008 ***
Prob (F-statistic)	0.0000

*** denote significance at the 1% level. Refer to Table 1: Variable definitions.

. Table 8 presents the results for the regression estimated for the sample with integrated reports with IRSCORE values above the median value. The coefficient for HighIRSCORE, the variable of interest, was positive but not significant (coefficient = 1.3013; p-value = 0.1704). Although the coefficient for HighIRSCORE, the variable of interest, is positive, the p-value of 0.1704 shows that it is not statistically significant. This means the study finds weak evidence that high levels of IR are positively associated with firm value when measured using Tobin’s Q. In other words, the results do not provide strong support that capital markets reward firms for higher alignment with the IIRF in the South African context. Unlike the earlier MVE model findings, firm leverage shows a significant positive association with firm value, while firm size shows a significant negative association when measured by Tobin’s Q. Firm profitability, measured by ROA, shows a significant positive relationship with Tobin’s Q (coefficient = 12.72; p-value = 0.00), while dividends have a marginal positive effect (coefficient = 0.48; p-value = 0.08). SALESG, CAPEXR, and INDSENS have negative but insignificant coefficients, indicating an unclear impact on firm value.

Likewise, Table 9 presents the results for the regression estimated for the sample with integrated reports with IRSCORE values below the median value. The coefficient for LowIRSCORE, is positive and insignificant (coefficient = 0.2106; p-value = 0.7198). These findings indicate that low levels of IIRF have no significant effect on firm value measured by the Tobin’s Q, all else equal. To further test whether integrated reports with high levels of alignment to the IIRF are evaluated differently by capital markets compared to those with low alignment, the Z-statistic test by Paternoster et al. (1998) was applied. The test produced a Z-statistic of 0.9795, confirming no significant difference in capital market evaluation between the two groups. This result supports the findings presented in Table 8 and Table 9. As a result, H₁ was rejected when firm value was measured using Tobin’s Q. At the same time, most control variable results remain consistent. The only exception is CAPEXR, which is now positive but still insignificant, reinforcing the limited role of capital expenditure in explaining firm value measured by Tobin’s Q in the South African sample.

When combined, the positive but statistically insignificant coefficients from the high and low IRSCORE samples showed conflicting results about whether capital markets reward or penalise businesses according to their degree of IR. The Z-statistic test result (Z = 0.9795), which indicates no discernible difference in the way capital market players assess companies with high and low alignment to the IIRF, supports this conclusion. These negligible results are more than just statistical noise, though. They show that alignment differences are minimal in South Africa, where IR adoption is required and average IRSCORE values are already high. Capital markets may view all JSE-listed reports as adequately credible due to the little heterogeneity, which reduces the marginal signalling effect of higher alignment. This suggests that in compulsory regimes, the market does not strongly differentiate between high and low levels of IR quality.

4.3.2. Regression Results—United Kingdom

• Model I: Market value of equity as a proxy for firm value

Results for the UK sample with MVE as the dependent variable are shown in

Table 10. Panel EGLS regression results: market value of equity and the level of integrated reporting in line with the IIRF: HighIRSCORE (United Kingdom).

Equation (1): MVEit = β0 + β1BVEit + β2EARNit + β3HighIRSCOREit + β4LNASSETSit + β5ROEit + β6LEVit + β7INDSENSit + β8LOSSit + YRit + INDit + Ɛit
Variable		Unstandardised Coefficient	Robust Standard Error	T-Statistic	p-Value
BVE		1.2070 ***	0.1714	7.0406	0.0000
EARN		2.9472 ***	0.6312	4.6691	0.0000
HighIRSCORE		19.3290	138.8797	0.1391	0.8894
LEV		4.6521 **	2.0421	2.2781	0.0236
ROE		12.4245	31.6936	0.3920	0.6954
INDSENS		92.8958 ***	29.6124	3.1370	0.0019
LOSS		−14.1225	11.5643	−1.2212	0.2232
LNASSETS		−55.7545 ***	15.1299	−3.6850	0.0003
N	265
Durbin-Watson statistic	1.2085
R-squared	0.4859
Adjusted R-squared	0.4528
F-statistic	14.6535 ***
Prob (F-statistic)	0.0000

*** and ** denote significance at the 1% and 10% levels, respectively. Refer to Table 1: Variable definitions.

and

Table 11. Panel EGLS regression results: market value of equity and the level of integrated reporting in line with the IIRF: LowIRSCORE (United Kingdom).

