Date: 7 July 2020
Where: Australia (virtually)
Board Member Ann Tarca delivered a speech at the virtual annual conference of the Accounting & Finance Association of Australia and New Zealand (AFAANZ). She discussed developments in digital reporting, including tagging of financial statements.
Did your life go online from March 2020? Or perhaps a significant part of your life was already online before March 2020. Some online activities move the physical activity into the virtual world, but the activity remains the same. My mother and I play Scrabble online and it’s the same game we’ve been playing for 40 years.
Are you reading company annual reports online via a PDF from the company’s website? That PDF looks much like a paper annual report, those weighty documents that were once posted to shareholders. For many Australian companies the digital part of the annual report is in the information system that produced the report. Companies’ internal accounting systems have been revolutionised over the past 40 years using sophisticated technology.
We have a technology, XBRL (eXtensible Business Reporting Language), and a taxonomy, the IFRS Taxonomy, that can be used to tag the financial information in a company’s report to make the information available in a structured electronic format. Why have we, in Australia and elsewhere, been so slow to take advantage of the technology and digitalise financial statements for the benefit of users?
My talk today is about digital reporting, about its benefits and costs. First, I will talk about what XBRL is and who uses it. Then I will ask four questions and look at different stakeholder groups' relations to tagged electronic reporting—listed companies and their auditors; accounting standard-setters and regulators; investors, analysts and data aggregators; and academics—teachers and researchers. At the end, I will mention some opportunities for further research that I see.
I will pose many questions for you. You are practitioners and academics—you’re good at answering questions. I challenge you to take on some of these questions—to think about them and to find answers. I’m a standard-setter. I have to think about these questions and about associated costs and benefits. I hope your research will help me and other standard-setters and regulators make those cost/benefit decisions.
XBRL is an international standard for digital business reporting. XBRL International, the organisation behind XBRL, describes XBRL as:
…a language in which reporting terms can be authoritatively defined. Those terms can then be used to uniquely represent the contents of financial statements or other kinds of compliance, performance and business reports. XBRL lets reporting information move between organisations rapidly, accurately and digitally.1
XBRL International says:
An annual report in PDF compared to an XBRL annual report is like comparing film photography to digital photography or a paper map to a digital map. XBRL tagging of financial statements is like ‘barcodes for financial reporting’.
Inline XBRL (iXBRL) is a form of XBRL technology that allows us to create a document that is human readable but also shows the tagging embedded in the financial statements. Here is a slide showing the IFRS Foundation’s annual report tagged using inline XBRL. You can find the 2019 annual report on our website and you can read it without special software.2 The tagging of the financial statements is based on the IFRS Taxonomy (which provides elements to tag disclosures required by IFRS Standards).3
Now you have the IFRS Foundation digital report, what can you do with it? You can extract the data from our financial statements into Excel and compare it with financial information from other years or standard-setters’ reports, if they have also been tagged. You can do this quickly and accurately. All the items are defined, with specific attributes, so the meaning of each item is clear—you can analyse the information presented and you can compare financial position and performance between entities.
XBRL tagging is used in the US, based on a US GAAP taxonomy developed by the Financial Accounting Standards Board (FASB). From 2009 the Securities and Exchange Commission (SEC) has required tagging of financial statements, first in a voluntary programme, then mandated for all SEC registrants (ie listed companies). Initially the primary financial statements were tagged along with block tagging of footnotes. Now there is detailed tagging of notes. US companies use the US GAAP taxonomy and IFRS filers (foreign private issuers in the US) use the IFRS Taxonomy. The SEC provides a central repository for the annual reports and tools to download the tagged data.
In Europe, the market regulator, European Securities Market Authority (ESMA), has mandated the European Single Electronic Format (ESEF). It requires all companies listed on an EU regulated market to prepare an inline XBRL filing from 1 January 2020. Tagging is required for consolidated financial statements prepared using IFRS Standards. Detailed tagging will apply to the primary financial statements and block tagging to the notes. Audit requirements apply to the tagged financial statements.4 The objective of ESEF is to make reporting easier for issuers and to facilitate accessibility, analysis and comparability of annual financial reports.5
Digital financial data can also be used by regulators and tax authorities. For example, in Japan all listed companies and investment funds have filed their financial statements with the regulator (the Financial Services Agency) using XBRL since 2008.6 In the UK, company tax returns are submitted using inline XBRL based on taxonomies approved for use by the UK government, which include a UK GAAP Taxonomy.7 The Danish Business Authority is using artificial intelligence to analyse the XBRL-tagged financial statements lodged with it—more than 230,000 per year.8
In Australia, the Australian Prudential Regulation Authority requires banks to provide digital reports. Australian Securities and Investments Commission (ASIC) requires lodgement of PDF (or paper) reports and permits the use of XBRL or iXBRL.9 ASIC’s taxonomy is based on the IFRS Taxonomy,10 but Australian companies don’t lodge with tagged data.
