SA Journal of Accounting Research formerly De Ratione
is an accredited, national academic journal which is published annually by the Independent Regulatory Board
for Auditors (IRBA), the South African Institute of Chartered Accountants (SAICA) and the Southern African Accounting Association (SAAA).
Chairman Colin McClelland - Petroleum industry consultant
Editor Enrico Uliana - University of Cape Town
Editorial Board Charl Kocks – Ratings Afrika
Minga Negash – Metropolitan State College of Denver and
University of the Witwatersrand
Wiseman Nkuhlu – Director of companies
Jeffrey Rowlands - Nelson Mandela Metropolitan University and
Southern African Accounting Association
Lesley Stainbank - University of KwaZulu-Natal
Graham Terry – The South African Institute of Chartered Accountants
Sandy van Esch – Independent Regulatory Board for Auditors
Marianne van Staden – University of South Africa and
Southern African Accounting Association
Annual subscription - R70.00 including VAT.
Payment must accompany all orders
Enquiries and orders:
SA Journal of Accounting Research,
PO Box 59873,
KENGRAY 2100
Tel: (011) 621 6600
Fax: (011) 622 5543
e-mail: [email protected]
Website: www.sajar.co.za
ISSN: 1029-1954 Copyright
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any
form or by any means, including electronic, mechanical, photographic, magnetic or other means, without the prior
written permission of the publishers.
The object of SAJAR is, however, to promote the wide dissemination of the results of research and other scholarly
inquiries into the broad field of accounting. Permission is, therefore, hereby granted to reproduce any of the contents of
SAJAR for use in courses of instruction provided the source and SAJAR’s copyright are indicated in such reproductions.
AD HOC REVIEWERS J Abor, University of Ghana
S Agrawal, University of Memphis
NSM Ahmed, Academy of Graduate Studies, Tripoli
U Anderson, University of Texas at Austin
E Arnold, al-Faisal University
K Barac, University of South Africa
G Barr, University of Cape Town
N Biekpe, Stellenbosch University
D Bradfield, Cadiz Holdings Limited
G Bradfield, University of Cape Town
W Bruwer, Stellenbosch University
C Cairney, University of Cape Town
E Chamisa, University of Cape Town
R Chivaka, University of Cape Town
AJ Cilliers, University of Brighton
D Coetsee, University of Johannesburg
S Coetzee, University of Pretoria
T Cohen, Berk Enterprises
C Correia, University of Cape Town
P Court, Rhodes University
P Cramer, University of Cape Town
S Davidson, RMIT University
C Deegan, RMIT University
A de Graaf, University of South Africa
H de Jager, University of Pretoria
P de Jager, University of Cape Town
C de Villiers, University of Auckland
A du Toit, Monash South Africa
L Esterhuyse, University of South Africa
C Firer, University of Cape Town
T Flegg, University of the West of England
D Forsyth, Nelson Mandela Metropolitan University
J Fouche, North-West University
E Gilbert, Cadiz Asset Management
S Goodman, University of Cape Town
D Gouws, University of Pretoria
M Graham, University of Cape Town
J Griffioen, University of Johannesburg
WD Hamman
L Hanner, University of Fort Hare
G Holman, University of Cape Town
J Kew, University of Cape Town
BS Khanna, Victoria University of Wellington
D Kolitz, University of Exeter
M Kolitz, University of the Witwatersrand
C Koornhof, University of Pretoria
H Kriek, University of Johannesburg
J Lancaster, Rhodes University
T Lodewyckx
W Lotter, Cape Peninsula University of Technology
W Maguire, University of Tasmania
M Mangena, Bradford University
B Marx, University of Johannesburg
J Miller
T Minter, University of Cape Town
G Modack, University of Cape Town
P Myers, Rhodes University
M Nieuwoudt, University of South Africa
L Olivier, University of Johannesburg
M Page, Bentley University
M Paxton, University of Cape Town
J Prather-Kinsey, University of Alabama at
Birmingham
A Prencipe, Bocconi University
F Prinsloo, Nelson Mandela Metropolitan University
J Pym, University of Cape Town
E Rabin, University of the Witwatersrand
G Radder, Nelson Mandela Metropolitan University
SE Radloff, Rhodes University
A Riahi-Belkaoui, University of Illinois at Chicago
J Roeleveld, University of Cape Town
G Samkin, University of Waikato
K Sartorius, University of the Witwatersrand
A Savage, California Polytechnic State University
A Saville, Canon Assets
H Scholtz, Stellenbosch University
J Shev A Singleton, Nelson Mandela Metropolitan University
C Smith, University of Cape Town
L Stack, Rhodes University
L Steenkamp, Stellenbosch University
N Stegmann, University of Johannesburg
A van der Watt, University of Johannesburg
M van Heerden, University of Johannesburg
P van Rensburg, University of Cape Town
L van Schalkwyk, Stellenbosch University
C van Staden, University of Canterbury
W van Zyl, Ernst & Young
S Visser, North-West University
T Voogt, University of Johannesburg
M Ward, University of Pretoria, GIBS
D Warneke, University of Cape Town
A Watson, University of Cape Town
N Waweru, York University
S Weil, Lincoln University
C West, University of Cape Town
J Winfield, University of Cape Town
N Wood, University of KwaZulu-Natal
M Wormald, University of Cape Town
Volume 25 Number 1 2011
Accounting academics’ multiple challenges: Issues-driven learning offers a way forward JH Hesketh
1-34
Students’ and lecturers’ perceptions of the effect of open-book examinations on the learning behavior of accountancy students SJ Kruger
35-57
Portfolio rebalancing in South Africa G Sher and GDI Barr
59-80
A comment on research frameworks applied in accounting research D Coetsee
81-102
Sustainability reporting at large public sector entities in South Africa B Marx and V van Dyk
103-127
The information content of the Financial Mail’s “Top Companies” announcements WD Esterhuysen and M Ward
129-143
Notes and Commentaries
The SAICA Part I Qualifying Examinations: Factors that may influence candidates’ success E van Wyk
145-174
The perceptions of a CTA class regarding the changes to the international accounting standards dealing with financial instruments G Steenkamp
175-189
Estimating the relationship between environmental performance and economic performance of South African mining companies TF Prinsloo and M Oberholzer
191-207
The SA Journal of Accounting Research is an accredited, national academic journal which is published annually by the Independent Regulatory Board for Auditors (IRBA), the South African Institute of Chartered Accountants (SAICA) and the Southern African Accounting Association (SAAA).
