Recognition and Disclosure of Intangibles Under International Financial Reporting Standards – Ibrahim – 2025 – Abacus

IAS 38: Accounting for Intangibles

The main standard that sets out the criteria for recognition and disclosure of intangibles (other than acquired goodwill) is IAS 38. Since its adoption in 2001, the standard has been subject to few modifications. The standard defines an intangible asset as ‘an identifiable non-monetary asset without physical substance’ (IAS 38.8), and states that an asset is identifiable when it is ‘separable’, or when it ‘arises from contractual or other legal rights’ (IAS 38.12).

The recognition criteria for intangible assets are quite stringent. Specifically, the standard states that: ‘an intangible asset shall be recognised if, and only if: (a) it is probable that the expected future economic benefits that are attributable to the asset will flow to the entity; and (b) the cost of the asset can be measured reliably’ (IAS 38.21).

Hence, it is more likely for intangible assets that are part of an acquisition to meet the ‘identifiability’ criterion as well as both requirements above, and hence be recognized. The accounting standard on accounting for business combinations also supports this. IFRS 3 states that ‘the acquirer recognises the acquired identifiable intangible assets, such as a brand name, a patent or a customer relationship, that the acquiree did not recognise as assets in its financial statements because it developed them internally and charged the related costs to expense’ (IFRS 3.13). For internally generated intangible assets, the standard specifically states that internally generated brands, mastheads, publishing titles, customer lists, and similar items should not be recognized as intangible assets. ‘Development costs’ are allowed to be recognized on the statement of financial position only if they meet further stringent requirements.

Furthermore, IAS 38 requires some disclosure in the notes to the financial statements before qualifying this by stating that: ‘an entity is encouraged, but not required, to disclose … a brief description of significant intangible assets controlled by the entity but not recognised as assets because they did not meet the recognition criteria in this Standard …’ (IAS 38.128(b), emphasis added). Hence, there is no formal requirement under the standard to disclose information about intangible items that are not recognized as assets.

Recognition of Intangible Assets: Background and Academic Evidence

Recent evidence points to a decrease in the value relevance of financial report information over time due to an increase in unrecognized intangible assets (e.g., Lev, 2018, 2019; Zambon et al., 2020). Academics and investors are of the view that a major problem in IAS 38 is that similar intangible items are accounted for differently depending on whether they are acquired or internally generated (Zambon et al., 2020; EFRAG, 2021; UKEB, 2023; Xie and Zhang, 2023), which can mislead investors and managers (Atkinson and McGaughey, 2006). Even if an item is defined as an asset under the Conceptual Framework (IFRS, 2018), it might not be recognized as such if it is internally generated. This is especially problematic in non-traditional new economy firms that invest heavily in intangibles (Xie and Zhang, 2023).

For acquisitive firms, at the time of acquisition, firms can recognize goodwill in addition to identifiable intangible assets. In fact, most firms recognize at least one intangible asset in the statement of financial position (Tsalavoutas et al., 2014). Academic evidence finds that acquired identifiable intangible assets that are recognized separately from goodwill are value relevant (Kallapur and Kwan, 2004; Sahut et al., 2011; Bauman and Shaw, 2018; King et al., 2023) and lead to future cashflow improvements (Deng and Lev, 2006).

For non-acquisitive firms, recognizing intangible assets is more problematic. IAS 38 reflects conservatism that encourages the expensing over the capitalization of internally generated intangible assets (Mazzi et al., 2019). Hence, firms that do not engage in acquisitions and which grow organically are unlikely to recognize a significant amount of intangible assets in their financial statements. Petkov (2011) argues that the principles used for identifying the existence of externally generated intangible assets resulting from a business combination can be adopted for assessing internally generated intangibles. However, reliability of measurement of these internally generated intangibles may limit their value relevance if they are recognized (Wyatt, 2008). Barth et al. (2003) indicate that basing recognition decisions on reliability is too simplistic. Xie and Zhang (2023) discuss several drawbacks from recognizing more internally generated intangibles, including their lack of reliability and difficulty in measurement. Barker et al. (2022) suggest that conditional recognition would be beneficial.

Development cost is the only internally generated intangible item that can be capitalized under IAS 38; hence, a significant amount of research focuses on research and development (R&D) costs. Recognizing these intangibles is informative (Kimbrough, 2007; Oswald et al., 2017; Mazzi et al., 2022a) and can have real effects (Dinh et al., 2019; Oswald et al., 2022). These recognized costs are also value relevant (Dedman et al., 2009; Tsoligkas and Tsalavoutas, 2011). Non-recognition leads to a decline in earnings usefulness (Lev and Zarowin, 1999; Dugar and Pozharny, 2021). However, the current recognition criteria for R&D are so stringent that they can disincentivize firms from capitalizing R&D costs (Ma and Zhang, 2023). Hence, capitalization of R&D costs in practice is not common (Mazzi et al., 2019; Campbell et al., 2023). There are managerial incentives to recognize more in some situations (Russell, 2017) possibly linked to opportunism (Mazzi et al., 2019).

A few studies have examined the value relevance of other internally generated intangibles and find that they are value relevant, for example, brands (Barth et al., 1998; Krasnikov et al., 2009), corporate reputation (Raithel and Schwaiger, 2015), and intellectual capital (Yang et al., 2015).

