Non-GAAP disclosing companies are typically characterized by greater incidence and magnitude of profit and loss items related to fair value asset remeasurement and impairment. These companies are also more likely to have their financial statements adjusted by the analyst. Analysts’ predictions that include such adjustments are more accurate.
What’s Inside?
The authors focus on the implications of non-GAAP earnings presentations and
adjustments in financial reports prepared in accordance with the International
Financial Reporting Standards (IFRS). Apparently, there are various measures of
non-GAAP earnings, which are designed to make financial statements more informative.
The widespread use of non-GAAP earnings indicates that the practical application of
the IFRS—namely, the fair value principle—is not an obvious fact. There
are a number of entries whose presence may distort the true image of a company.
These include one-offs, unusual or nonrecurring items, volatility associated with
events beyond the managerial control, and the effects of accounting treatments.
The non-GAAP earnings, according to management’s belief, remove such items to
disclose the true, underlying performance of a company. The authors investigate the
extent to which the release of non-GAAP earnings metrics is associated with fair
value measurement under the IFRS. Another area of interest is the relationship
between the adjustments made by management and those made by analysts as well as to
what extent the adjustments are actually useful to the analysts.
How Is This Research Useful to Practitioners?
The authors touch on one of the fundamental questions in the area of accounting and
reporting. As the results of their research show, companies providing non-GAAP
earnings are more likely to have a greater incidence and magnitude of IFRS
remeasurements included in their financial statements. However, analysts are active
in making similar adjustments. The companies offer them a helping hand by providing
their own estimates of these adjustments. Obviously, non-GAAP adjustments can also
be used by managers as a tool for reaching their personal goals. This misuse may
happen when the results calculated in accordance with the IFRS are less than the
previously set targets. But the results of the authors’ research indicate a
relatively high value of the adjustments, proven by the lower forecast error and
dispersion made by the analysts.
The most important question concerns the underlying rationale for making such
adjustments. It appears that both managers and analysts do not see IFRS
remeasurement items as part of the underlying earnings. Although the scope of the
research is limited to Australian companies, the results should be carefully
analyzed by any standard-setting bodies.
How Did the Authors Conduct This Research?
The authors study large Australian companies from the ASX 200 Index. They collect
data about non-GAAP earnings from companies’ annual reports, earnings
announcements, and investor presentations. The authors follow the Australian
Securities & Investments Commission’s definition of non-IFRS financial
information (i.e., any information that is presented other than in accordance with
financial standards). The authors search for non-GAAP earnings using the Adobe
Acrobat Pro text search. The total sample includes 576 firm-years from 2008 to 2010.
Out of the sample, 371 firm-years contain non-GAAP earnings information. The profit
or loss information is sourced from the Aspect FinAnalysis database. The
analysts’ adjustments are sourced from the list of adjustments compiled by
Aspect Huntley analysts. Share prices are obtained from the share price and price
relative database from the Securities Industry Research Centre of Asia-Pacific.
Other financial data are sourced from Aspect Huntley’s database.
The authors apply quantitative methods for the research. In particular, to test the
relationship between IFRS remeasurements and non-GAAP earnings releases, they apply
a binary logistic regression. The forecast error and dispersion are measured by
using multiple linear regressions.
An obvious limitation of the research is its scope, which is limited to only large
Australian companies, and the observations include only a limited time period of
three years. Hence, it would be necessary to retest the models as well as the
validity of the hypotheses in a broader context. Also, there is no information about
how the companies included in the sample may represent the whole population of
companies.
Abstractor’s Viewpoint
The authors raise a fundamental issue from the practice of financial analysis. The
fair value remeasurements and the way the financial market participants understand
and include them in their analyses has been a regular discussion point since the
recent global financial crisis. My personal assessment of the paper is high, and I
look forward to further research in the area, particularly with relation to the
European financial markets.