JH Hesketh 1
Accounting academics’ multiple challenges:
Issues-driven learning offers a way forward
JH Hesketh
Management Studies Education Unit, University of KwaZulu-Natal
Received: November 2010 SAJAR
Revised: June 2011, September 2011 Vol 25 No. 1
Accepted: September 2011 2011
pp.1 to 34
Accounting academics in South Africa are under pressure from their stakeholders, particularly
the South African Institute of Chartered Accountants (SAICA) and the Department of Education
(DoE), to make urgent changes. Challenges include the need for increased research output and
for new teaching, learning and assessment techniques that require and enable students to develop
additional competencies beyond core technical knowledge, thereby improving graduate
attributes and student retention rates. Changes needed involve an educational focus and mind-
shift. The purpose of this article is to establish how issues-driven learning (IDL), as an example
of experiential or experience-based learning theory (ELT) can be implemented to meet the
challenges.
KEY WORDS
accounting education; South African Institute of Chartered Accountants (SAICA); competency
framework; professional and pervasive skills; experiential learning theory; issues-driven
learning; teaching and assessment techniques
Contact
INTRODUCTION
Accounting education in South Africa might well feel under siege faced as it is with
seemingly contradictory stakeholder demands for change. These come from SAICA,
other professional bodies and graduate employees and the DoE.
SAICA, in line with the International Federation of Accountants (IFAC), has issued a
new Competency Framework Detailed Guidance for Academic Programmes1 (SAICA
2010a) which “identifies and describes the professional competencies (knowledge,
skills and attributes)” required by candidates2 entering the profession (SAICA, 2010a).
Assessing the additional competencies in professional examinations will involve new
assessment approaches and these will influence the way Accounting Schools teach and
1Hereafter referred to as the competency framework.
2Candidates striving to be admitted into the profession through professional examinations will hereafter
be referred to as candidates.
2 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
assess their students in order to prepare them incrementally for professional
examinations. The required competencies reflect closely the desired graduate attributes
put out worldwide in university missions, by other professional bodies (Irby, 2000;
Johnstone and Vignaendra, 2003; Kirkland, 2000), by prospective employers (Griesel
and Parker, 2009), confirming the relevance of SAICA’s (2010a) competency
framework also for educators of professional practitioners more broadly in
accountancy-related fields. The DoE (1997; 1998) targets very similar competencies for
development of required graduate attributes. Hence agreement amongst stakeholders on
what has been lacking and what is needed in students strengthens SAICA’s (2010a) call
for change. At the same time the DoE (2004; 2007) also requires academics to increase
their research outputs, improve student retention rates and ensure suitable scholarly
activity in honours programmes in order to meet funding-related criteria.
The scholarly, professional and student retention requirements involve a variety of
seemingly conflicting challenges. The honours research requirement might appear to
conflict with the need to develop prospective accountants and to address low student
retention rates. However, research requires students to gather, examine and interpret
information and ideas critically before using them to build logical arguments and these
academic capacities mirror the professional skills required by SAICA (2010a:26) and
prospective employers (Clinebell and Clinebell, 2008; Griesel and Parker, 2009). Hence
by considering the requirements through an educational lens coherence amongst the
requirements is revealed. An IDL lens illustrates consistency also with the student
retention requirement since research-related activities are a central element of IDL-
based curricula3, designed to empower students as effective learners, and thereby
improve student retention rates significantly (Hesketh, 2003).
While coherence offers a more containable target than disparate requirements, it does
not in itself solve accounting academics’ problems. The purpose of this article is to
establish how IDL can be applied to meet the challenges posed by SAICA’s
competency framework to develop the required qualities, skills and depth of knowledge
beyond the core technical competencies and second, to improve student retention rates
and increase scholarly activity in students and staff as required by the DoE.
The article hence aims to introduce a helpful trans-disciplinary slant to debate around
demands facing accounting academics that draws on curriculum theory relevant to the
requirements, namely IDL as an example of ELT.
To this end the article first considers literature that establishes the need for teaching,
learning and assessment strategies associated with IDL to address current and imminent
stakeholder demands. Second, it argues that experiential learning theory (ELT) and
specifically IDL as an example of ELT provides a suitable theoretical framework for
educational practices that meet stakeholder requirements, and third, the article
demonstrates how IDL can be applied in practice to resolve the challenges posed by
SAICA and the DoE.
3The term ‘curriculum’ refers in this article to the whole teaching, learning and assessment experience,
as it is used in education disciplines.
JH Hesketh 3
STAKEHOLDER REQUIREMENTS
This section draws on literature to demonstrate the additional competencies required of
students and hence the kinds of changes in accounting education urged by first SAICA,
and second the DoE.
Professional and employer requirements
The pervasive qualities and skills required of entry level South African chartered
accountants CA(SAs) as set out in the competency framework (SAICA, 2010a) reflect
the values and requirements of IFAC (2010) and are closely aligned with the graduate
attributes that other professional bodies and employers require internationally. Hence
SAICA’s requirements, and its new-style professional examinations from 2013, can be
contextualised within a widespread recognition of graduate shortcomings and the need
for new kinds of teaching, learning and assessment. It is important for accounting
academics to recognise that they are not alone in their educational challenges and that
SAICA’s (2010a) required competencies reflect wide and current thinking if they are to
commit themselves philosophically as well as technically to applying new teaching,
learning and assessment approaches. SAICA’s new requirements are considered first,
followed by employer requirements.