Equation (2): MVEit = β0 + β1BVEit + β2EARNit + β3LowIRSCOREit + β4LNASSETSit + β5ROEit + β6LEVit +β7INDSENSit + β8LOSSit + YRit + INDit + Ɛit
Variable		Unstandardised Coefficient	Robust Standard Error	T-Statistic	p-Value
BVE		1.5848 ***	0.143041	11.07995	0.0000
EARN		2.5330 ***	0.742028	3.413663	0.0008
LowIRSCORE		−79.3413	52.92886	−1.499018	0.1352
LNASSETS		−77.2075 ***	16.85019	−4.581996	0.0000
LEV		11.8488 ***	2.796181	4.237505	0.0000
ROE		258.5635 ***	52.02163	4.970307	0.0000
INDSENS		108.9763 ***	41.76518	2.609262	0.0096
LOSS		−2.0724	29.10986	−0.071194	0.9433
N	260
Durbin-Watson statistic	1.3395
R-squared	0.5524
Adjusted R-squared	0.5230
F-statistic	18.7496 ***
Prob (F-statistic)	0.0000

*** denote significance at the 1%level, respectively. Refer to Table 1: Variable definitions.

. The HighIRSCORE coefficient for high-IRSCORE enterprises (Table 10) was positive but not statistically significant (19.33; p = 0.89). This suggests that greater business value was not correlated with greater alignment with the IIRF. Firm size (LNASSETS) was negative, but industry sensitivity (INDSENS), book value of equity (BVE), earnings (EARN), and leverage (LEV) were all significant and positive. On the other hand, the coefficient on LowIRSCORE was negative but not statistically significant for low-IRSCORE enterprises (Table 11) (−79.34; p = 0.14). The following were all favourable and significant in this case: book value of equity (BVE), earnings (EARN), profitability (ROE), leverage (LEV), and industry sensitivity (INDSENS). The Z-test (Z = 1.51) confirmed no statistically significant difference between high and low IRSCORE groups. Therefore, Hypothesis 2 was rejected when MVE was the proxy for firm value.

• Model II: Tobin’s Q as a proxy for firm value

Table 12. Panel EGLS regression results: Tobin’s Q and the level of integrated reporting in line with the IIRF: HighIRSCORE (United Kingdom).

Equation (3): TOBINQit = β0 + β1HighIRScoreit + β2LEVit + β3ROAit + β4CAPEXRit + β5LNASSETSit + β6SALESGit + β7DIVit + β8INDSENSit+ β9LOSSit + YRit + INDit + Ɛit
Variable		Unstandardised Coefficient	Robust Standard Error	T-Statistic	p-Value
HighIRSCORE		0.1919	0.7518	0.2553	0.7987
LEV		0.0309 ***	0.0086	3.5624	0.0004
ROA		4.6942 ***	0.6685	7.0215	0.0000
CAPEXR		−2.7507 ***	0.5910	−4.6537	0.0000
LNASSETS		−0.4490 ***	0.0837	−5.3632	0.0000
SALESG		0.0312	0.0248	1.2550	0.2107
DIV		0.4149 **	0.1668	2.4867	0.0136
INDSENS		0.1709 *	0.0959	1.7809	0.0762
LOSS		0.0012	0.0804	0.0156	0.9875
N	265
Durbin-Watson statistic	1.4359
R-squared	0.5737
Adjusted R-squared	0.5425
F-statistic	18.3972
Prob (F-statistic)	0.0000

***, ** and * denote significance at the 1%, 5% and 10% levels, respectively. Refer to Table 1: Variable definitions.

and

Table 13. Panel EGLS regression results: Tobin’s Q and the level of integrated reporting in line with the IIRF: LowIRSCORE (United Kingdom).

