Market Reaction to Goodwill Impairments

Abstract

This paper examines the information content of goodwill write-downs under International Accounting Standard (IAS) 36 (Impairment of Assets) and Statement of Financial Accounting Standards (SFAS) No. 142. We investigate whether the informational value depends on the reliability of the news. Using a sample of 564 goodwill write-down announcements issued from 2005 to 2009, we find a negative capital market reaction to announcements of unexpected goodwill write-offs. Our results indicate that investors react more negatively when a country's level of legal protection is low and allows more management discretion. We further report that market reaction is associated with managers explaining the write-down decision and depends on the verifiability of these explanations. Investors react more negatively when an unverifiable internal explanation is given and less negatively when a verifiable external explanation is provided. We do not find significant differences between write-down announcements under SFAS 142 and IAS 36.

Acknowledgements

We gratefully acknowledge helpful comments and suggestions by Florin Vasvari, the associate editor, and two anonymous reviewers. This paper has benefited from comments and suggestions at the 2012 CAAA Annual Conference, the 2012 International Accounting and Auditing Conference, the 2012 AAA Annual Meeting, and the 2013 EAA Annual Congress and by workshop participants at the University of Muenster. All remaining errors are ours.

Notes

1 Throughout the paper, we refer to partial and complete downward goodwill revaluations with the term write-down.

2 An important caveat of this study, therefore, is that a goodwill impairment announcement must have been available. If a firm did not release such information or if the press release was not included in the database or identified during our search, it is not included in our sample.

3 Our study focuses on unexpected write-offs where the announced impairment amount exceeds the write-off expected by investors. Our study contains no predictions for situations where the announced impairment amount is below the expected amount.

4 We thank an anonymous reviewer for this line of inquiry.

5 Prior to the introduction of SFAS 142, US GAAP also allowed firms that met certain criteria to account for business combinations by using the pooling of interest method. Under this method, firms did not have to amortize purchased goodwill, so that the income statement was not affected in subsequent years. Since the pooling of interest method is of little relevance to our study, we refer to Ramanna (2008) for an extensive discussion of the background of SFAS 142.

6 Francis et al. (1996) provide the only evidence that goodwill write-downs do not cause a significant market reaction. Yet, the insignificant market reaction can be attributed to the fact that these authors analyze goodwill impairments during 1989–1992, when regulatory guidance for goodwill impairments was absent. We thank an anonymous reviewer for suggesting this explanation.

7 Mercer (2004) synthesizes the literature on management disclosures and identifies factors that impact reliability. The author concludes that disclosure credibility is affected by the disclosure characteristics themselves, including the supporting information. The underlying rationale is that supplementary statements inform about the likelihood of disclosures being subject to opportunistic actions.

8 We assume that all explanations are classifiable. That is a firm either provides an external reason, an internal reason, an external and internal reason or no reason for the impairment.

9 Values for non-euro currency firms are converted into euros.

10 We excluded smaller goodwill write-downs for two reasons. First, we wanted to ensure that our LexisNexis search referred to write-downs likely to be included in the database. Second, this procedure also enhances the power of our tests. We acknowledge that this procedure also biases in favor of rejecting the null. However, we believe that we still follow a conservative approach based on widely observable firm behavior. For instance, Bens et al. (2011) set a lower limit of 5% of total assets as the minimum goodwill write-down.

11 The LexisNexis full text information database is often used in accounting research (e.g. Beatty & Weber, 2003; Vafeas, 2009). It contains news and business articles from more than 4000 international publications and is updated on a daily basis.

12 Computing the expected returns with Fama–French factors and momentum was not feasible due to the lack of (daily) data availability.

13 We thank an anonymous reviewer for pointing this out. A further chi-squared test provides evidence that managers in common-law countries provide an explanation more often than do managers in civil-law countries (p = 0.005).

14 We computed both the Pearson and Spearman correlations between the independent variables of the regression analysis. Since some of our classification variables are correlated (e.g. CIVILLAW and USGAAP), we test our hypotheses using separate regression models, as explained in the previous section (Bens et al., 2011).

15 The World Bank also uses the extent of disclosure as another measure of investor protection. We do not include this alternative measure because, in our study, a higher level of disclosure does not imply less management discretion when computing the goodwill write-off.

16 We acknowledge that IFRS are applied in many other countries around the world. In our study, we focus on their application in Europe, since they have a long tradition as accounting standards in the European Union.

Additional information

Funding

This work was supported by the German Research Foundation (Deutsche Forschungsgemeinschaft) under Grant KN 1041/1-1.