SAICA’s new requirements
Like other professional bodies SAICA has reflected iteratively on its qualification
model and revised its assessment policies and practices continuously. Many crucial
elements of the competency framework (SAICA, 2010a) can in fact be found in earlier
syllabuses. SAICA’s syllabus of 2005, for instance, was informed similarly by the
recognition that “characteristics essential to a profession” needed to be developed and
that professionals also need “to possess intellectual, analytical and advisory skills that
enable them to apply core knowledge” (SAICA, 2005:2), listing as desired outcomes
capabilities like critical thinking, effective communication, identifying, selecting and
integrating appropriate information, contextualising knowledge. The document also
specifically links such capabilities to the desired outcome of lifelong learning,
demonstrating strong similarities to the new competency framework (SAICA, 2010a).
However examiners’ comments (SAICA, 2009) illustrate some of the difficulties
CA(SA) candidates have had in developing the kind of skills that have been found
lacking, despite their inclusion in syllabuses. They include:
Responses to [requirements for recommendations / interpretation] are generally
poor, either because candidates are unable to explain principles that they can apply
numerically or because they are reluctant to commit themselves to one course of
action. It is essential to make a recommendation when a question calls for it, and
to support it with reasons. Not only the direction of the recommendation (i.e. to do
or not to do something) is important, but particularly the quality of the arguments
– in other words, whether they are relevant to the actual case and whether the final
recommendation is consistent with those arguments. Unnecessary time is wasted
by stating all the alternatives.
4 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
A serious problem experienced throughout the examination was that candidates
were unable to apply their knowledge to the scenarios described in the questions
...
Many candidates did not address what was required by the questions and, for
example, provided answers in the form of statements while calculations were
required or presented financial statements where a discussion of the appropriate
disclosure was required.
(SAICA, 2009).
The new competency framework project, initiated in 2007 in response to the fact that
candidates were not demonstrating adequately the required skills and qualities
appearing in their syllabuses (SAICA, 2009), has four key changes that demonstrate
new thinking.
First, ‘leadership’ has been specifically identified as a fundamental attribute of a
CA(SA) who “should have the full range of technical competencies of a professional
accountant and also those which will enable the development of leadership qualities”.
The competency framework hence “focuses on those pre-qualification competencies
which provide a foundation for the acquisition of leadership ability after entry to the
profession” with the objective of raising “technical competencies to a level applicable at
the strategic level” and ensuring that “technical disciplines be taught and assessed with
this objective in mind” (SAICA, 2010a:4-5).
Second, while the 2005 syllabus shows that “pervasive qualities and skills” set out as
requirements in the new competency framework (SAICA, 2010a: 19-30) are not by any
means new to SAICA’s thinking, their presentation and conceptualisation within a
competency framework rather than a “knowledge-based syllabus” ( SAICA, 2010a:3;
IFAC, 2010:19) represents a significant shift in focus in terms of teaching, learning and
assessment. The pervasive qualities and skills listed under “ethical behaviour and
professionalism ... personal attributes ... (and) professional skills” (SAICA, 2010a:19-
30) are to combine with specific competencies “to produce the technical excellence,
integrity, objectivity, and commitment to public interest for which the CA profession is
known (SAICA, 2010a:19). This conceptualisation of what is required in terms of
detailed competencies (SAICA, 2010a:25-30) calls for a heightened level of
accountability in educators and learners to begin preparing for professional
examinations from the start of undergraduate studies. It is explicitly stated that “the
knowledge base provides a foundation for the development of the competencies”
indicating that knowledge will be assessed in terms of the kind of understanding that
prohibits rote learning (SAICA, 2010a:11) and competencies will be demonstrated
“within (the) context of tasks” that depend on sound technical knowledge (Olivier and
Kleinhans, 2008:11).
The swing from knowledge-based to competency-based assessment reflects the shift
from assessing “whether a candidate has obtained the professional knowledge required
to perform the required tasks as a professional accountant” to assessing competence in
candidates with capacities to draw on competencies to perform realistic tasks “to a
defined standard, with reference to real working environments” (IFAC, 2010:19). The
swing also reflects IFAC’s requirement that its member bodies develop teaching and
assessment methods that are effective in developing and testing required competencies.
JH Hesketh 5
Third, while the new competency framework might not look significantly different from
previous syllabuses, the critical difference will become apparent when new-style
specimen questions and examinations are made public as guidelines for new approaches
to teaching, learning and assessment. It is not yet possible to provide a detailed analysis
of the changes since the development of specimens is currently work-in-progress but
from 2013 (for Part 1) the assessment questions and the shape of the examinations will
reflect the guidelines in terms of what is to be assessed and how it will be assessed
(SAICA workgroups, 2011). Beyond demonstrating appropriate application of
knowledge in questions that involve integration across and within competency areas, for
example, candidates will be rewarded for demonstrating specific professional skills. All
competencies are considered assessable in one form or another at some stage in the
education and training programme but the new focus for accounting academics is
clearly on developing and assessing those professional skills that are demonstrable in
examination forums.
A fourth change from past syllabuses is Dewey’s philosophical approach to education
that explicitly underpins the competency framework and which has also informed the
IDL approach. By grounding theory in practice, the Deweyan approach leads to students
learning to “think like business people” by requiring them to demonstrate “both
technical expertise and an understanding of the significance of the solutions arrived at”
and understandings of the “implications of new knowledge in relation to current
contexts” (SAICA, 2010a:8-9). It points to a shift in educational understanding that
recognises the need to move from traditional, teacher-centred, knowledge transmission
approaches and associated knowledge-based assessment (Hesketh, 2003) to approaches
like IDL. A principle-based curriculum, for example, emphasises conceptual principles
which can be applied in the future, as opposed to learning rules or training students
according to past experience.
Principle-based learning, like IDL, enables and requires students to develop a
conceptual basis for analysis; application of theoretical knowledge; exercising of
judgement; and understanding of the kind of information needed to reach a decision, the
implications of the decision and why an approach or calculation is appropriate, rather
than simply how it is conducted (Barth, 2008, 2011; Clinebell and Clinebell, 2008;
Watson, 2010). Hence the focus on principles or conceptual frameworks or theoretical
understandings, results in students being able to develop “their own approaches based
on their understanding of (the) principles” (Watson 2010:3), appropriate responses to
unknown future problems (Clinebell and Clinebell, 2008), and “knowledge that is more
enduring” (Watson, 2010:2).