This leads to the first question of the session.
(And that ‘we’ includes Australia and some other countries without requirements for mandatory lodgement of listed companies’ annual reports.)
First, let’s look at the proposed benefits of digital financial reporting, that is, the tagging of financial statements. These expected benefits are based on market efficiencies arising from accessibility, speed and accuracy in data transmission. Regulators, such as security market regulators, can gather data on the entities they regulate. Business can provide information to regulators, tax authorities and government agencies in an efficient way (ie greater speed and accuracy; less duplication, manual processing and consumption of resources). Data providers can collect information about companies using automated processes. Financial institutions, analysts and investors can improve their access to information that enables analysis and comparison, for example, through the detailed tagging of information in financial statement notes.
Next, let’s look at a number of stakeholders to examine the costs and benefits of digital reporting for them and to address question 1. The stakeholders to which I refer are: listed companies and their auditors; accounting standard-setters and regulators; investors, analysts and data aggregators; and academics—teachers and researchers.
Around the world, various government initiatives have required or permitted entities to file financial information in digital format. ASIC gives companies choice regarding lodgement of financial reports—paper, PDF, XBRL, iXBRL. Given a choice, listed companies need a compelling case to take on an activity that consumes resources. Tagging financial statements will involve software, systems, expertise, staff and consultants. Many in Australia believe that the compelling case exists. However companies in Australia have not generally responded—it appears no XBRL-tagged reports have been lodged with ASIC.11 Other jurisdictions that have introduced a voluntary electronic reporting programme share this experience (for example Canada).
Nevertheless, some Australian companies, notably from the healthcare sector, prepare XBRL-tagged financial statements because they are also listed in the US. Inline XBRL filings will be required by the SEC for foreign private issuers for fiscal periods ending on or after 15 June 2021.
Let's move to our second question.
Following the SEC requirements for registrants to file tagged financial statements, early research pointed to some significant problems relating to the accuracy of tagging and excessive or erroneous use of extensions.12 Later studies suggested errors have become less prevalent and concluded ‘learning’ has taken place.13 Other studies have investigated the use of extension tags—this is where the company creates its own company-specific tag for a line item because the company does not consider the taxonomy contains a suitable tag. Extension tags may be used because the company fails to find the correct tag in the taxonomy or when the company has a disclosure that is specific to its own circumstances or business model. Based on interviews with preparers, Harris and Morsefield (2012) reported that some companies viewed XBRL tagging as an unnecessary incremental cost. However, some filers reported benefits for XBRL implementation relating to their internal data analysis and financial reporting and closing processes.14
Some researchers argued that some preparers have not supported electronic filing by not giving sufficient attention to the accuracy of coding, possibly because SEC XBRL filings are not audited, nor have they been the subject of enforcement actions by the SEC. However, the SEC has provided tools to help improve the accuracy of tagging. Hoitash et al. (2020) concluded that auditing is an essential next step for electronic reporting in the US, particularly when the SEC mandates inline XBRL (2019-2021).15
The IFRS Foundation has been a long-time supporter of digital reporting—it has been working on the IFRS Taxonomy for more than 15 years. But like the IFRS Standards, the IFRS Foundation develops its Taxonomy but does not mandate its use in any jurisdiction. Taxonomy use is the decision of national regulators. So what makes a country or jurisdiction mandate a taxonomy?
XBRL International provides a detailed list of benefits from use of a taxonomy for many stakeholders in the financial reporting community. Regulators can benefit from analysis of performance and risk data about the institutions they regulate. Government agencies can simplify the reporting process and standardise the data available. Information about risk and performance can assist analysts, and investors can benefit from being able to compare companies.16
I can hear you asking: what is the evidence of benefits of tagged data for capital markets? We’ll go there shortly! First, let’s talk about investors by considering a third question.