In the current study, we first examine the extent of recognition of intangible assets in an IFRS-compliant regime, comparing firms that engage in acquisitions with those that do not. We expect the former to recognize more intangible assets than the latter given the restrictions in IAS 38 to recognize internally generated intangibles.

Disclosure of Intangible Items: Background and Academic Evidence

Disclosure of information about intangibles in the accompanying notes to the financial statements can be for informative, legitimate reasons, or for impression management (Abeysekera, 2014; Appleton et al., 2023). These disclosures can relate to both recognized/capitalized intangible assets and unrecognized intangible items (such as R&D expenses). For the former, firms should complement their recognition of intangible assets with required disclosure under IAS 38. For the latter, IAS 38 does not require, but only encourages, entities to provide a brief description of intangibles that are not recognized as assets because they do not meet the recognition criteria (EFRAG, 2021; UKEB, 2023). Hence, whether there would be adequate disclosure about these unrecognized intangible items is an open question.

There are two competing expectations about the level of disclosure of unrecognized intangible items, which are more likely in non-acquisitive firms. On the one hand, firms may supplement non-recognition of intangibles with additional disclosure in the accompanying notes, in which case firms with higher levels of internally generated intangibles would have higher levels of disclosure. On the other hand, firms may limit their disclosure since this is not required by the standard. In addition, firms may be reluctant to provide information that is proprietary so as not to impair their future cashflows (Bhattacharya and Ritter, 1983) and information-processing costs may be high (Blankespoor et al., 2020). In this case, higher levels of internally generated intangibles would be associated with lower levels of disclosure.

Empirical evidence on this is mixed. Some studies support the first view that firms supplement non-recognition with disclosure (Schiemann et al., 2015), although this may not always be in the form of accounting disclosures (Gelb, 2002; Castilla-Polo and Ruiz-Rodriguez, 2017). Mehnaz et al. (2023) argue that greater disclosure of unrecognized intangibles facilitates a better understanding of capitalized intangible assets, providing further insight into value creation for market participants. Ciftci and Zhou (2016) find that value relevance of disclosures of patent counts and citations is greater than that of capitalized R&D expenses. Disclosure on intellectual capital is found to be both informative (Ousama et al., 2011) and value relevant (Vafaei et al., 2011). Chen et al. (2017) find that voluntary disclosure of forward-looking information on development costs is value relevant beyond earnings, book value, and capitalized R&D.

Other studies find evidence of low disclosure when firms have unrecognized intangible assets. For example, Ho et al. (2023) find that Australian firms that were required to de-recognize internally generated intangible assets after the adoption of IAS 38 did not choose to provide alternatives or substitute disclosure elsewhere in their annual reports or financial statements. Market forces alone may not be sufficient to ensure timely disclosures on unrecognized intangible items (Dedman et al., 2009). Furthermore, items that are not considered material may not be disclosed to avoid disclosure overload (Saha et al., 2019).

Many studies find that disclosure of intangible items is not adequate. For example, Gerpott et al. (2008) find that annual report intangible disclosure quality is relatively low in a sample of firms in the telecommunications industry. Zambon et al. (2023) surveyed users and preparers of financial statements about the adequacy of disclosure of intangible assets and find that both groups of stakeholders agree that not enough information is provided. Mazzi et al. (2022a) present similar findings with respect to development costs. Mazzi et al. (2022b) find that firms do not provide a high quantity of R&D-related disclosures in the (audited) financial reports section. Tsalavoutas et al. (2014) find that disclosure of some information related to internally generated intangibles is limited in a sample of non-financial companies in the EU, Australia, China, Hong Kong, New Zealand, Brazil, South Africa, and Malaysia in the financial year 2010/11.

Therefore, it is not clear if firms that engage in acquisitions would disclose more or less information related to intangible items compared to firms that develop their intangibles internally. We expect that non-acquisitive firms may supplement non-recognition with more disclosure of unrecognized intangible items; or they may disclose less intangible information since they are not required to do so.

Sample

Our sample selection process is similar to that employed in Mazzi et al.’s (2022b) study on R&D capitalization. We first used Refinitiv Eikon to obtain lists of dead and active firms in the UK during the period 2017–2022. We kept only securities that were classified as equity. We removed firm-year observations where reporting was not based on IFRS, or reporting was not in British Pounds (GBP). We also excluded observations for firms belonging to the oil and gas and mining industries as the database may classify the relevant extraction costs as development costs. We also excluded firms in the financial services industry. Next, we excluded firm-year observations with negative book values and accounting periods of more than 380—or less than 350—days (Mazzi et al., 2022b). Finally, we excluded observations before the initial public offering (IPO) and with negative values of net intangible assets. Our final sample with financial data is 3,475 firm-year observations (803 firms).