Hence principle-based curricula are closely aligned with both IDL and SAICA’s
(2010a) requirements and their underpinning Deweyan interest in discovery-learning,
contextualising knowledge, developing the intellectual attitudes and approaches of
lifelong learners (Hesketh, 2003) and combining relevance to business practice with
academic rigour Clinebell and Clinebell (2008). Similarly aligned is the the ‘Part 2’
examination’s new focus on “strategic and managerial aspects of the accounting
disciplines” (SAICA, 2010b) rather than on technical competence alone. This further
highlights SAICA’s (2010a) shift in thinking and the importance of requiring students
to understand conceptual principles.
The four key change areas confirm Olivier and Kleinhans’ (2008) argument that new
teaching and learning approaches are needed but that nothing significant in the content
6 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
requires changing. The significance of the pervasive skills (SAICA, 2010a) lies in their
explicit link to new assessment approaches that will influence teaching, learning and
assessment practices in universities.
SAICA is not alone, however, in finding mismatches between competencies required
and competencies demonstrated in its candidates and it is important for South African
accounting academics to recognise the new demands in the context of those facing
IFAC and other accounting bodies. The Canadian Board of Evaluators, for example,
have reported very similar problems in their candidates (Chartered Accountants of
Canada, 2007). Professional bodies relating to law and medicine are also seeking to
address competencies found lacking in their graduates and academics in these
disciplines have indeed made considerable progress in developing curricula accordingly
along with supportive structural and policy changes (Irby, 2000; Johnstone and
Vignaendra, 2003; Kirkland, 2000).
Graduate attributes required by employers
Employer demands for graduates with attributes similar to professional competencies
identified by SAICA confirm the relevance of SAICA’s requirements for both CA(SA)
candidates and graduates from practitioner programmes ’s in accounting fields.
Mismatches, similar to those above, between graduate attributes achieved and those
sought by commerce, industry and society are evident in countries as varied as the
United States of America, the United Kingdom (UK), Australia and South Africa
(Fenton O’Creevy, Knight and Margolis, 2006; Griesel and Parker, 2009; Hilton 2008;
Louw, 2008; Naudé, 2008; Skinner, 2005; Wessels and Steenkamp, 2009). Griesel and
Parker, (2009:9) draw on Yorke to explain the notion of graduate attributes in terms of
‘employability’ which goes beyond demonstrating key skills. ‘Employability’ refers
also to capacities e.g. to apply a “mix of personal qualities and beliefs, understandings,
skilful practices and the ability to reflect productively on experience… in situations of
complexity and ambiguity”.
Again, as in the case of SAICA’s requirements, ideas on what is needed in our
graduates are not new and have been expressed widely over many years. What is
relatively new is a shift in focus to the notion of employability and to the kind of
teaching, learning and assessment required to develop it in students (D’Andrea,
Gosling, Scott and Tyeku, 2002; Griesel and Parker, 2009; Scott, Yeld and Hendry,
2007). Specifically, Griesel and Parker (2009: 24) argue for IDL-aligned approaches
that can result in “intellectually well-grounded individuals who are flexible and can
readily adapt to new demands and challenges”, linking these competencies to lifelong
learning and capacities to work effectively in rapidly changing environments.
Griesel and Parker (2009) echoed misalignments elsewhere when they found even
reading and writing skills to be lacking, with obvious implications for learning.
Inadequate knowledge and intellectual ability also has been found across disciplines.
The authors noted an inability in graduates to learn meaningfully and little evidence of
lifelong learning with the greatest gap being abilities to “choose appropriate information
to address problems” and to “plan and execute tasks independently”. There were
significant inadequacies in communication, in summarising key issues, in relating their
knowledge to the world of work and in demonstrating skills and understanding in the
JH Hesketh 7
work context. Hence the authors identified problem areas also targeted by SAICA
(2010a) that called for new, IDL-aligned approaches to teaching and learning.
South Africa is not facing this challenge alone hence the problem cannot be blamed
simplistically on factors like student under-preparedness or poor work ethic. American
business school graduates have also been criticised for lacking skills and attributes
required in business (Hilton, 2008), lacking creativity in applying content knowledge
and in capacities to apply critical analytical skills needed to cut to the essence of a
problem (Rubin and Dierdorff, 2008). This concern was widely shared as was the
inability to demonstrate global perspectives, ethical behaviour, systemic thinking and
‘soft’ skills that include team-building, communications and interpersonal skills,
negotiation and leadership, (Atwater, Kannan and Stephens, 2008; Beenen and Pinto,
2009; Birkinshaw, 2009; Navarro, 2008; Shareef, 2007; Zhu, 2009) again in line with
SAICA’s requirements and call for new kinds of learning opportunities.
Authors internationally argue similar points around curriculum shortcomings in terms
of requiring and enabling students to develop the desired graduate attributes, thereby
supporting the competency framework requirements and philosophy (SAICA, 2010a)
and the argument that IDL provides a means of addressing current educational
challenges (Atwater et al., 2008; Beenen and Pinto, 2009; Birkinshaw, 2009; Boud and
Feletti, 1998; Giacalone and Thompson, 2006; Grey, 2004; Hesketh, 2003; Jacobs,
2007; Kolb and Kolb, 2005; Learmonth, 2007; Liang and Wang, 2004; Naudé, 2008;
Oberholzer, 2007; Reynolds, 1999; Reynolds and Vince, 2004; Samuelson, 2006;
Shareef, 2007; Zhu, 2009; Wessels and Steenkamp, 2009). Echoing the arguments
above of Watson (2010) and Barth, 2008, 2011) the authors show particular concern
that students should develop proper understanding of business principles in order to
make good, conscious choices about why an action is taken and whether it suits specific
circumstances.
Graduates who have been required, for instance, to consider why certain accounting
practices are appropriate rather than learning only how to apply them are more likely to
make personal sense of their learning, adjusting their understanding accordingly and
relating it to their values. The above authors also argue that graduates with enquiring
minds and analytical skills are more likely to develop systemic understandings required
in business; awareness both of seemingly harmless mechanisms that allow
organisational corruption to take place and of their own value systems; and to
demonstrate willingness to change a mind-set. These are clearly important attributes in
an environment of accounting failures like Enron in the USA, Shell in the Netherlands
and UK, and Parmalat in Italy.
