Press Release Disclosure of ‘Pro Forma’ Earnings Metrics by Large German Corporations – Empirical Evidence and Regulatory Recommendations

Abstract

This paper is set against the background of recent regulatory action and standard-setting activities pertaining to the disclosure of so-called ‘pro forma’ earnings. For a sample of large corporations listed on the Frankfurt stock exchange, I individually analyze quarterly earnings announcements published for the fiscal years 2005 and 2006. Given voiced concern about the potential of pro forma metrics to misrepresent firm performance and thus to potentially mislead investors, research questions pertain to the use, the calculation and the presentation of pro forma earnings, and the impact of recent recommendations issued by European securities commissions. The results indicate that firms make extensive use of so-called ‘EB’ (earnings before) metrics and, more importantly, of pure non-GAAP performance measures, both in terms of frequency and reporting emphasis. The transparency of adjustments to GAAP earnings turns out to be low. Year-to-year analysis suggests that recent recommendations by European securities regulators have had no discernible impact on these disclosure patterns. Taken together, this evidence suggests that regulatory concern may be warranted. At the same time, it points out the need for more research into the determinants and the investor reception of pro forma earnings disclosures in EU jurisdictions.

Acknowledgement

I thank Ulfert Gronewold, Christoph Kuhner, Ralf Sabiwalsky, Thorsten Sellhorn and an anonymous referee for their valuable comments on earlier versions of this paper. Also, I gratefully acknowledge support by the European Commission Research Training Network INTACCT.

Notes

Management latitude in reporting subtotals can be expected to persist. In their ‘Preliminary Views on Financial Statement Presentation’, a discussion document issued by the IASB and the FASB in October 2008, the standard-setters propose inclusion of subtotals and headings in the financial statements ‘if such presentation is helpful to understanding [the entity's] financial position and changes in its financial position and if those subtotals and headings are presented consistently in the three statements’ (para. 2.23).

For instance, IAS 1.81 requires the EBIT components tax expense and financing items to be shown on the face of the income statement, while EBITDA information on depreciation and amortization amounts can be gleaned from the notes, either directly if the nature of expense method is used, or from nature of expense data which must additionally be reported if firms opt for the cost of sales approach (IAS 1.93).

An ever growing body of empirical studies addresses the effects of country-specific variations in institutions and security regulations. In this literature, Germany is consistently categorized as a country with low enforcement quality (e.g. Hope, 2003, p. 253; LaPorta et al., 2006, p. 16). Recent studies demonstrate that enforcement quality is positively associated with economic effects (liquidity, cost of capital) of (mandatory) financial reporting, particularly under IFRS (e.g. Daske et al., 2008, p. 1120).

A large part of this literature pertains to so-called ‘street earnings’ – adjusted GAAP earnings provided by analyst tracking services such as Zack's oder I/B/E/S. Since such metrics reflect analysts' expectations rather than adjustments conducted by the reporting firms, they are not systematically surveyed here.

In contrast to these studies, Andersson and Hellman (2007) find experimental evidence that, for a setting of more pronounced GAAP–non-GAAP differences, professional investors, proxied by financial analysts, may also be adversely influenced by pro forma earnings disclosures.

Ernst et al. (2005, pp. 21, 58) provide survey-based evidence that German equity investors in general assign higher importance to newspapers and television broadcasts than to annual reports as sources of information. Notably, subsample analysis shows that less experienced, small investors in their investment decisions assign significantly higher importance to print and television media than more experienced investors, and comparatively less importance to annual reports and financial statement information. Also, empirical evidence suggests that analyst tracking services rely on earnings announcements rather than on the related financial statements (Brown and Sivakumar, 2003, p. 567).

The relative importance of the annual earnings number is for example reflected by the degree of analyst coverage. To illustrate this, the IBES database on analyst forecasts has been analyzed for the reporting period under investigation. I find that for the firms in the sample, on average, only 1.84 analyst forecasts are available for quarterly earnings, whereas the consensus forecast for annual earnings on average reflects the predictions of 17.82 analysts.

Metrics such as ‘operating income’, that correspond to an EB metric which is reported as a subtotal in the income statement, are also counted as EB metrics. Particularly, many firms use the terms operating income and EBIT interchangeably under IFRS.

Hitz and Jenniges (2008) investigate the voluntary section of the 2006 annual reports for the 110 largest listed German firms, excluding financial and cross-listed firms. For their final sample of 86 firms, they find that 72 firms (84%) disclose either an EB or a non-GAAP pro forma earnings metric. Specifically, 18 firms (21%) report at least one non-GAAP metric. These findings, though anecdotal in nature, suggest a slightly lower proliferation of pro forma earnings measures in annual reports compared to the sample of quarterly earnings announcements investigated here, and may be attributed, for example, to the auditor review of the annual report.

Also, EB metrics are more predictable and determined in their difference to GAAP earnings. Since they typically exclude expense items, they can be expected to be higher in amount than GAAP earnings. On rare occasions, this may not be the case, for example, for EBT when tax benefits are consumed, or for EBIT when interest income exceeds interest expenses, for instance, due to low leverage and substantial investments in interest-bearing securities.

Untabulated analysis of absolute reporting emphasis, based on a score which recognizes the prominence of the reporting location, the use of performance metrics for earnings forecasts and the display of financial panel data, ranging from zero to five, yields scores of 2.45, 2.71 and 2.58 for GAAP, EB and non-GAAP metrics, respectively.