International Variations in Tax Disclosures

Abstract

We examine the tax disclosures of 161 large IFRS-reporting companies in five countries by studying in detail how they carry out two numerical reconciliations mandated by IAS 12. Using a variety of approaches, we conclude that there are systematic differences in IFRS reporting practice between companies from different countries. However, we also find varied reporting practices within certain industries such as the extractive industry. Based on our observations, we question whether the requirements of the present standard are sufficiently explicit and concise to lead to understandable and comparable tax disclosures. We identify a number of elements of IAS 12 that could be improved.

Acknowledgements

Previous versions of this paper have been presented at workshops at the BI Norwegian Business School, the University of Stavanger, and at the 2011 AAA annual meeting in Denver. The authors are grateful for comments from John Christian Langli, Andrew Lennard, Christian Stadler, Maria Wieczynska, two anonymous reviewers and the editor of the journal.

Notes

1 In March 2009, the IASB issued an exposure draft (ED/2009/2 Income tax) for a new standard to replace IAS 12. Some of the companies whose accounting practice is studied in this paper had a balance sheet date for the 2008/2009 financial statements subsequent to the publication of the ED, and hence they may have been influenced by it. However, the disclosure requirements of the exposure draft are nearly the same as in IAS 12 for the elements discussed in this paper.

2 IAS 12, para. IN2. Among other things, all timing differences are included as temporary differences.

3 IAS 12 does not explain what to do with deferred tax assets and liabilities in the case of deconsolidation of a subsidiary following a loss of control. Normally, the consolidated statements contain deferred tax assets and liabilities arising from temporary differences of the assets and liabilities of each subsidiary. In the absence of a clear exemption from income recognition, the derecognition of a deferred tax liability through deconsolidation should create a deferred tax income. However, the company sample of this study shows a widespread practice of treating the derecognition of such items in the same way as their initial recognition, i.e. as a balance sheet entry only. Any other solution is hardly meaningful. In a going concern context, a deferred tax liability in a subsidiary will gradually become deferred tax income as temporary differences reverse, but in a deconsolidation such a reversal does not take place. Nevertheless, even though the balance sheet entry seems to be the only sensible solution at a deconsolidation, IAS 12 could usefully be made more explicit on this point.

4 Hanlon writes in the context of US GAAP, but the IFRS requirement is very similar.

5 An example of IFRS tax note information being used in accounting research is Kager et al. (2011) who calculate tax values of balance sheet items from data on temporary differences.

6 As a result of requirements for uniformity of accounting policies (e.g. IFRS 10, para.19), the inclusion of accounting numbers from foreign subsidiaries might reduce BTC. For example, a group reporting under IFRS would have to adjust (to first-in-first out or weighted average) any inventories which had been measured on an last-in-first-out basis by a US subsidiary.

7 For example, the up-lift of subsidiaries' assets to fair value on consolidation creates (or increases) a book-tax difference.

8 As mentioned in Section 2, EFRAG/ASB (2011) proposes a standardised format for the reconciliation of tax expense to pre-tax profit. The format consists of: (1) income exempt from taxation, (2) non-deductible expenses, (3) effect of tax losses (with three subcategories), (4) effect of foreign tax rates, (5) effect of tax rate changes, (6) prior year adjustments and (7) other items. Our format resembles closely the EFRAG/ASB proposal, with somewhat more disaggregation.

9 We define foreign as meaning not legally registered in the country. For example, we exclude from the French sample, Belgian-registered companies that prepare IFRS statements under Belgian law.

10 That is, 255 companies in the indexes, less 16 foreign, less 22 lost from the indexes due mainly to takeovers.

11 We used English language reports in all cases. Work by one of the authors on smaller companies included a comparison of some English language reports with the originals for French and Spanish companies. No differences in figures or disclosures were detected.

12 IAS 12, para. 81(c)(i).

13 In the UK, a part of this variation is due to the fact that many companies have reporting periods unsynchronised with the tax years at a time when the corporate tax rates were reduced.

14 Cintra and Repsol.

15 Most IFRS reporters also refer to a ‘statutory rate’ for the reconciliation of tax expense to pre-tax profit.

16 Raedy et al. (2011), who count of the number of items in the reconciliation of US Fortune 250 firms, find a mean (and median) of four, which is substantially lower than the overall mean of 6.5 observed in our sample.

17 Raedy et al. (2011), when counting the number of temporary differences reported in their sample of US Fortune 250 companies, find a mean of 10 items, which is the same as the overall mean in our sample.