The thinking captured in this sub-section offers wide support for, and therefore adds
credence and weight to, the new competency framework’s requirements and its call for
approaches to teaching and learning that result in “the creation, analysis, evaluation and
synthesis of information and ideas; problem-solving and decision-making skills;
communication ... vital to the professional success of CAs” (SAICA, 2010a:25) and the
relevance of IDL as a means of addressing graduate shortcomings. The arguments also
underpin the DoE’s policies discussed below adding coherence to the multiple
challenges facing accounting academics.
8 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
DoE requirements
As was the case with SAICA’s development of the competency framework, recent DoE
policies and frameworks (DoE, 2004; 2007) did not so much represent new ideas for
higher education as new strategies to ensure their implementation and this sub-section
argues that IDL provides a means of meeting the DoE requirements.
It is not a new idea for instance that postgraduate students should conduct research, but
it is now a formal requirement; decent pass rates have always been sought and generally
linked to funding but are now linked to funding in new ways; research outputs have
always been expected from academic staff and these too are now more strongly
enforced; good teaching, learning and assessment practices have always been discussed
but there is now more recognition of their direct relationship with student retention and
pass rates if ‘standards’ are to be maintained or raised.
The new policies and frameworks (DoE, 2004; 2007) were intended to assure quality
and consistency and to increase scholarly activity generally and the challenges
associated with their implementation are currently being felt. Accounting schools which
had not yet done so were required to incorporate a research component in Accounting
Honours or the Certificate in the Theory of Accountancy (CTA). In addition, funding
formulae were linked to both the publishing of academic research and to student
throughput rates (DoE 2004), the latter reflecting the educational thinking and goals set
out in the 1997 White Paper (DoE, 1997:13-14). Some of the goals were: “to improve
the quality of teaching and learning throughout the system...” and “to produce graduates
with the skills and competencies that build the foundations for lifelong learning,
including, critical, analytical, problem-solving and communication skills, as well as the
ability to deal with change ...”.
Hence the DoE linked throughput rates to new approaches to teaching and learning and
the educational goals were not new but were conceived in terms of critical outcomes
(DoE, 1998) that demonstrate some important synergies with competencies required by
SAICA (2010a) and other stakeholders and indicate the suitability of IDL as a means of
meeting those learning outcomes. Difficulties in achieving educational goals and critical
outcomes have led to more recent legislation.
Through the Higher Education Qualifications Framework and Higher Education Act
101 of 1997, the DoE (2007:25), has called for bachelor honours degrees to
“consolidate and deepen the student’s expertise” in each discipline and to “develop
research capacity in the methodology and techniques of (the) discipline” and “a high
level of theoretical engagement and intellectual independence” and thereby lifelong
learning approaches. These requirements demonstrate direct synergies with those of the
stakeholders discussed above and similarly indicate the usefulness of IDL in developing
in students the required capacities. The renewed emphasis on research components at
honours level has implications for some accounting schools in terms of course
nomenclature but also in terms of the intellectual capacities required in research that
need to be developed incrementally from the start. Research involves identifying
information that is relevant to arguments and problematising knowledge in relation to
reality. The research process involves professional skills required in the workplace,
specified as: “gathers...analyses...evaluates information and ideas...verifies and validates
information...integrates ideas and information from various sources...draws
JH Hesketh 9
conclusions/forms opinions...” (SAICA, 2010a:25-27) and linked to lifelong learning,
demonstrating again the synergies amongst stakeholder demands and with IDL.
Linking state funding to student retention rates (DoE, 2004) which might on its own
threaten standards, also adds to incentives to academic departments to develop curricula
and teaching approaches like IDL that promotes effective learning, with the kind of
meaning-making that results in full understanding of the subject matter and hence
capacities to apply knowledge and recognise the implications of applying it one way or
another, or choosing one treatment over another in reality. There are thus clear
synergies amongst the DoE, the Accounting profession and employers in terms of the
kind of graduate competencies and attributes sought and the recognition that new ways
of teaching and learning are required.
Synergy can also be found between the need for new teaching approaches with the
DoE’s drive for research, linking funding allocations to academics’ research output
(DoE, 2004). However, most accounting schools worldwide have traditionally focused
more on developing aspirant accountants than on their own research. Hence the
requirement for scholarly activity, which is generally associated with research published
in suitably rated journals, offers serious challenges to accounting academics concerned
with issues of tenure and promotion. Clearly discipline-related research is an ideal in
terms of research feeding back into theoretical understandings and thereby enriching
teaching practices. However, additional research prospects lie in opportunities to
combine teaching and research interests by researching the effective educational
implementation of technical aspects of a course or by developing teaching practices
based on explicit educational theory and reflecting critically on them as practitioner-
researchers (Hesketh, 2003; Jarvis, 1999, 2004), with or without collaboration with
educational specialists. (Recommendations are given below for policy changes in
universities to encourage practitioner-research of this nature).
A Council for Higher Education (CHE) paper (Scott et al., 2007) supports this strategy.
The authors identified the need for developing new kinds of “teaching expertise...based
on systematic knowledge of teaching and learning processes in higher education
acquired through literature, reflection and research” (Scott et al., 2007:61). They found
that academics relied largely on their ‘craft knowledge’ of education, some more
successfully than others, for developing their courses, when the ‘craft knowledge’ was
associated with “excellence in the discipline, and personal charisma” and indeed
‘excellence in teaching’ awards, rather than the kind of ‘teaching expertise’ defined
above (Scott et al., 2007:61). The ‘craft knowledge’ with its common-sense approach to
teaching and assessment has been found inadequate in directing academics towards rich
alternatives for addressing the challenges facing higher education. Theoretical or
systematic educational understandings could have informed, for example, the
development of teaching and assessment practices that specifically required and enabled
students to learn in ways that led to mastery of the knowledge and competencies
required of them. What was needed was academics’ recognition that new kinds of
learning and therefore new kinds of teaching and assessment were needed in order to
meet the required learning outcomes of current stakeholders.