For our analysis on disclosure, we further required the firm to have publicly available annual reports on annualreports.com, a free online repository of global annual reports. This repository has the most complete listing of annual reports and simplifies the automated process of annual report collection using the Python programming language. We cross-referenced available annual reports against our final sample of 3,475 observations. We then limited our analysis to the financial statements and accompanying notes of each annual report. We did so through the identification of the page number on which the term ‘key audit matters’ appeared within the annual report, as it represents a term that commonly appears in the audit report preceding the audited financial statements.1^,2 Once the corresponding page number had been obtained, we then counted the frequency of intangible-related key words from that point in the document onwards. By following this process, we arrived at a sample of 1,460 firm-year observations (360 firms). This includes 1,223 observations that are classified as belonging to an acquisitive firm and 237 classified as non-acquisitive. Table 1 provides details of the sample selection process. The level of recognized intangible assets (net of amortization) across the full sample with available financial data (3,475 observations) and the reduced sample with matched annual reports (1,460 observations) are similar, and therefore the reduced sample is representative of the full sample.

Measures of Recognized Intangible Assets

We first examine the extent of recognition of intangible assets in our sample. We are interested in investigating trends of recognition of identifiable intangibles over time and across industries, and especially comparing firms that engage in acquisitions and those that do not. Data for the following items were collected from Refinitiv/Eikon (Datastream): recognized net intangible assets (net of amortization) (WC02649); amortization of intangible assets (WC02649); capitalized development costs (WC02504); capitalized brands (WC02507); capitalized licences (WC02510); capitalized computer software (WC18299); capitalized other intangibles (WC02513); and R&D expenses (WC01201). All variables are divided by total assets for standardization. Most firms do not report a breakdown of intangible items and therefore the number of observations for all variables other than net intangible assets are much lower than 3,475.

We compare recognition in firms that engage in acquisitions and those that do not. We would expect firms that have engaged in acquisitions to have more identifiable intangible assets recognized in the financial statements. The variable used to classify each firm according to its acquisitive nature is defined as ACQU, an indicator variable that takes the value one if the firm reports during the year a non-zero value of goodwill (on the balance sheet—WC18280), minority interest (on the balance sheet—WC03426), or an acquisition (on the cashflow statement—WC04355). If none of these conditions is met, then a value of zero is assigned.

Measures of Disclosure of Intangible Information

Table 2 presents the list of key words that are included within each proxy along with the sum of occurrences in the sample of 1,460 firm-year observations. Of the top 10 most frequently disclosed key words, six are from Category A, namely: brand(s) (N = 6,774), licence(s)/license(s) (N = 6,109), patent(s) (N = 2,250), computer software (N = 2,230), franchise(s) (N = 2,060), and trade mark(s)/trademark(s) (N = 1,859). The key words related to R&D (N = 10,312) and intellectual property (N = 1,732) are the only key words amongst the top 10 that are in Category B, while the remaining two most frequently mentioned key words are from Category C: customer relationship(s) (N = 3,083) and business model(s) (N = 2,955). There are also 990 mentions of internally generated or developed intangibles.

Multivariate Analyses

In model (3), we include ${\mathit{INT}}_{\mathit{REC}}$ to capture intangible asset intensity and RDI to control for R&D intensity. These two variables control for the recognized intangible assets and R&D expenses in the financial statements. These are included to identify the disclosure differences in acquisitive and non-acquisitive firms, after excluding amounts recognized in the financial statements.

We include several firm characteristics to control for determinants of recognition and disclosure. These factors include firm age (AGE) and size (MVE), as older and larger firms tend to recognize fewer intangibles (Russell, 2017) and prefer to expense rather than capitalize intangibles (Oswald et al., 2022). Larger firms also disclose more information on intangibles (Oliveira et al., 2006). We also include leverage (LEV) as more leveraged firms tend to report more intangible assets (Lim et al., 2020) and disclose less intangible information (Kang and Gray, 2011). We include competitiveness as firms may be less willing to disclose information on a voluntary basis when they have more proprietary information or higher proprietary costs (Verrecchia, 1983; Leuz, 2004; Nichols, 2009; Lang and Sul, 2014). Therefore, we would expect less recognition and disclosure of intangible items when proprietary costs are higher. Proxies of proprietary costs in the literature include market concentration measures such as the Herfindahl-Hirschman index (HHI) (Nichols, 2009; Chen et al., 2015) and market-to-book values (MTB) (Frankel et al., 2018).

Similarly, entry barriers (e.g., Leuz, 2004) can also affect reporting and disclosure, and this is proxied using capital intensity (PPE) and capital expenditures (CAPEX). Audit quality can also impact upon reporting and disclosure choices (Rajgopal et al., 2021). For example, audit fees (AUFEES) are higher when there is more recognition of intangible assets, implying these reports are more difficult to audit (Visvanathan, 2017; Datta et al., 2020). We also include the type of auditor (BIG4) and audit opinion (AUOPIN) as additional auditability variables, and control for profitability (EBEX). We also include a dummy for the COVID-19 years (COVID) as firms may have written off large amounts of intangibles during that period of uncertainty. The threat of litigation can affect recognition of intangibles (Lev, 2019). However, this is more relevant in the US than it is in other countries with less litigious environments (Arena and Ferris, 2018). Finally, industrial membership is important (e.g., Russell, 2017; Makrominas, 2017), with some industries such as technology and healthcare expected to have more intangible items than others (e.g., Wyatt, 2005; Xiong et al., 2022) while also being more prone to litigation (Arena and Ferris, 2018). Hence, we include dummy variables for industry membership, using SIC classifications. All variables are defined in the appendix.