Equation (4): TOBINQit = β0 + β1LowIRScoreit + β2LEVit + β3ROAit + β4CAPEXRit + β5LNASSETSit + β6SALESGit + β7DIVit + β8INDSENSit + β9LOSSit + YRit + INDit + Ɛit
Variable		Unstandardised Coefficient	Robust Standard Error	T-Statistic	p-Value
LowIRSCORE		−0.6175 *	0.3471	−1.7789	0.0765
LEV		0.0543 ***	0.0089	6.0773	0.0000
ROA		4.3574 ***	0.9062	4.8082	0.0000
CAPEXR		−2.8897 ***	0.6712	−4.3047	0.0000
LNASSETS		−0.6173 ***	0.0842	−7.3272	0.0000
SALESG		−0.2069 **	0.0966	−2.1413	0.0332
DIV		0.1284	0.1338	0.9593	0.3383
INDSENS		0.3534 **	0.1429	2.4728	0.0141
LOSS		0.0413	0.1000	0.4137	0.6794
N	260
Durbin-Watson statistic	1.5430
R-squared	0.5926
Adjusted R-squared	0.5622
F-statistic	19.4801 ***
Prob (F-statistic)	0.0000

***, ** and * denote significance at the 1%, 5% and 10% levels, respectively. Refer to Table 1: Variable definitions.

report results for the UK sample using Tobin’s Q as the dependent variable. For high-IRSCORE firms (Table 12), the coefficient on HighIRSCORE was positive but insignificant (0.19; p = 0.80). Key controls such as profitability (ROA), leverage (LEV), dividends (DIV), and industry sensitivity (INDSENS) were significantly positive, while capital expenditure (CAPEXR) and firm size (LNASSETS) were significantly negative. For low-IRSCORE firms (Table 13), the coefficient on LowIRSCORE was negative and weakly significant (−0.62; p = 0.08). This indicates that poor alignment with the IIRF was associated with lower firm value. Controls behaved largely as expected: profitability (ROA), leverage (LEV), and industry sensitivity (INDSENS) were significantly positive, while capital expenditure (CAPEXR) and firm size (LNASSETS) were significantly negative. Interestingly, sales growth (SALESG) was negatively associated with firm value, suggesting investor scepticism about the sustainability of rapid growth. The Z-test (Z = −2.47) confirmed a significant difference in valuation between high- and low-IRSCORE groups when Tobin’s Q was used. Therefore, Hypothesis 2 was supported under this specification.

In summary, unlike South Africa, UK results show that poor IR alignment is penalised by the market under Tobin’s Q. This suggests that in voluntary settings, investors use IR quality as a screening mechanism—rewarding credible reports and punishing poor ones.

4.4. Synthesis Across Countries

The findings from South Africa support the idea that required IR establishes a high standard of disclosure quality since they reveal no discernible market difference between companies with high and low alignment. The UK data, on the other hand, show that poor IR is penalised, particularly when Tobin’s Q is used, underscoring the significance of alignment in voluntary contexts.

Table 14. Summary of hypothesis tests by country and valuation proxy.

Country	Proxy	High IRSCORE	Low IRSCORE	Z-Test Result	Hypothesis Outcome
South Africa	MVE	Insignificant	Insignificant	Not significant	H1 rejected
South Africa	Tobin’s Q	Insignificant	Insignificant	Not significant	H1 rejected
UK	MVE	Insignificant	Insignificant	Not significant	H2 rejected
UK	Tobin’s Q	Insignificant	Negative and significant	Significant difference	H2 supported

presents a more comprehensive summary of the findings for both nations and value proxies.

One important result from this synthesis is that financial markets are not always affected by IR alignment. In accordance with signalling and legitimacy theories, poor alignment is penalised in voluntary regimes like the UK, whereas alignment variation has less impact in mandatory regimes like South Africa.