The notion of accounting educators considering and developing their curricula and
specifically their teaching and assessment approaches in light of appropriate educational
theory brings a rich opportunity for collaborative trans-disciplinary work between
discipline specialists and higher education specialists. Given appropriate incentives and
10 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
rewards, accounting academics could thereby gain access to educational theoretical
frameworks like ELT or IDL in which to locate their practices and from which to reflect
critically on them and undertake practitioner-research. Teaching and learning principles
could then be used to guide curriculum development processes, allowing academics to
implement optimal approaches as best they could in less than optimal circumstances
(like large classes, staff shortages and unsuitable teaching venues). At the same time the
reflective processes would allow discipline specialists to use their practices to push the
boundaries of educational theory (Hesketh, 2003) in relation to their contexts. New
interests in curriculum and learning theory would hence expand research opportunities
to incorporate trans-disciplinary collaborative work and help academics to implement
curricula based on educational theories.
Hence approaches like IDL could provide a means of meeting DoE demands, namely
developing research skills in students, giving academics an additional research
opportunity and thereby the possibility of increased research output, and improving
student retention.
ELT AND IDL: A THEORETICAL RESPONSE TO STAKEHOLDER
REQUIREMENTS
This section argues that ELT provides a suitable broad theoretical framework for
addressing the stakeholder requirements discussed above and accommodates a variety
of approaches, including IDL. It is intended that by making explicit the principles
underpinning ELT readers will consider ELT in relation to their own practices and
contexts and make decisions about its suitability to their purposes and the applicability
of IDL as an example of ELT.
Drawing on the work of Barth (2011), Hesketh (2003), Scott et al. (2007), van Esche
(1998) and Watson (2010) it is strongly indicated that ‘traditional’ teaching practices in
accounting schools have been widely based on the assumption that content-driven or
knowledge-based syllabuses make sense whereas they are now to be competency-based
(IFAC, 2010; SAICA, 2010a). At the same time teaching approaches have been
informed generally and implicitly by the theory that teacher-centred education is an
effective means of enabling learning, knowledge is effectively transmitted through
lectures, and concepts and facts ‘covered’ in the lecture can be considered learned and
understood. Rote learning has often resulted from this approach with more emphasis on
the ‘how’ than the ‘why’ and on the whole academics have taken little responsibility for
enabling students to develop the depth of understanding and competencies now
explicitly identified as stakeholder requirements. At the same time the literature of the
previous section has shown that the ‘traditional’ approach has proved unsuccessful in
terms of developing the required understandings, competencies and the kind of attitudes
and approaches to learning associated with lifelong learners.
ELT provides a broad framework for educational theories and approaches that require
students to come to grips with and make personal sense of knowledge, thereby
enhancing knowledge retention and active participation in learning processes. Set tasks
require students to grapple with issues emerging from realistic scenarios or case studies
to reach decisions or form justified opinions, usually in small-group learning contexts
within tutorials of preferably not more than 30 students. ELT aims to enable students
inter alia to: understand the underlying principles of what they are learning and thereby
to develop full understanding; apply knowledge to different situations; identify
JH Hesketh 11
knowledge relevant to a situation; know why a certain treatment or calculation is
appropriate as well as how it is conducted. ELT hence provides a suitable theoretical
framework for curricula that require learners to develop the competencies identified
above.
ELT accommodates approaches and theories like adult, lifelong, student-centred,
discovery, problem-based, case-based and issues-driven learning. Central to the theory
are questions about ‘exactly how people learn what they learn through experience and
from experience’ (Gregory, 2002:94) and how to promote criticality, meaningful
understanding that impacts on the way people see the world, consciousness-raising and
personal development.
Jarvis (2004:104) drawing on Miller and Boud (1996) defines experiential learning as
“the process by which individuals, as whole persons, are consciously aware of a
situation and make sense, or try to make sense of what they perceive, and then seek to
transform it and integrate the outcomes into their own biography”.
In other words experience is the basis and catalyst for learning where learners are
confronted with real or realistic complex scenarios or case studies including, for
instance, a real or simulated stock-count or genuine documentation that can help in
making the transition from student to trainee, and give a degree of ‘experience’ on
which students can draw in making sense of new knowledge. The scenarios present
students with experiences to which they cannot respond automatically but which they
can transform into “knowledge, skills, attitudes, values, emotions, beliefs and senses”
(Jarvis, Holford and Griffin, 1998:46). The situations are always accessible yet
challenging to students and preferably involve both reason and emotion as the two in
combination enhance learning, meaning-making and memory. ELT recognises that
when learning starts with experience students are intrinsically motivated and when tasks
require active reflection on new knowledge in relation to experience, the learning is real
(Dewey, 1916, 1938; Kolb, 1984; Kolb and Kolb, 2005).
Given the stakeholder requirements discussed above, ELT thus offers a highly suitable
theoretical framework for developing teaching, learning and assessment practices that
can address the stakeholder requirements.
IDL: A development from ELT
This sub-section explains why IDL was developed from the broad ELT framework as
an appropriate vehicle for meeting stakeholder requirements. Principles underpinning
the approach are made explicit so that readers can judge for themselves the principles’
applicability to their purpose, practice and context. It also invites readers to consider
whether or how they might implement IDL differently from ways discussed below.
Hence the focus here is on how IDL, as an example of and development from ELT is
suited to meeting stakeholder requirements while the next section demonstrates how
IDL can be applied for this purpose. The process of developing IDL-based practices has
involved and continues to involve ongoing critical reflection on practice in relation to
learning theory and iterative course improvement (Goodier, 2005; Hesketh, 2003).
Hence theory influences practice but at the same time practice informs theory in the
process of iteratively improving an approach to teaching, learning and assessment in
12 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
light of stakeholder requirements. In this way the IDL concept has been developed
within the framework of ELT in response to the need to find a more critical framework
for implementing ELT principles than is generally demonstrated in the literature, most
often round problem-based learning (Hesketh, 2003).