Descriptive Statistics

Table 3 presents the descriptive statistics for all variables used in the study across the acquisitive (N = 1,223) and non-acquisitive (N = 237) firms. We find that, on average, recognized intangible assets (net of amortization) represent 32% of total assets, ranging between zero and 94% in acquisitive firms. This contrasts with 9% of total assets in non-acquisitive firms, ranging between zero and 83%. This difference is significant at the 1% level (p < 0.001). The median values are 30% in acquisitive firms and 1.4% in non-acquisitive firms and this difference is significant at the 1% level (p < 0.001).

Disclosure of intangible items constitutes a very small percentage in the financial statements and accompanying notes, representing less than 0.07% on average in both acquisitive and non-acquisitive firms. Disclosure of contractual and broad intangible items is higher in acquisitive firms (mean ${\mathit{INT}}_{\mathit{DiS}\_A}$ is 0.027 and 0.023 in acquisitive and non-acquisitive firms, respectively, p = 0.051; mean ${\mathit{INT}}_{\mathit{DiS}\_C}$ is 0.016 and 0.009 in acquisitive and non-acquisitive firms, respectively, p < 0.001). However, disclosure of non-contractual intangible items is lower in acquisitive firms (mean ${\mathit{INT}}_{\mathit{DiS}\_B}$ is 0.020 and 0.034 in acquisitive and non-acquisitive firms, respectively, p < 0.001). Total disclosure does not differ between both groups (mean is 0.063 and 0.066 in acquisitive and non-acquisitive firms, respectively). The maximum count of disclosure key words in the sample is 0.393 of the total word count.

There are significant differences between acquisitive and non-acquisitive firms with respect to several firm characteristics. For example, R&D intensity is higher in non-acquisitive firms, in line with these firms investing internally (mean RDI is 0.085 and 1.043 in acquisitive and non-acquisitive firms, respectively, p < 0.001). Acquisitive firms have been listed for around 25 years on average, whereas non-acquisitive firms are younger (average age 16 years). Acquisitive firms are also more valuable firms (mean MVE is 12.628 and 11.273 in acquisitive and non-acquisitive firms, respectively, p < 0.001). Acquisitive firms are also more leveraged than non-acquisitive firms (mean LEV is 0.217 and 0.156 in acquisitive and non-acquisitive firms, respectively, p < 0.001). In addition, there are significant mean and median differences in terms of auditability and the level of profit or loss. Almost 60% of the acquisitive sample firms are audited by ‘Big Four’ firms, while this figure is only 27% for non-acquisitive firms (p < 0.001). Acquisitive firms are also more profitable (mean EBEX is 0.017 and –0.157 in acquisitive and non-acquisitive firms, respectively, p < 0.001).

Table 4 presents the correlation coefficients between variables used in the regressions. The values in bold represent correlation coefficients that are significant at the 10% level or below. From the table, we see that recognized intangible assets $({\mathit{INT}}_{\mathit{REC}}$) is significantly positively related to all disclosure proxies and negatively related to R&D intensity. However, the coefficients are small, the highest being 0.355 (correlation between ${\mathit{INT}}_{\mathit{REC}}$ and ${\mathit{INT}}_{\mathit{DiS}\_C}$). Most coefficients are less than 0.5, with only three higher. These are between ${\mathit{INT}}_{\mathit{DiS}\_A}$ and ${\mathit{INT}}_{\mathit{DiS}\_\mathit{ALL}}$ (coefficient = 0.801), between ${\mathit{INT}}_{\mathit{DiS}\_B}$ and ${\mathit{INT}}_{\mathit{DiS}\_\mathit{ALL}}$ (coefficient = 0.747), and between market capitalization (MVE) and the type of auditor (BIG4) (coefficient = 0.629). Therefore, there are no issues with multicollinearity.

Trends in Recognition of Intangible Assets

Figure 1 presents the trends in capitalized net intangible assets, as a percentage of total assets, over the sample period (Panel A), across industries (Panel B), and in acquisitive firms (Panel C) and non-acquisitive firms (Panel D) in the full sample with available financial data (N = 3,475). In Panel A, we find that recognized net intangibles assets (${\mathit{INT}}_{\mathit{REC}}$) represents only around one third of total assets and has not increased over the sample period. In fact, it decreased slightly from 31.06 in 2017 to 26.75% in 2022. This decrease is not due to amortization, which remains constant at around 2% of total assets across the years. The decrease could be indicative of a shift in asset structuring or perhaps a change in how intangibles are evaluated in the wake of evolving market conditions.

In addition to net intangible assets, we present data when available on capitalized development costs (WC02504), brands (WC02507), licences (WC02510), computer software (WC18299), and other capitalized intangible assets (WC02513). A notable observation from the figure is the relative stability of computer software assets, reflecting perhaps an unwavering investment in technology despite changing economic climates. Similarly, the capitalized development costs are stable across the years, so there is no indication of an increased investment in innovation, a critical driver of long-term growth. Brands and licences maintain lower yet stable proportions (brands are 3.04% of total assets in 2022 and licences are 4.08% of total assets in 2022). Other capitalized intangibles do not demonstrate significant variation, suggesting a steady valuation of these assets. Overall, recognition over the sample period for either specific items or overall intangible assets does not appear to have increased over time. The amount of R&D expenses reported shows a slight increase in the earlier years (7.93% in 2017 up to 9% in 2019) but then decline in later years (to 6.95% in 2022).