A compelling case for requirements for companies to provide tagged financial statements would arise if investors found tagged financial statements useful. Benefits for investors could justify the costs for preparers of tagging data and having the tagged data audited. Harris and Morsefield (2012) interviewed a large number of stakeholders involved in the use of XBRL. Feedback from investors and analysts indicated that some had accessed tagged data from the SEC website or data feeds but all who tried to use XBRL data for large-sample research had negative views about the data usability and data quality (please note the interviews were conducted prior to 2012).
Considering the situation in Australia, Professor Peter Wells opined:
companies do not lodge digital financial reports because analysts have not updated their systems to use digital financial reports, and analysts are not updating their systems because no entities produce digital financial reports.17
If there is low demand from investors for regulators to make tagged data mandatory, one possible explanation is that the financial data investors use is already digital—they get it from database providers. Investors demand information that is: timely, accurate, detailed, covers many countries and years, is comparable across companies and consistent from year-to-year. Many investors have systems to draw this data directly from databases, and then feed it into their models. A CFA Institute survey in 2016 reported that 48% of respondents obtained most of the data/information used in their evaluation and analysis of companies from third party data providers, and 10% obtained all their data or information from third party providers. This survey also asked about use of XBRL-tagged data. Considering interim or annual reports lodged with a national stock exchange or securities market regulator, 13% of respondents indicated they made direct use of XBRL information and 10% indicated that they extract or import XBRL-tagged data directly into financial analysis models.18
Louis Matherne, Chief of Taxonomy Development at the FASB, noted in 2019 that although it is difficult to quantify, analysts, hedge fund managers and others are using tagged data in their systems. Others agree with his view. Christine Tan from Idaciti says that her customer base indicates the tagged data is used by academics, investors, analysts and corporations. Lou Rohman from Toppin Merrill (IT consulting services company) said that the data was not used in the early years of the SEC mandate, however investors and the SEC are now using the data and efficiencies are being realised. In his view, waiting for investors to ask for the tagged data is the wrong strategy.19
Tagging of data would likely help the database providers. Harris and Morsefield (2012) reported that data aggregators were using or experimenting with XBRL data but since tagged data was not available for a full set of companies (ie international companies) some aggregators needed to continue with their current processes for collecting and validating data. Louis Matherne stated in 2019 that he observed data aggregators were consuming the tagged data into their systems and new data aggregators have built platforms based on the tagged data.20 One would assume that data providers have benefitted enormously from the SEC’s initiatives to have data tagged because this makes it more structured, accessible and machine readable. Data providers can then focus more on ‘standardising’ and ‘normalising’ data and providing their various other value-adding activities for their clients. Data providers potentially can provide more timely and granular data for more companies to their customers because the data collection costs would be significantly reduced.
Finally, let’s look at the question you have been waiting for.
The first wave of research, following SEC requirements for registrants to lodge tagged financial statements, was very positive about the benefits of the SEC’s mandate, which made data available electronically. For example, Kim et al. (2012) examined 1,536 10-K/10-Q filings from 428 firms in 2009, the first year of mandatory XBRL use in the US. They investigated various market measures in the pre/post adoption periods, namely event return volatility (absolute value of daily abnormal returns), information efficiency (absolute deviation between actual return and expected return), and standard deviation of stock returns. They found an increase in information efficiency, a decrease in event return volatility and a reduction in change in stock return volatility. They concluded that XBRL has the potential to decrease information risk and information asymmetry through greater transparency.21
In a similar vein, researchers claimed that mandatory XBRL adoption reduced information processing costs, which reduced stock price synchronicity (the degree to which companies’ returns co-moved with market returns) by facilitating company-specific information acquisition by investors.22
Other research claimed benefits for analysts. For example, considering SEC Phase I and Phase II filers in 2009 and 2010, Liu et al. (2014) reported a significant positive association between mandatory XBRL adoption and analyst following and analyst accuracy (despite the errors in data). The associations are stronger for Phase I filers in the second year, implying learning and better tagging (block tagging in 2009, detailed tagging in 2010).23 Other research has claimed benefits for smaller investors compared to larger investors and for smaller institutions compared to larger institutions.24
However, one must approach the findings with caution because it is unclear how much capital market participants are using XBRL-tagged data in the post-SEC mandate period. We do not know to what extent analysts and investors are obtaining electronic data directly (and therefore could possibly be benefiting from it) or if they are continuing to source data from data providers, as indicated by the CFA survey I previously mentioned. If the latter is mainly true, then the research conclusions imply the data aggregators have changed the data they are providing to clients following the SEC XBRL mandate. I do not have evidence on this point, but it means the findings must be approached with caution. In addition, the timing of studies is important. The SEC XBRL mandate was in 2009. In the immediate pre-XBRL period, analyst forecast accuracy and dispersion were greatly affected by the financial crisis (2007-2008) so an improvement in the attributes of analyst forecasts could reflect problems before 2009 rather than improvements after 2009.