IDL is distinct from both issues-based learning and problem-based learning, as widely
interpreted (Hesketh, 2003), in that it is neither directed towards pre-conceived
solutions nor set within a particular set of issues or assumptions. Instead IDL involves
a number of inter-related and mutually reinforcing educational methods which
distinguish it from widespread practice and which promote required professional skills
like “examin(ing) and interpret(ing) information and ideas critically (SAICA, 2008:22-
23). Dictionary definitions of ‘issue’ include: ‘the point in question’ and ‘an important
topic of discussion’. IDL centrally involves both of these understandings. There is a
primary focus on grasping an issue as ‘the point in question’ or principal concept(s)
underlying any topic being studied and, through the sense-making process, developing
full understanding of the principle and how and why it is applied. There is also a focus
on contextualizing business knowledge in the ‘issues of the day’.
IDL has been applied through lectures, tutorials and assessment practices and resulted
in students engaging meaningfully with ideas and text, thinking independently and
engaging in thoughtful discussion around issues (Goodier, 2005; Goodier and
Parkinson, 2005; Hesketh, 2003; Skinner, 2009). The approach has been found effective
in enabling students from a range of backgrounds to achieve good marks and pass rates
at various levels and to develop competencies and attributes sought by accounting
education stakeholders. It has been argued before that much of what has been learned is
transferrable to other contexts with core similarities (Hesketh, 2004) and doctoral
findings (Hesketh, 2003) have indeed been confirmed, strengthened and expanded
through ongoing research and experience around IDL-based projects (Goodier, 2005;
Skinner, 2009).
The critical interest and the resultant development of the IDL approach echoes Drinan’s
(1998) argument that progressive educators implementing ELT principles are not
necessarily interpreting the approach fully nor achieving its full potential. Drinan argues
(1998:333) that academics who are applying experience-based learning principles are
generally aiming to
“creat(e) active interdependent and independent learners; holistic, divergent, creative
thinkers; people who can solve problems or improve situations; better communicators;
people who are able to entice the best from others; people who are aware of their own
talents and who are confident in using them”.
However Drinan also maintains that this is not being widely achieved. His argument
resonates with the aims of IDL, and the competencies SAICA (2010a)4, the DoE (1997,
4For instance, SAICA identifies the need for problem-solvers and decision-makers who seek “to
understand, identify and analyse the nature and context of a problem or issue, and to understand the
factors contributing to the problem, before drawing conclusions or considering potential solutions or
courses of action”; and a person who “considers and combines ideas and information from a variety of
sources to create a design, formulate a plan, arrive at a solution..., obtain a broader understanding of an
issue...” (SAICA, 2010a:27); collaboratively develops potential solutions to address root causes of
problems... “exercises professional judgement by selecting or recommending a course of action or by providing advice that is likely to contribute the most to achieving the stated goals...analyses and
JH Hesketh 13
1998, 2007) and other stakeholders seek, when he maintains that educators need to
embrace ‘higher purposes’. These include “generating the desire and ability to think
deeply and holistically” and encouraging a search beyond one’s own preconceptions, so
becoming ultimately innovative and positively critical with respect to self and one’s
profession and society” (Drinan, 1998:334-335).
IDL is highly suited to SAICA and other stakeholders’ requirements. It works precisely
as SAICA’s (2010) competency framework requires accounting education to work. It
works from within existing syllabuses but involves an increased emphasis on
contextualization, the interrelatedness of knowledge areas and the origins, underpinning
principles and implications of practice, thereby promoting knowledge elaboration
(Coles, 1998). The development of these competencies occurs in the context of course
content and hence enhances the mastery of that knowledge. IDL is well suited to
learning accounting because of its essentially context-bound nature, with underpinning
theory easily grounded in current business practice and links easily made to business
issues-of-the-day. Students at any level have some personal experience of business, and
by linking new knowledge to prior experience and to current issues (and by setting tasks
that require students to familiarise themselves with current issues) the knowledge
becomes more accessible and more interesting than if it were de-contextualised.
By locating a new topic within the context in which it is applied, and requiring students
to reach a decision about a related issue, students can grasp the topic’s significance and
its broader implications, and hence understand it fully. At the same time students
develop the higher order technical competencies needed to reach and defend their
decision. Thus, by requiring both technical expertise and understanding of the
significance of their decisions in assessment tasks, students are introduced from the start
to business discourses, the way business people think and competencies required by
SAICA and other stakeholders.
The IDL approach is also suited for curricula from first-year undergraduate to post-
graduate levels of study though clearly the depth and range of conceptual knowledge on
which the issues draw differs according to the level. At the start of their studies, for
instance, the issues require students to make links to practical knowledge of business
and its environment whereas more senior students might be challenged by issues
specifically relevant to auditing principles and practices. At whatever level, the issue is
selected for its capacity to “live fruitfully and creatively in subsequent experiences”
(Kolb and Kolb, 2005, quoting Dewey 1938:28) and to catalyse interest, independent
thought, debate and ongoing reflection.
Issues are contextualised within realistic, topical scenarios which are relevant both to
students’ experience and their academic learning. The issues hence provide meaningful,
‘authentic’ (Jarvis, 1992; 2004) learning experiences, as close to primary experiences as
possible (Jarvis, 2004), given the limitations of secondary experience which is further
removed and requires mediation. Challenging, interesting tasks then provide
disjunctures (Jarvis, 2004) or what Mezirow called ‘disorientating dilemmas’ (Jarvis et
al., 1998) that make students stop and think before they can make sense of new
concepts in relation to past experience or other knowledge areas. Being required to
synthesises the comments of all parties to develop a complete and insightful understanding of the issues
at hand” (SAICA, 2010a:28); “negotiates and reconciles differing views to find acceptable
compromises leading to agreement” (SAICA, 2010a:29).
14 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
interrogate, integrate and internalise knowledge thus enables students to understand the
underlying principles and to help them, first, to see what they still need to know and,
second, to ask questions that will help fill those gaps in a quest for full understanding.
Organising learning around issues hence incorporates Vygotsky’s notion (Jarvis 2004;
Moll, 1990) of working within students’ ‘zone of proximal development’ to challenge
them, with support, to reach beyond what can be learned effortlessly to achieve the
desired level of knowledge and understanding. Support or ‘scaffolding’ includes a
supportive learning environment; well-designed, incrementally challenging tasks;
appropriate resources like interesting, relevant and informative readings; and formative
assessment through which students’ levels of understandings and competencies, or lack
thereof, are demonstrated and addressed where necessary throughout the educational
process. Formative assessment has been found to support rapid development in students
from a range of backgrounds and at different levels, including first year students for
whom English is an additional language and those whose poor schooling left them
under-prepared (Goodier, 2005; Hesketh, 2003).