In Panel B, we present the distribution of capitalized net intangible assets across the industries in our sample. We find that the services industry exhibits a prominent and consistent lead in recognizing net intangible assets (34.53% in 2022), potentially indicative of the sector’s reliance on intangible assets such as intellectual property and customer relationships. Manufacturing and wholesale trade industries also maintain a notable share (24% for manufacturing firms and 16% for wholesale trade firms in 2022), suggesting a significant role for intangibles such as patents and brands in these sectors. Interestingly, the agriculture and fishing sector, while smaller in comparison, shows a downward trajectory, reflecting a decreasing recognition of intangibles in an industry traditionally focused on tangible assets. Similarly, retail trade demonstrates a decline over the observed period (18.54% in 2017 to 16.29% in 2022), which could be attributed to sector-specific challenges or shifts in asset capitalization strategies. The figure shows the divergent trends in intangible asset capitalization across industries, highlighting the importance of intangibles in the modern economy and reinforcing the need for sector-specific analysis when evaluating the value and performance of firms.

In Panels C and D, we compare the level of capitalized net intangible assets and capitalized specific intangibles across acquisitive (Panel C) and non-acquisitive firms (Panel D). As expected, we find that recognized net intangible assets are more prominent in acquisitive firms. Specifically, they represent 34.6% of total assets in 2017, declining to 31.2% in 2022. Conversely, Panel B illustrates a much lower percentage of recognized net intangible assets in non-acquisitive firms (8.6% of total assets in 2022 but a higher incidence of capitalized development costs (13.9% of total assets in 2022 compared to around 6.9% in acquisitive firms)). They also spend more on research as shown by the 18.6% R&D expenses in 2022 (compared to 4.1% in acquisitive firms).

Overall, these findings align with investor concerns regarding the heterogeneity of intangible asset recognition between firms expanding through acquisitions compared to those bolstering internally generated intangibles. This variance underpins the broader discourse on the harmonization of intangible asset reporting, illuminating the necessity for refined standards that reflect the underlying economic realities of firms’ intangible asset portfolios.

Trends in Disclosure of Intangible Information

In the previous section, we find diversity in recognition of intangible assets across firms that rely on acquisitions compared to those that grow internally. It is possible though that non-acquisitive firms compensate for restrictions in recognition of intangible assets through higher levels of disclosure. Hence, in this section, we present details on the level of disclosure of intangible items in the UK over the sample period for all firms (Figure 2, Panel A), industries (Figure 2, Panel B), and acquisitive (Figure 2, Panel C) and non-acquisitive firms (Figure 2, Panel D) in the reduced sample with annual report data (N = 1,460).

In Figure 2, Panel A, we present the count of the key words in the categories of intangibles as a percentage of the total word count in the financial statements and accompanying notes for all firms in the sample. The figure shows that disclosure of intangibles in all categories has not increased over time. Specifically, disclosure of all key words represents 0.063% of the total word count in 2017 and 0.062% in 2022. A third of the disclosure in 2022 (0.026%) relates to contractual intangibles. In addition, in 2022 only 0.022% words within these sections are key words of non-contractual intangibles while 0.015% relate to broad intangibles. The overall trend indicates no increase in the disclosure of intangibles, which suggests these could be more boilerplate in nature and therefore there is no enhancement in transparency and reporting practices over the sample period. The disclosure also remains minimal, below 1% of the total word count in the relevant sections of the annual report.

In Panel B, we present the diversity of disclosure across industries. We find that the services sector exhibits the most significant level of disclosure of intangibles, suggesting a higher emphasis on intangible assets such as software, customer databases, and proprietary technologies (0.079% of total word count in 2022). The industries with the highest level of disclosure after the services industry are manufacturing (0.064% of total word count in 2022), followed by transportation (0.047% of total word count in 2022), and wholesale trade (0.04% of total word count in 2022). Interestingly, there is a decline in disclosure in the retail industry post-2018, which may reflect the sector’s reassessment of intangible asset values in the face of changing consumer behaviours and market dynamics brought on by external events such as the global pandemic. Only the construction and wholesale trade industries show a slight increase in disclosure (0.024% in 2017 to 0.029% in 2022 in the construction industry, and 0.037% in 2017 to 0.04% in 2022 in the wholesale trade industry).

In Panels C and D, we present the level disclosure across acquisitive and non-acquisitive firms, respectively. We find that acquisitive firms (Panel C) demonstrate a lower level of intangible asset disclosure than non-acquisitive firms (Panel D) for non-contractual intangibles. Specifically, 0.019% of the total word count relates to category B intangibles in acquisitive firms in 2022 compared to 0.034% of the total word count in non-acquisitive firms. These non-contractual intangible items reflect mainly internally generated intangibles. This perhaps reflects an attempt by non-acquisitive firms to supplement non-recognition of intangible assets with disclosure. The level of disclosure does, however, remain small. We also find that the overall level of disclosure is higher in non-acquisitive firms, but only marginally (0.067% in 2022 in non-acquisitive firms compared to 0.061% in acquisitive firms). Overall, the trend of disclosure is stable across the years for both samples, with some variation in non-acquisitive firms in 2019. Specifically, there is a spike of disclosure in 2019 in non-acquisitive firms for all key words (0.075% of total word count) then a decline in future years, possibly due to the COVID-19 effect.