Hoitash et al., (2020) provide an extensive literature review and include many studies that have investigated the impact of adoption of XBRL for SEC filers.25 The authors point to mixed results in investigations of the capital market impact of XBRL in the US. They conclude this could be ‘because of limited use of XBRL data by financial statement users’. Javrin and Mascha (2014) reported doubts by financial officers that investors were using the tagged data.26 If this is the case, it makes it difficult for researchers to provide evidence about the impact of the XBRL mandate that is not affected by a range of factors, not just availability of tagged data.
To date there is limited research in jurisdictions where entities use IFRS Standards because requirements for electronic reporting using XBRL tagging are not widespread. Three such studies are described below.All claim benefits from providing XBRL-tagged data.
In Japan tagging financial statements using XBRL was mandated in 2008 for listed companies and investment funds. Bai et al. (2014) reported that adoption of XBRL improved the information environment in Japan, shown by reduction in returns volatility, absolute cumulative abnormal returns, changes in the standard deviation of returns and abnormal bid-ask spread.27
In Belgium tagging financial statements using XBRL was mandated in 2007 for listed non-financial companies using local GAAP. Based on a sample of 28,711 firm-year observations for the period 2005-2010, Liu et al. (2017) demonstrated that the reduction in information asymmetry, measured by increased liquidity, was stronger for larger companies with available resources and expertise to implement the technology. The effect was also stronger for low technology companies whose financial statements were key sources of information for investors.28
In China the Shanghai Stock Exchange and the Shenzhen Stock Exchange introduced requirements in 2008 for companies to file documents tagged using XBRL along with traditional PDF reports. For a sample of 5,203 firm-year observations starting in 2004 (pre XBRL adoption) and ending in 2012 (post XBRL adoption), Songsheng et al. (2017) found that tagging financial statements using XBRL was associated with lower information asymmetry, measured by returns during the post-earnings announcement drift period.29
The final group of stakeholders I would like to discuss are academics—teachers and researchers. Many are thought leaders, so they may have views on costs and benefits related to digital reporting. Some of our colleagues are cited in the Australian parliamentary inquiry I mentioned previously.30
Academics may be able to provide evidence relevant to policy developments such as requirements for mandatory tagging of financial statements in Australia. Researchers can show the impact of requirements in other countries that could be relevant in Australia. The way taxonomies are set up and used vary across countries. There is much to be learned about what works best and why; such evidence is not currently available.
Researchers could also investigate broader questions about whether XBRL is the best technology for digital reporting and the role of software vendors in providing tools to facilitate the use of digital reports by investors and others. To what extent is lack of these tools holding back use of electronic data? What challenges are involved in using non-standardised data and how can these be overcome? It would be useful to know more about users’ experiences of the FASB’s taxonomy and the IFRS Taxonomy. Given decades of investment into the current structure of each taxonomy, changing the approach of either would be difficult. However, are there other ways in which the taxonomies can evolve to be more useful for users?
ESMA’s ESEF initiative means a large amount of tagged data will become available. This has the potential to revitalise research about use of IFRS Standards so I look forward to seeing many new projects. It’s now 15 years since IFRS Standards were adopted in the EU, Australia and elsewhere (13 years for New Zealand). Research about application (as distinct from adoption) of IFRS Standards is challenging but very important to the ongoing work of standard-setters.
XBRL-tagged data provides a new IFRS data source—more data, more coverage, more complete, more ‘as reported’, less standardised. XBRL data gives opportunities for several reasons.
Tagged financial data opens up research opportunities by making data accessible in new ways and allowing further manipulation and analysis. Opportunities relate to comparability, quality, financial statement presentation and disclosure. I list a few examples.
