Market valuations of bargain purchase gains: are these true gains under IFRS?

Abstract

This study investigates stock market valuations for bargain purchase gains (BPGs) in the context of International Financial Reporting Standards (IFRS) between 2005 and 2014. Motivated by the increased frequency and high concentration of BPGs in Europe, we study a sample of acquirers listed on the London Stock Exchange to assess the value relevance of BPGs (a) under discrepant disclosure practices (i.e. disclosure versus non- disclosure of the reasons for the gains), (b) before and after the revision of IFRS 3, and (c) considering different income classifications for BPGs (operating or non-operating earnings). BPGs, on average, are not significantly valued by the stock market. However, the post-IFRS 3 revision period, marked by stricter measurement criteria and additional disclosure requirements, witnessed a significant shift in firm valuations. BPGs for which the reason for the gain is disclosed are positively valued only in the post-IFRS 3 revision period. BPGs are consistently perceived as value irrelevant for those firms which fail to comply with mandated IFRS 3 disclosure requirements regarding the reason for the gain. Finally, BPGs classified as a component of non-operating income with sufficient note disclosure on the reason for the gain are significantly associated with prices and returns.

Keywords:

• Bargain purchase gains
• negative goodwill
• IFRS 3 revision
• accounting disclosure
• value relevance

Acknowledgement

We would like to thank the editor and two anonymous reviewers for their constructive feedback and directions. We also thank Adam Aoun, John Zimmermann and Joshua Riand for research assistance in the data collection. We are grateful to participants of the EIASM-Accounting and Economics workshop in Segovia, 2012 and the NWDTC conference at Lancaster University in 2013. We are thankful to Paul André for his constructive comments. We also benefited from the comments of Stergios Leventis, John O’Hanlon and Steven Young on an earlier draft of the paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Value relevance reflects the ability of an accounting measure to capture and summarise information that affects equity value (Wyatt 2008). It is characterised as one of the most important attributes of accounting quality (Francis et al. 2004). Barth et al. (2001) emphasise on the importance of the value relevance literature in providing fruitful insights for standard setting. However, Holthausen and Watts (2001) argue that, in the absence of descriptive theories to justify empirical associations, the value relevance literature has limited implications for standard setting.

2 According to the European Securities and Markets Authority (ESMA 2014), out of 56 European listed companies with 66 business combinations in 2012, BPGs were reported in 11% of all transactions.

3 Comiskey et al. (2010) uses a sample of 43 firms that recognised negative goodwill over the period 2000–2007. Dunn et al. (2015) employs a sample of 79 acquisitions of failed banks over the period 2009–2010.

4 BPGs are only associated with purchase accounting (Wines et al. 2007).

5 Under Financial Reporting Standard (FRS) 10, acquiring firms were required to separately disclose negative goodwill on the balance sheet under the heading of goodwill and to recognise it in the profit–loss account during the periods in which the non-monetary assets acquired were depreciated or sold. Any negative goodwill in excess of the non-monetary assets was written back in the profit–loss account over the period expected to benefit from the negative goodwill (FRS 10 1997).

6 International Accounting Standard (IAS) 22 stipulated that the amount of negative goodwill arising from bargain purchases should not exceed the aggregate fair values of the identifiable non-monetary assets acquired. This negative goodwill was allocated on a systematic basis over the remaining weighted average useful life of the identifiable depreciable assets. Any remaining excess was immediately recognised as income (IAS 22 1998).

7 However, the prohibition of the recognition of ‘possible obligations’ for contingent liabilities might lead to the recognition of higher BPGs (IFRS 3R 2008; BC 275). On the other hand, in line with this prohibition, the IASB concluded that contingent assets should not be recognised. This might lead to

the recognition of lower BPGs.

8 For instance, the IASB states that the lacking or delayed recognition of contingent consideration under IFRS 3 is unacceptable and unreliable and leads to incomplete financial reporting, thus lessening its usefulness in making economic decisions (IFRS 3R 2008; BC 346–347 349). In contrast, IFRS 3R requires the measurement and recognition of all liabilities at their fair values on the acquisition date (IFRS 3R 2008; BC 379). This approach is expected to result in more accurate and, more verifiable measurement of the different components of the business combination.

9 In a similar vein, the U.S. SEC has questioned the classification of $66 million of BPG arising from the acquisition of General Motors–Strasbourg by General Motors as part of the non-operating income. The company responded that, given the unique nature of a BPG, the company does not believe recording BPGs as operating income to be representationally faithful; hence, these gains should be classified under other non-operating income (SEC Comments Letter to General Motors 2011; SEC File No. 001-34960, Comment 7, 2011).

10 Other benefits, for example, are related to tax advantages such as tax-loss carry-forwards. This refers to accumulated losses incurred from the acquiree that the acquirer may use to claim for a tax expense reduction post-acquisition date. These benefits are normally carried forward for up five years depending on the jurisdiction. Acquisition of tax-loss carry-forwards by profitable companies may reduce the present and future tax liability of the combined firm, resulting in significant tax savings for the acquirer.

11 The BPGs are disclosed pre-tax in the annual reports. We use the firm- and year-specific effective tax rates to calculate the after-tax BPGs.

12 Although an event-study approach would be suitable to assess how the market evaluates a business combination, there are several reasons for not adopting this approach in our study. First, the identification and exclusion of observations related to confounding events would dramatically decrease our relatively limited sample of BPG-reporting firms. Second, after checking several announcements related to our sample acquisitions, we noticed that press releases lack any information related to the existence of BPGs. More importantly, there is no disclosure about the reasons for and, thus, the source of the BPGs.

13 Since we are unable to identify the exact announcement date for 82 firm-year observations, our returns analysis is based on 128 observations.

14 In untabulated sensitivity tests, we perform mean comparison (t-tests) for other common firm characteristics used in the previous literature by using all sub-samples. We find no significant differences in terms of indebtedness, growth opportunities, auditor type, alternative investment listing status, and ownership structure. In addition, we also find no significant differences in the acquisition activity (proxied by the change in positive goodwill) of BPG-reporting firms across the different sub-samples examined.

15 We find that BPG_RET is significantly (10%) more material in the crisis than in the recovery period. This finding suggests that the increased value relevance of BPG_RET is not related to its potentially higher materiality in the recovery period.

16 The industry fixed effects controls employed in the main analysis are based on the Datastream ICB Industry classification level 2, which identifies ten industries (details are provided in Table 2). The BANKS control variable employed in this analysis is based on the ICB Supersector classification level 3 and identifies banks and financial services firms only.

17 In this subsample, BPG_DISC (BPG_NONDISC) has a mean value of 0.027 (0.038, t-test for the difference = −1.579).

Additional information

Funding

We acknowledge financial support for this project from Newcastle University Business School.