Measuring financial reporting quality: an approach based on qualitative characteristics

Type of content
Theses / Dissertations
Publisher's DOI/URI
Thesis discipline
Accountancy
Degree name
Doctor of Philosophy
Publisher
University of Canterbury
Journal Title
Journal ISSN
Volume Title
Language
English
Date
2020
Authors
Rathnayake Mudiyanselage, Saman Bandara
Abstract

I have developed a Financial Reporting Quality (FRQ) measurement index within the scope of the 2018 Conceptual Framework for Financial Reporting of the International Accounting Standards Board (IASB), and I used it to measure FRQ of annual reports from Sri Lankan listed companies. My study is motivated by i) the seminal work of Beest, Braam, & Boelens (2009) who used Qualitative Characteristics (QCs) to measure FRQ, ii) the lack of a comprehensive measurement tool from which to quantitatively derive the degree an annual report complies with the postulated (by the IASB) characteristics of decision-useful information, and iii) the different classification interpretations of QCs and the inconclusive results about the perceived importance user groups ascribe to the QCs within decision usefulness theory: useful to whom and useful to make what decisions.

A first important realisation to make is that QCs and FRQ are latent constructs, which immediately suggest that the relationship between QCs and FRQ may be complex, non-linear and hierarchical. The process of developing the FRQ measurement index is then formulated through Research Question (RQ) 1, in which I use three steps. In Step 1, I searched the literature to identify measures for the QCs, and I obtained 54 so-called sub-information items under 17 information dimensions. In Step 2, I surveyed Sri Lankan investment (N=235) and lending (N=214) decision-makers on the usefulness of the identified sub-information items to their particular decision roles, and the respondents validated the selection identified in Step 1. In Step 3, the structural relationships between the 54 sub-information items, the 17 information dimensions, the 6 QCs and FRQ were tested by confirmatory factor analysis using SmartPLS. The factor analysis results revealed that the 54 sub-information items are measures of the 17 information dimensions and that they each factorise statistically satisfactorily with one of the 6 QCs.

The 2018 Conceptual Framework postulates a particular 2-group (fundamental and enhancing) classification the 6 QCs belong to. I thus have tested the postulated classification and also formed and tested 2 alternative models of how the 6 QCs affect FRQ. The results revealed that enhancing QCs affect FRQ indirectly through fundamental QCs, as postulated by the Conceptual Framework, but importantly they also make strong and significant direct contributions to FRQ. In particular, understandability has the highest direct contribution to FRQ from all 6 QCs. This finding challenges the IASB 2-group classification. A further utility of the 3 models, which in essence are variants of an FRQ measurement index, is the explicit relative contributions obtained that each of the QCs makes towards FRQ.

In supporting the development and validation of the FRQ measurement index, in RQ2, I also investigated several secondary research questions. I surveyed Sri Lankan investing (N=235) and lending (N=214) decision-makers to examine their use of annual reports, their perceived importance of QCs, and their perceived impact of International Financial Reporting Standards (IFRS) on FRQ. My results revealed that on average and ahead of ‘annual reports’, lending decision-makers rate highest ‘the direct communication with clients’, and investment decision-makers rank ‘stock market publications’ as the prime source for investment decisions; within annual reports, both types of decision-makers identified financial statements as the most useful sections and both groups stated that the main factor that restricts the usefulness of annual reports is the delay in publishing annual reports after year-end. When asked directly, both groups challenged the IASB’s current classification of QCs into ‘fundamental’ and ‘enhancing’, and both groups identified understandability as the most important QC, followed by timeliness. Relevance ranked sixth and last, surprisingly. These results complement the findings from RQ1. With respect to the impact of IFRS adoption in Sri Lanka in 2012, both groups believe that FRQ improved compared to the earlier Sri Lanka Accounting Standards (SLAS) reporting regime.

In RQ3, I also put in practice the derived FRQ measurement index by assessing the FRQ of annual reports of 53 listed Sri Lankan companies for the years 2010, 2014 and 2018. I find that Sri Lankan companies recorded on average an FRQ of 56% in 2010, rising to 61% in 2014 and to 66% in 2018. These differences are statistically significant, which allows me to conclude that the FRQ of Sri Lankan entities improved after IFRS adoption in 2012 compared to the period before adopting IFRS. This result complements the finding in RQ2. I identified that the total number of pages, the size of the firm as measured by total assets, and her market capitalization all positively correlate with the level of FRQ.

Through my work, I have made several useful contributions: I challenge the classification of QCs as fundamental and enhancing, which should also lead to a re- examination of the interpretation various authors of accounting textbooks give to this issue in the corresponding ‘IFRS and Conceptual Framework’ chapters; my results further challenge the widely held assumption that relevance and faithful representation rank supreme in the importance ranking among the 6 QCs; next, I provide numerical equations with which i) users can measure, i.e. calculate, FRQ and the change in FRQ over time, and ii) the IASB can measure to which degree their objective has been achieved of setting standards intended to improve the quality of decision-useful information for investors and lenders. While the processes for the derivation of an FRQ measurement index apply generally, the data have been collected and obtain within the Sri Lankan context. Thus, I invite other researchers to use, test and validate the measurement of FRQ in jurisdictions of their interest.

Description
Citation
Keywords
Ngā upoko tukutuku/Māori subject headings
ANZSRC fields of research
Rights
All Rights Reserved