5. Discussion

Hypotheses 1 and 2 predicted that integrated reports with a high level of IR in line with the IIRF are evaluated differently by capital markets compared to integrated reports with a low level of IR in line with the IIRF in South Africa and the United Kingdom.

5.1. South Africa

The regression analyses did not support Hypothesis 1. High-quality integrated reports (HighIRSCORE) and low-quality integrated reports (LowIRSCORE) were not positively associated with firm value, whether proxied by market value of equity or Tobin’s Q. The Z-statistic test confirmed that differences in coefficients between HighIRSCORE and LowIRSCORE firms were not statistically significant. This pattern indicates a boundary effect, which makes it significant analytically. The baseline assumption of conformity in South Africa’s mandated IR framework seems to be so ingrained that further alignment does not give investors new information. This result can be explained by signalling theory, which states that the marginal signalling value of improved alignment decreases as all firms comply. Comparably, the information asymmetry hypothesis suggests that minor increases in disclosure lose market significance after a regime-level action lowers uncertainty generally. The findings differ from Zhou et al. (2017), who reported reduced analyst forecast error and cost of capital with greater IR alignment. The divergence can be attributed to different outcomes (forecast errors vs. firm value) and contextual market differences. The South African capital market may already embed assumptions of high IR quality, particularly among the JSE Top 100. Eccles et al. (2019) and Marrone and Oliva (2020) suggest that South African reports consistently demonstrate high quality, which means the observed variance in alignment may be too narrow to influence valuation.

Another possible explanation for the absence of valuation disparities across South African enterprises is the so-called “competitive flattening” effect. Most businesses, notably those in the JSE Top 100, are expected to adhere to a high standard of compliance in a mandated IR regime. Consequently, there is less opportunity for significant distinction based on IR quality. Capital markets find it challenging to differentiate between companies based only on IR alignment because of the homogeneity in compliance levels, reducing integrated reports’ perceived signalling value. Additionally, this suggests that the integrated report in a required regime may eventually serve more as a compliance artefact than a strategic communication tool.

These results extend the disclosure literature by clarifying a boundary condition: although IR quality has been hypothesised to improve legitimacy and lessen information asymmetry, our data indicate that the anticipated benefits might not be realised in required situations. This enhances our contribution by proving that the regulatory framework affects the market relevance of IR quality, a point that has frequently been missed in earlier research.

5.2. United Kingdom

The regression results partially supported Hypothesis 2. When firm value was proxied by Tobin’s Q, low-quality IR (LowIRSCORE) was negatively associated with firm value, while high-quality IR was not significantly positive. The Z-statistic test confirmed a significant difference between the two groups. This indicates an asymmetrical association: markets may penalise poor alignment more heavily than they reward strong alignment. The result is consistent with signalling theory—high-quality IR sends a positive signal that enhances credibility, while low-quality IR sends a negative signal that erodes investor trust. Legitimacy theory also supports this interpretation: poor alignment may undermine legitimacy in the eyes of stakeholders, prompting market penalties. The results are support those of Moloi and Iredele (2020) and Lee and Yeo (2016), who documented valuation gains for high-quality IR, while Wahl et al. (2020) argue that low-cost, superficial IR can damage firm reputation and valuation. The UK evidence demonstrates that in voluntary settings, investors use IR alignment as a screening mechanism, punishing firms that fall short of credible disclosure expectations. This evidence supports the idea that voluntary regimes foster a market-driven environment where the calibre of disclosure becomes a differentiator is supported by this study. Capital markets penalise low-quality integrated reports in these situations because they perceive a lack of alignment with the IIRF as a symptom of inadequate governance, transparency, or prospects for the future.

5.3. Cross-Country Synthesis

A boundary condition for disclosure theory is shown by the cross-country comparison. Baseline compliance in mandatory regimes reduces the incremental value of alignment by flattening differentiation. Meanwhile, investors in voluntary regimes reward reliable signals and penalise unreliable ones, underscoring the importance of IR quality in terms of legitimacy and information. This theoretical extension demonstrates that IR’s relevance to the capital market is conditional rather than universal. By demonstrating how institutional context affects the applicability of signalling and legitimacy viewpoints, the study deepens theory. The findings show that generalisations about the market relevance of IR are incorrect for both practice and policy.

