Prioritising disclosures in the annual report

Abstract

Drawing upon information economics, this paper presents a relative assessment of 24 of the most common disclosure items in the management commentary and notes sections of the annual report. We design and conduct an Internet survey using a large representative sample of users with an investment focus (n = 288) and preparers of annual reports (n = 89). Using cost-effectiveness analysis, an evaluation method widely used in healthcare economics, the balance between preparation costs and user satisfaction, relative to user demand is assessed. Our main findings show that corporate social responsibility and corporate governance, the least demanded disclosure items in the management commentary, are also costly items to prepare. Further, preparers do not consider indirect costs (i.e. competitive position costs and potential litigation costs) of information provided in the management commentary to be a major concern. With regard to the notes, we find that business combinations (IFRS 3), financial instruments (IFRS 7) and impairment tests (IAS 36) are highly demanded but are also among the items most costly to prepare, and users are less satisfied with these notes. The findings have important implications for practitioners and policy-makers and can be used for setting priorities.

Keywords:

• annual report
• cost-effectiveness analysis
• disclosure
• management commentary
• IFRS
• notes

Acknowledgements

We acknowledge financial support from PwC. We would like to thank Vivien Beattie (the editor), Pauline Weetman (the former editor) and an anonymous reviewer for very constructive comments. We further thank Grace Huang, Henrik Steffensen, Jakob Uldall, Jeppe Christoffersen, Kim Füchsel, Lars Kiertzner, Mei Hua, Pernille Stokholm Bøgh, Peter Andersen, Simone Stæhr, participants at the PwC annual report seminar, the annual meeting of FSR and the EFRAG/IASB outreach event at Dansk Regnskabsforum for very helpful comments and assistance.

Notes

1. The UK Financial Reporting Council defines clutter as (a) immaterial disclosures that inhibit the ability to identify and understand relevant information and (b) explanatory information that remains unchanged from year to year.

2. The evaluation of disclosures by examining both current disclosure practice and user assessment of disclosure importance is, in principle, similar to the study by Hooks et al. (2002).

3. The survey includes users with an investment focus as these are specified as the main users of the annual report. Other users may be more interested in CSR and corporate governance disclosures.

4. The issue of litigation costs is primarily of interest in countries in which these costs are considered to be high, such as in the USA.

5. There may be a range of combinations based on different levels of cost, satisfaction and demand. For simplicity, we only illustrate the eight combinations based on high and low levels. The responses to other combinations are likely to fall in between the indicative responses in the table.

6. EFRAG and ASB (2011) mainly consider the effects of the introduction of an accounting standard or amendment and therefore suggest that effects are considered relative to the status quo.

7. We attempt to motivate an approach in which the resource allocation to a set of items is considered. WHO (Edejer et al. 2003) acknowledges this as a ‘CEA approach’. This is most explicit in its focus on the use of CEA in allocating a fixed budget between interventions.

8. A statement in which the management and the supervisory board declare that the annual report is prepared in accordance with the legislation and any further accounting and reporting requirements and that the financial statements provide a true and fair view.

9. We consulted with PwC to ensure (a) that the 24 disclosure items and their explanatory descriptions reflected disclosure requirements and common practice, (b) that the appropriate terms were applied and (c) that the questions were easily understood.

10. The adjusted scores may further be adjusted by the standard deviation across items. However, as Fischer (2004) indicates, this adjustment is only relevant if an extreme response bias is expected.

11. We acknowledge that cost-effectiveness ratios (rather than differences) are often used in CEA in order to analyse the balance between cost and effectiveness. We have decided to use a difference measure for two reasons. First, a ratio-based measure will provide misleading information as the mean adjusted scores can be both positive and negative. Second, we further consider the use of negative/positive scores as helpful for the presentation of the cost-satisfaction score.

12. A copy of the questionnaire is available on request from the authors.

13. We do not report the results from the two types of bank respondents separately as the results are qualitatively the same.

14. Because we are interested in individuals' views on the management commentary and the notes, we allow more than one observation for each firm. We acknowledge that this could potentially bias the results for professional investors if individuals from the same firm provide identical (firm-specific) answers. We have therefore also run tests where only one observation from each firm is included. However, the results remain similar to the ones reported in this study.

15. Graham and Harvey (2001) and Graham et al. (2005) obtain response rates of 9.0% and 10.4%, respectively. Vergoossen (1993) obtains a response rate of 43%, Gassen and Schwedler (2010) obtain a response rate of 1.9% and Beattie and Smith (2012) obtain a response rate of 9.3%.

16. We considered applying the ‘100-point’ scale for disclosure items in notes. However, preliminary tests revealed that the increase in items to 15, which had to be evaluated simultaneously, was problematic as users had difficulties in allocating the 100 points across 15 notes.

17. This result is most likely driven by the fact that private investors are not as trained as professional investors in annual report analysis and therefore have a need for more processed information.

18. Results not tabulated suggest that the preparation costs of notes are the same across industries.

19. The pilot test of the questionnaire showed that both accounting specialists and preparers considered the indirect costs of notes to be immaterial. As the length of the questionnaire was critical, we decided not to include 15 additional questions that could capture the indirect costs of notes. Thus, we have not collected information on the indirect costs of notes.

20. We have also included indirect costs of disclosure in the management commentary as part of the robustness check. The results remain similar.