We pick the top 10 key words by incidence in the full sample (from Table 1) and provide further details on the diversity of the incidence of these particular key words across years (Figure 3, Panel A), industries (Figure 3, Panel B), and acquisitiveness (Figure 3, Panel C). Panel A shows no evidence of an increase in disclosure of any of the key words across the years. In general, the disclosure remains stable across time. There is a large jump in the disclosure of licences and R&D in 2019, with a fall in the following years. Specifically, disclosure of R&D increases from 0.015% to 0.018% of the total word count between 2017 and 2019, but declines to 0.016% in the remaining three years. This could potentially be due to additional disclosures in 2019 at the beginning of the COVID-19 pandemic, which required discussions about its impact on R&D activities and possible write-downs of its value as well as that of licences. R&D is by far the most commonly disclosed intangible across the years, followed by licences and brands.

Panel B sheds light on the variation in key word usage across different industries, suggesting sector-specific approaches to intangible asset reporting. Notably, the services sector features high disclosure of R&D (0.022% of total word count) and licences (0.015% of total word count). The manufacturing sector also features high disclosure of R&D (0.021% of total word count). The transportation sector has high disclosure of franchises (0.009% of total word count) and licences (0.007% of total word count). Business model is mentioned most by firms in the retail sector, but disclosure remains low at 0.005% of the total word count in 2022.

Lastly, Panel C contrasts the disclosure in acquisitive and non-acquisitive firms; the figure shows discrepancies in key word frequency between acquisitive and non-acquisitive firms, which could reflect differing asset capitalization and internal development. Acquisitive firms tend to report higher instances of brands, computer software, and customer relationships, possibly due to the integration and valuation of acquired assets, whereas non-acquisitive firms show a higher count of licences, R&D, and intellectual property, emphasizing organic growth and proprietary assets. This contrast in reporting practices shows the interplay between firm growth strategies and intangible asset disclosure.

Multivariate Analyses

Table 5 presents the results of regressions (1) and (2) investigating the relationship between the extent to which intangible assets are recognized and disclosed and the acquisitiveness of the firm, controlling for factors that would influence recognition of intangible assets and disclosure of intangible items. Disclosure is encapsulated within different models based on the nature of the intangibles—contractual (${\mathit{INT}}_{\mathit{DiS}\_A}$), non-contractual (${\mathit{INT}}_{\mathit{DiS}\_B})$, broad (${\mathit{INT}}_{\mathit{DiS}\_C})$, and an aggregate measure of all types (${\mathit{INT}}_{\mathit{DiS}\_\mathit{ALL}})$. Focusing on recognized intangibles in column (1), the analysis indicates a statistically significant positive correlation with acquisitive behaviour, evidenced by a coefficient of 0.188 (p < 0.001), suggesting that firms engaging in acquisitions are more inclined to recognize intangible assets. This is expected and is in line with the univariate results presented in earlier sections as well as the requirements under IAS 38. In column (2), where the dependent variable is disclosure of contractual intangibles (${\mathit{INT}}_{\mathit{DiS}\_A})$, acquisitive companies exhibit a tendency towards more disclosure but the effect is not significant (coefficient of 0.003, p = 0.193). This trend is the opposite for non-contractual intangibles (column (3)), where non-acquisitive firms have a higher level of disclosure, a relationship affirmed by a coefficient of –0.009 (p < 0.001). In column (4), we find that acquisitive firms tend to disclose more of the broad intangible items (coefficient of 0.006, p < 0.001). In terms of overall disclosure in column (5), we find no significant differences between acquisitive and non-acquisitive firms (coefficient of 0.000, p = 0.932). Overall, it appears that non-acquisitive firms supplement non-recognition of intangible assets by increasing disclosure of non-contractual intangible items. Given that non-acquisitive firms invest in higher amounts of R&D expenditures (see Table 3), they may be providing details of these expenditures and other intangible items in their reports.

In terms of the control variables, we find that the age of a firm presents an inverse relationship with the recognition of intangible assets, with a coefficient of –0.001 (p = 0.003) in column (1), suggesting that older firms are more conservative in recognizing such assets. Moreover, age positively affects the breadth of disclosure of contractual and non-contractual intangibles, which may reflect a more traditional approach to financial reporting among established entities. The variable representing the size of the firm, as measured by the market value of equity, does not show a uniform influence across the models, indicating that larger firms might not consistently translate their size into more recognition or disclosure of intangible assets. In terms of leverage, firms with higher leverage have a significantly higher propensity to recognize intangible assets, evidenced by a coefficient of 0.253 (p < 0.001) in column (1). They also disclose more contractual and broad intangibles (coefficient of LEV is 0.010, p = 0.037 and 0.004, p = 0.091 in columns (2) and (4), respectively). However, this propensity does not extend to disclosure of non-contractual intangibles in column (3), where higher leverage correlates with less transparency (coefficient of –0.021, p < 0.001). This cautious approach to disclosure may be a strategic decision to mitigate financial vulnerability from stakeholders’ scrutiny.