5.4. Implications for Policymakers and Practitioners

Evidence from South Africa indicates that while requiring IR may ensure baseline compliance, it does not always result in value gains. Therefore, regulators must go beyond adoption criteria and consider supplementary measures that meaningfully differentiate across organisations, including independent assurance, standardised metrics, or monitoring procedures. In the meantime, the results from the UK indicate that poor alignment may be penalised by market discipline. Therefore, regulators in comparable situations may rely on principle-based regulation, allowing capital markets to promote high-quality disclosure through guidance and light-touch inspection.

6. Conclusions

This study set out to examine whether integrated reports with high alignment to the IIRF are valued differently by capital markets compared to those with low alignment in South Africa and the United Kingdom. Using data from the Top 100 firms on the JSE and LSE for the period 2011–2018, the findings reveal that the market relevance of IR quality is context-dependent.

Both high and low alignment are not associated with company value as determined by the market value of equity (MVE) or Tobin’s Q in South Africa, where IR adoption is required. This implies that capital markets do not distinguish across IR quality levels, most likely due to the prevalence of high-quality reporting and the negligible variations amongst enterprises. In contrast, IR quality is important in the voluntary sector in the UK. When Tobin’s Q is employed as a proxy, low alignment is punished, but strong alignment does not substantially increase firm value. This asymmetrical effect demonstrates that investors react more strongly to disclosures of low quality than to small adjustments made to reports that are already of high quality.

By elucidating a boundary condition, that the institutional context shapes the influence of IR quality on company value, these findings advance disclosure theory. Because compliance is universal under mandated regimes, the benefits of high-quality disclosure in terms of signalling and information asymmetry are diminished. However, in voluntary regimes, markets penalise companies that do not meet credible norms by using IR alignment as a screening tool. By showing that disclosure consequences vary and depend on the regulatory environment, this expands on current thinking.

The study also carries implications for policymakers, practitioners, and investors. Evidence from the UK indicates that markets already penalise poor-quality reports, which is important for regulators in voluntary systems. Although greater incentives might be required to persuade businesses to strive for high-quality disclosures rather than just avoiding penalties, principle-based regulation can be useful in this situation. Regulators should understand that market advantages are not produced by adoption alone in mandated environments like South Africa. To establish significant differentiation, it might be necessary to strengthen enforcement, provide independent assurance, and encourage creativity in reporting. The findings highlight how crucial it is for managers to see IR as more than just a compliance activity. While companies should incorporate IR into strategy communication in required regimes to maintain legitimacy and investor trust, poor alignment in optional regimes runs the danger of direct market penalties. While other indicators like ESG ratings, governance quality, and financial performance may be more informative than IR scores in mandatory contexts, the results advise investors to pay close attention to low-alignment firms in voluntary contexts as these may indicate higher risk or weaker transparency.

Lastly, we highlight limitations to this study. Based on the initial 2013 IIRF, the dataset spans 2011–2018. There would have been problems with comparability if the period had been extended further, introducing the updated 2021 framework. Further studies may examine the role of IR assurance, look at longer horizons once frameworks are harmonised, or expand the analysis to more jurisdictions to ascertain whether the observed border requirements hold across different institutional and cultural contexts. Second, although the IR alignment index has been validated against prior research, it is dependent on content analysis and may not accurately indicate the scope of integrated thinking or the dependability of disclosures. This constraint encourages further research that looks at how external assurance can increase the validity of alignment measurements or that combines qualitative and quantitative evaluations. Third, the study focuses on large, listed companies in South Africa and the UK, which are already subject to intense reporting scrutiny and may not fairly reflect the businesses of smaller or developing countries. Other researchers are thus encouraged to expand the scope to include small and mid-sized enterprises or other nations where IR adoption is less developed to ascertain whether the tendencies seen are comparable across company sizes and institutional levels.