A research report submitted
In partial fulfilment of the degree Master of Commerce (Accounting)
University of the Witwatersrand / This research report evaluates the nature of, and gathers evidence of, the potential
misuse of the non-GAAP 'adjusted earnings' by JSE-listed firms in South Africa. The prior
literature is explored and applied to the South African context which is a unique
environment due to the mandatory use of the non-GAAP Headline Earnings . The prior
literature provides the grounding for the research methods which enhance the validity of
the study.
Adjusted earnings are analysed through 3 research questions and sub-questions. The
first research question focuses on the nature of the use of adjusted earnings in South
Africa, by examining the extent of use of adjusted earnings by a population of JSE firms,
as well as the most common types of adjustments used. It is evaluated using descriptive
statistical methods from data from databases and company annual financial reports.
Research question 2 gathers evidence for misuse through the identification of 'valid' and 'invalid' adjustments
made in the determination of adjusted earnings, as well as the
identification of the repeated use of particular adjustments, which are indicators of misuse
from the prior research of Bhattacharyaa, Black, Christensenb and Larsonc (2003) and
Doyle, Lundholm and Soliman (2003). This question uses an ANOVA and repeated
measure approach respectively using the same data from research question 1. The third
research question examines whether there is an association between adjusted earnings
and whether firms meet or beat analyst earnings forecasts more often (the dependent
variable) as set out in Doyle, Jennings and Soliman (2013). This is assessed using logistic
regression analysis using analyst earnings forecast data and company results data
The results indicate that types of firms and adjustments made in South Africa are similar
to U.S. literature. It raises questions around use of adjusted earnings as a performance
metric and the use of Headline Earnings in South Africa. Evidence of misuse of adjusted
earnings was found. In addition, a strong relationship similar to the Doyle et al. (2013)
findings was found between the use of upwardly adjusted earnings and the propensity of
firms to meet or beat analyst forecasts. Whether a firm s accounting earnings met or beat
the forecast was also found to have significant influence on the dependent variable. It
was also found that South African firms met or beat analyst forecasts significantly less
often than U.S. firms, suggesting that there may be structural differences in the analyst
forecasts environment in South Africa when compared to the U.S. The results suggest
that adjusted earnings may be misused in South Africa, and one of the motivations to do
so is to meet or beat analyst earnings forecasts. / MT2017
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/22365 |
| Date | January 2016 |
| Creators | Howard, Michael |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | Onliine resource (92 leaves), application/pdf |
Page generated in 0.0113 seconds