Interestingly, the variable that represents competitiveness within the industry, the Herfindahl-Hirschman Index (HHI) has a pronounced effect on the disclosure. Specifically, firms in less competitive (more concentrated) industries are less willing to disclose non-contractual intangibles (${\mathit{INT}}_{\mathit{DiS}\_B})$ (column (3)), which tend to include more internally generated intangibles, which may imply protecting themselves against new market entrants and an unwillingness to disclose proprietary information (coefficient of –0.020, p < 0.001). The same is true for ${\mathit{INT}}_{\mathit{DiS}\_C}$ and ${\mathit{INT}}_{\mathit{DiS}\_\mathit{ALL}}$, which both have negative significant HHI coefficients (coefficient of –0.006, p = 0,011 and a coefficient of –0.018, p = 0.014 in columns (4) and (5), respectively).

Capital intensity is negatively associated with recognition (column (1)) and disclosure (columns (2)–(4)) of intangible assets. For example, the coefficient for PPE in column (1) is –0.516 (p < 0.001). The high capital intensity likely signifies a focus on tangible investments, aligning with less need or capacity to identify intangible assets. Audit-related variables, such as audit fees, exhibit a nuanced impact. Specifically, higher audit fees and the presence of a qualified audit opinion correlate with less recognition of intangible assets, but have no association with any of the disclosure proxies.

The results above indicate that acquisitive firms recognize more intangible assets but disclose fewer non-contractual intangible items in their reports. To further test whether there is a substitution effect between recognition and disclosure, the regression analysis (model 3) in Table 6 controls for the intensity of intangible asset and R&D expense recognition. This offers insights into the relationship between acquisitiveness and the disclosure of intangible assets, while controlling for intangible asset recognition and other firm characteristics. Starting with the model for disclosure of contractual intangibles (${\mathit{INT}}_{\mathit{DiS}\_A}$) in column (1), the coefficient for acquisitive firms remains insignificant (0.000, p = 0.977). This suggests that disclosure of contractual intangibles is similar across both groups of firms. The coefficient of ${\mathit{INT}}_{\mathit{REC}}$ is positive and significant (0.016, p < 0.001), in line with the guidance under IAS 38 to discuss further details of recognized intangible assets in the accompanying notes to the financial statements. We do not find a significant association with R&D intensity (coefficient of 0.000, p = 0.335).

When considering non-contractual intangibles (${\mathit{INT}}_{\mathit{DiS}\_B}$) in column (2), the findings show that acquisitive firms disclose less, reflected by a coefficient of –0.010 (p < 0.001). This supports our findings above, that non-acquisitive firms supplement non-recognition of intangible assets by providing further details on R&D expenses and other non-contractual intangibles. This holds even after controlling for intangible asset recognition (coefficient of 0.010, p = 0.006) and R&D expenses intensity (coefficient of 0.001, p = 0.003). The same holds for the aggregate measure of disclosure (${\mathit{INT}}_{\mathit{DiS}\_\mathit{ALL}}$) in column (4), showing that acquisitive firms tend to disclose less, reflected by a coefficient of –0.007 (p = 0.040).

In contrast, the results for disclosure of broad intangibles (${\mathit{INT}}_{\mathit{DiS}\_C}$) in column (3) indicate that acquisitive firms tend to disclose more than non-acquisitive firms (coefficient of 0.003, p = 0.008). Hence, acquisitive firms may discuss matters such as customer relationships, business models, or other broad items that may be considered intangibles.

In terms of other firm characteristics, the age of the firm is positively related to the extent of disclosure of all items, other than broad intangibles (column (3)). The relationship potentially illustrates the conservative nature of less established firms when it comes to disclosing internally generated intangibles, where no market exists, and ownership rights are less defined. The negative association with leverage in column (2), with a coefficient of –0.023 (p < 0.001) hints at a cautious approach by firms in disclosing obligations that could expose financial stress.

In terms of competitiveness, we find similar results as in Table 5 regarding the alternative intangible disclosure measures. The coefficient of the Herfindahl-Hirschman Index (HHI) is negative and significant for all proxies other than for contractual intangibles (${\mathit{INT}}_{\mathit{DiS}\_A})$in column (1). For example, the coefficient is –0.020 (p < 0.001) in column (2), where ${\mathit{INT}}_{\mathit{DiS}\_B}$is the dependent variable, indicating an unwillingness for firms in less competitive (more concentrated) industries to reveal proprietary information.

Overall, we find that acquisitive firms, although they recognize more intangible assets, disclose less information related to non-contractual intangible items. Since these include internally developed intangible items and development costs, the implication is that non-acquisitive firms supplement non-recognition of assets with additional disclosure of intangible items.

Findings

Matching the sample with available price data reduces it to 1,450 firm-year observations, which we use in our analysis. The findings from models (4), (5), and (6) in the full sample are presented in Table 7. These provide insight into the interplay between intangible asset disclosures (DISC) and firm valuation metrics. The results of model (4) are presented in column (1) (including the overall disclosure measure) and column (2) (including the three separate disclosure measures). In column (1), we find that, consistent with the literature, the core financial metrics of earnings per share (EPS) and the book value of equity (BVE) exhibit strong and significant associations with firm value (p < 0.001), reinforcing their fundamental role in equity valuation, with more weight placed on earnings than book value (coefficient of 2.147 for EPS, compared to 1.900 for BVE). In addition, disclosure (DISC_ALL) has a positive and significant association with price (coefficient of 1.229, p = 0.008), indicating that aggregated disclosures provide decision-useful information to investors.