Tagged data (which is structured data) provides new ways to construct comparability metrics. Promoting comparable financial information is part of the objective of many standard-setting initiatives (and of the work of XBRL International too). Comparability of information matters greatly to investors and analysts. But measuring comparability has been difficult for researchers. Opportunities provided by tagged data have been investigated by Henry et al. (2018).33 They analysed various pairs of companies to see the extent to which the same items appeared in the financial statements. They concluded that ‘structural comparability’ mediated the relationship between earnings comparability and analyst coverage, forecast accuracy and forecast dispersion.
It has always been challenging to measure the ‘quality’ of financial reporting, because the concept of quality has many dimensions and data to measure these dimensions is not easily accessible. Some researchers have investigated the use of extension tags as a way of investigating quality. For example, does use of extensions improve usefulness of financial statements or does it detract from financial statements? Debreceny et al. (2011)34 studied use of extensions by US filers in 2009-2010. They concluded 40% of extensions were unnecessary, suggesting a negative outcome for comparability and quality of reporting. Other researchers have used tagged data to explore financial reporting complexity. For example, Hoitash et al. (2018) created a complexity measure based on the number of tagged items, which they concluded outperforms other measures of complexity.35
XBRL data is structured. All the line items sit under headings and directly reflect the presentation and disclosure requirements of IFRS Standards. Researchers can investigate what and where items are presented for selected topics and companies of interest. There is also scope to take new approaches to investigating policy choice and compliance with accounting standards because the tagged data can provide a manageable way to get a comprehensive picture of firms’ recognition, measurement and disclosure practices. For example, there is diversity in accounting for exploration and evaluation expenditure by companies with extractive activities. Many of these companies are listed in the US (US GAAP filers and IFRS filers) so tagged data is available. To what extent are exploration and evaluation expenditures capitalised and/or written off? How does this differ between companies, industries and over time? What is the usefulness of the information (recognised and disclosed) to investors?
Notes may be block tagged and/or detail tagged. XBRL gives researchers opportunities to find and analyse the notes more easily and to look into the notes using technology-based tools. This has huge potential compared to manual coding of disclosure data. More analysis is possible on a larger sample. For example, consider the lengthy and complex notes associated with financial instruments disclosures. With XBRL filings, researchers can locate notes more easily (even when they are spread throughout the report), identify particular line items of interest, along with amounts and details (such as classifications of level 1, 2 or 3 fair value). They will be able to locate, search and analyse the narrative content of the text blocks and to use natural language processing and other text analysis tools on the text. There will be many opportunities for research relevant to standard-setting and, consequently, opportunities for you to develop and demonstrate the impact of your research. Research about tagged data has huge scope—it is relevant to all countries with capital markets. It involves a broad range of stakeholders and it stands at the bridge between academia and practice. Researchers in the field can contribute evidence relevant to practice as well extending academic literature.
Thank you for the opportunity to address you today on the topic of digital financial reporting. I hope the presentation has given you many things to think about. I’ve posed some questions for your ongoing consideration. One final question—are we, as members of the accounting profession, change facilitators? Do we lead thinking and action on new and better ways to do things? I think we can do more to facilitate digital reporting.
A recent inquiry has recommended that the Australian Government take appropriate action to make digital reporting standard practice in Australia.36 As members of the accounting profession—practitioners, teachers and researchers—this is an area where we can make a valuable contribution. I hope you will be change facilitators in the discussions ahead on this initiative and that your research will provide evidence that leads to sound policy making.
Thank you and good afternoon.
6Bai, Z., Sakaue, M. and Takeda, F. (2014), 'The impact of XBRL adoption on the information environment: Evidence from Japan', The Japanese Accounting Review, 4, 49-74.
11See Parliamentary Joint Committee on Corporations and Financial Services. (2020) Regulation of Auditing in Australia: Interim Report. February. pp. 101-106. Available at https://parlinfo.aph.gov.au/parlInfo/download/committees/reportjnt/024330/toc_pdf/RegulationofAuditinginAustralia.pdf;fileType=application/pdf
12Debreceny, R., Farewell, S., Piechocki, M., Felden, C. and Graning, A. (2010) ‘Does it add up? Early evidence on the data quality of XBRL filings to the SEC’. Journal of Accounting and Public Policy, 29(3), 296-306.