In column (2), including the separate measures of disclosure, we find that the positive value relevance of disclosure found in the previous regression is driven by non-contractual disclosures (coefficient for DISC_B is 1.444, p = 0.003), whereas the remaining two disclosure measures are not significant in the regression. This implies that investors only value disclosures related to non-contractual items such as internally generated intangibles and R&D costs. The other types of disclosures are not informative beyond the financial metrics of the firm.

In column (3), we divide the book value into its components: intangible assets, other assets, and liabilities as in model (5). We find that recognized intangible assets (INTANG) shows a significant positive relationship with price, highlighting the informational relevance of these assets for investors (e.g., Alshehabi et al., 2024). The same holds for OA and LIAB, which are significantly related to share price. Furthermore, the overall disclosure measure (DISC_ALL) is significantly associated with firm value (coefficient of 1.677, p < 0.001), indicating that aggregate disclosures provide decision-useful information to investors beyond recognized intangible asset amounts.

The results in column (4) present the regression results of model (5) with the separate disclosure measures. The results once more indicate that investors only value disclosures related to non-contractual intangibles (coefficient for DISC_B is 1.978, p < 0.001), whereas they place no significant value on the other two types of disclosure.

We investigate the moderating effect of the disclosure on the value relevance of EPS and BVE in model (6) which is presented in columns (5) (overall disclosure measure) and (6) (separate disclosure measures) of Table 7. In column (5), we find that there is a negative association with EPS in firms that disclose higher amounts of information on intangibles. Specifically, the coefficient of EPS*DISC_ALL is negative and significant (coefficient of –5.448, p < 0.001). This suggests that these disclosure practices might signal risk or uncertainty, which may impact future earnings. These findings align with prior literature, such as Baboukardos (2018), who emphasizes the importance of aligning disclosure strategies with investor expectations. Conversely, there is a significant positive coefficient for BVE*DISC_ALL (coefficient of 0.869, p < 0.001), which highlights the role of disclosure in enhancing the informational relevance of book value for equity valuation. This finding emphasizes the complementary nature of financial and non-financial disclosures. These results also corroborate the findings of Barth et al. (2023), who highlight the importance of detailed disclosure in amplifying the value relevance of traditional accounting metrics. Hence, high disclosure of intangible information provides useful information about the firm’s book value but also potentially creates uncertainties about the usefulness of earnings, which leads to investors relying less on this metric.

In column (6), including the three separate disclosure measures, we find that the negative moderating effect on EPS is driven by contractual and broad intangible information (coefficient for EPS*DISC_A is –4.151 and for EPS_DISC_C it is –4.616, both significant at p < 0.001). In addition, the positive effect on BVE is driven by non-contractual intangible disclosure (coefficient for BVE_DISC_B of 0.798, p < 0.001). Hence, the additional information revealed through disclosure related to book value is driven by internally generated intangibles, R&D, and similar intangible items.

Overall, these findings reveal that investors value disclosures related to intangibles that may not be recognized as intangible assets, given the strict requirements under IAS 38 for capitalizing internally generated intangibles.

In Table 8, we provide results for model (6) across acquisitive and non-acquisitive firms to further understand the contextual impact of these disclosures. In column (1), the results for acquisitive firms indicate that investors of firms that disclose more intangible information rely less on earnings in their valuation (coefficient for EPS*DISC_ALL of –7.177, p < 0.001), which may imply information about future poor performance is revealed, in line with findings in Gregory (1997). However, disclosure practices enhance the value relevance of book value of equity (coefficient for BVE*DISC_ ALL of 0.798, p < 0.001), reflecting stronger informational alignment with firm valuation in acquisitive firms. In column (2), we see that the negative effect of disclosure on the value relevance of earnings is driven by both contractual and broad intangibles (coefficient for EPS*DISC_A of –7.116 and for EPS*DISC_C of –5.667, p < 0.001 for both), whereas there is a positive impact on relevance of both earnings and book value through disclosure of non-contractual intangibles (coefficient for EPS*DISC_B of 1.209, p = 0.075 and for BVE*DISC_B of 0.529, p = 0.010). Hence, disclosure of non-contractual intangibles is value enhancing in acquisitive firms.

In contrast, in non-acquisitive firms, in column (3), there is no negative coefficient for the interaction term of disclosure with earnings, but BVE*DISC_ALL remains positive and significant (coefficient of 2.578, p < 0. 001), reinforcing the finding that disclosure of intangible information reveals further information about book value. In column (4), surprisingly, we find that the positive impact on the value relevance of book value is driven by contractual and broad intangibles (coefficient for BVE*DISC_A of 2.881 and for BVE*DISC_C of 1.803, both significant at p < 0.001). Hence, investors of non-acquisitive firms value disclosures related to intangible items that would not be recognized under IAS 38 unless acquired as well as broad intangible items.

The difference between acquisitive and non-acquisitive firms further supports the view that firm-specific characteristics shape the interpretive value of disclosures, as highlighted by Onali et al. (2017). These findings contribute to the literature by emphasizing the contingent nature of value relevance, where the interplay between disclosure characteristics and firm-level strategic dynamics significantly influences market perceptions.