13Bartley, J., Chen, A., and Taylor, E. (2011) 'A comparison of XBRL filings to corporate 10-Ks--Evidence from the voluntary filing program', Accounting Horizons, (2), 227-45; Debreceny, R., Farewell, S., Piechocki, M., Felden, C., Gräning, A. and d'Eri, A. (2011), 'Flex or break? Extensions in XBRL disclosures to the SEC', Accounting Horizons, 25 (4), 631-57; Du, H., Vasarhelyi, M., and Zheng, X. (2013), 'XBRL mandate: Thousands of filing errors and so what?', Journal of Information Systems, 27 (1), 61-78.
14Harris, T. and Morsefield, S. (2012) An Evaluation of the Current State and Future of XBRL and Interactive Data for Investors and Analysts. Columbia Business School. Center for Excellence in Accounting and Security Analysis White Paper Number Three. Available at https://academiccommons.columbia.edu/doi/10.7916/D8CJ8NV2.
15Harris and Morsefield (2012) ibid; Hoitash, R., Hoitash, U. and Morris, L. (2020) ‘eXtensible Business Reporting Language: A review and directions for future research’. Working paper, available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3595401
17Parliamentary Joint Committee on Corporations and Financial Services. (2020) op. cit. pp. 102.
18CFA (2016) ‘CFA Institute Member Survey: XBRL’. CFA Institute. See p. 7 and p. 14. Available at cfainstitute.org/survey/survey_extensible_business_reporting_language_xbrl.pdf
19Toppan Merrill Insights (2019) ‘Ten Years of XBRL: Financial Reporting Experts Reflect on Benefits, Successes and Remaining Challenges (Part 1)’. June. Available at https://blog.toppanmerrill.com/blog/ten-years-of-xbrl-part-one
20Toppan Merrill Insights (2019) ibid.
21Kim, J., Lim, J.H. and No, W. (2012), 'The effect of first wave mandatory XBRL reporting across the financial information environment', Journal of Information Systems, 26 (1), 127-53.
22Dong, Y., Li, O., Lin, Y. and Ni, C. (2016), 'Does information-processing cost affect firm-specific information acquisition? Evidence from XBRL adoption', Journal of Financial and Quantitative Analysis, 51 (2), 435-62.
23Liu, C., Wang, T. and Yao, L. (2014), 'XBRL’s impact on analyst forecast behavior: An empirical study', Journal of Accounting and Public Policy, 33 (1), 69-82.
24Blankespoor, E., Miller, B. and White, H. (2014), 'Initial evidence on the market impact of the XBRL mandate', Review of Accounting Studies, 19 (4), 1468-503; Bhattacharya, N., Cho, Y. and Kim, J. (2018), 'Leveling the playing field between large and small institutions: Evidence from the SEC's XBRL mandate', Accounting Review, 93 (5), 51-71.
25Hoitash et al. (2020, p. 16) op. cit.
26Janvrin, D. and Mascha, M. (2014), 'The financial close process: Implications for future research', International Journal of Accounting Information Systems, 15(4), 381-399.
27Bai et al. (2014) op. cit.
28Liu, C., Luo, X. and Wang, F. (2017), 'An empirical investigation on the impact of XBRL adoption on information asymmetry: Evidence from Europe', Decision Support Systems, 93, 42-50.
29 Songsheng, C., Jun, G. and Xiaoxiao, T. (2017), 'XBRL Implementation and Post-Earnings—Announcement Drift: The Impact of State Ownership in China', Journal of Information Systems, 31 (1), 1-19.
30Parliamentary Joint Committee on Corporations and Financial Services. (2020) op. cit.
31Boritz, J. and No, W. (2013), 'The quality of interactive data: XBRL versus Compustat, Yahoo Finance, and Google Finance', Working paper, available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1288376
32Li, K. and Sloan, R. (2017) ‘Has goodwill accounting gone bad?’ Review of Accounting Studies 22, 964–1003.
33Henry, E., Liu, F., Yang, S. and Zhu, X. (2018), 'Structural comparability of financial statements', Working paper, available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3133324
34Debreceny et al. (2011) op. cit.
35Hoitash, R. and Hoitash, U. (2018), 'Measuring accounting reporting complexity with XBRL', The Accounting Review, 93 (1), 259-87.
36Parliamentary Joint Committee on Corporations and Financial Services. (2020) op. cit.