Assessment strategies associated with IDL are also highly suited to ensuring students
develop the required competencies. The ongoing critique and development of IDL
assessment practices demonstrates its crucial place in the curriculum. The way in which
students are assessed makes explicit what knowledge and competencies are valued in
the various courses, the discipline and the profession. In this way assessment drives
how and what students learn and the depth of understanding they strive to gain.
Appropriate learning is rewarded through ongoing feedback from the start of each
course.
If, for instance, assessment is couched in terms of students advising a client, even at a
basic level, students will develop the technical knowledge and competencies needed to
support their advice. Students are therefore faced with realistic, complex dilemmas and
tasks in homework and tutorial questions, tests and examinations that require them, for
example, to prioritise risk management over reward or social responsibility over
shareholder satisfaction; or to advise ‘clients’ on appropriate accounting treatment. In
order to argue soundly and to debate the issues in the process of reaching a decision,
students need to understand fully the technical knowledge i.e. the principles behind it,
and its significance and implications in reality. Hence IDL assessment requires students
to acquire the competencies in a grounded and meaningful way as required by SAICA
(2010a). It also holds promise for improving student retention rates and promoting
attitudes and approaches associated with research and lifelong learning.
IDL: how it can be applied to meet stakeholder requirements
Having established its suitability theoretically for the purpose of meeting stakeholder
requirements this sub-section demonstrates how IDL can be applied to meet the
challenges posed by SAICA’s competency framework, the DoE and other stakeholders,
providing a means of:
enabling students to develop the knowledge, skills and qualities that have generally
been found lacking;
developing research skills in undergraduate students to enhance learning and to
equip them for honours research projects;
JH Hesketh 15
providing learning opportunities for underprepared students that would improve
pass rates;
offering additional research opportunities for accounting academics who are
interested in developing research-led teaching practices suited to stakeholder
demands.
Because there are logistical obstacles to implementing IDL, recommendations are then
made for changes in university policies, cultures and structures that are obstructive to
IDL’s implementation and hence to meeting stakeholder requirements.
The IDL model (Figure 1) below, developed from earlier models (Hesketh, 2003; In
Press) illustrates how IDL enables students to develop the required competencies and
attributes through three mutually supportive aims: academic, intellectual and personal
growth. Each growth strand is considered separately, somewhat artificially, for purposes
of clarifying the links to IDL principles, making explicit how each aspect of the
approach can be implemented in practice.
Academic growth Aligned to the aim to promote knowledge and competence associated with academic
growth is the DoE (1997; 2004; 2007) demand for helping students make the transition
from reproductive learning and to become lifelong learners, improved pass rates and
research skills. There are also direct links to stakeholder interests in skills like effective
communication, self-directedness, time-management, logical organisation of tasks,
meta-cognition, argument-building, problem-solving, knowledge gathering and
application (SAICA, 2010a: 24-28).
The academic focus involves teaching, learning and assessment techniques that help and
require students to develop the knowledge and competencies needed at their stage of
study. Clearly, the sooner curricula require and offer opportunities to develop
appropriate approaches to knowledge and learning, the sooner students can start
developing them. Hence students challenged with small research tasks from the start of
undergraduate studies will meet the DoE’s honours research requirement relatively
seamlessly. In order to develop academic competencies IDL selects issues and designs
learning processes, tasks and assessment strategies that require and enable students to
build arguments, solve problems, apply knowledge, draw abstractions and make
generalisations to other situations, as reflected in Kolb’s (1984) learning cycle.
Self-reflective processes, through e-journals for example, require students to consider
shortcomings in their learning styles and knowledge bases so that they can remedy the
situation, thereby developing meta-cognitive capacities. The e-journals involve students
in regular email dialogues with their tutors, where writing exercises include
commenting critically on lecture material; current news events; and their own academic
progress.
16 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Figure 1: How issues-driven learning has enabled students to develop required
knowledge, competencies and attributes
PROBLEM-POSING CATALYST
Demanding but Scaffolded Tasks
Social, Environmental & Ethical Issues
Learning Community
Learner-Centred Environment
ISSUES-DRIVEN LEARNING
Real or Realistic Complex Issues
Relevant to New Conceptual Knowledge – Related to Practical Prior Knowledge
INTELLECTUAL ATTITUDE TO KNOWLEDGE & LEARNING
Meaningful Understandings
Independent Thought, Problem-posing, Owning Knowledge
Knowledge elaboration
Critical awareness
Implications for environmental & social responsibility
Autonomous thought Defending viewpoints
Constructing knowledge
Sense-making
Problem-posing
Uncertainty of knowledge
ACADEMIC GROWTH
PERSONAL GROWTH
INTELLECTUAL
GROWTH
Self-directedness
Knowledge application
Metacognition
Knowledge, Skills, Approaches, Problem-solving,
Argument-building
Group work
Assertiveness
Self-concept
Risk-taking
Meaning-negotiation
Emotional intelligence
CRITICALLY REFLECTIVE LIFELONG LEARNER
EFFECTIVE DECISION-MAKER ‘HIGHER SKILLS’
CRITICALITY
WITH MORALITY
KNOWLEDGE &
COMPETENCE
PERSONAL
CONFIDENCE
JH Hesketh 17
Tutors’ responses probe for further understanding, clarification, and more careful
expression and defence of opinions. In this way e-journals provide a safe environment
for students to immerse themselves in the discourse of the discipline, practising relevant
writing skills while exploring issues and making personal sense of new knowledge.
Where the time-consuming nature of the exercise becomes an issue ‘journal buddies’
selected from more senior students are trained to encourage their protegés’ academic
development, with normal moderation procedures implemented. An on-line learning
system is also used to promote reflection and written communication amongst peers
through structured tasks with tutors overseeing these activities (Goodier, 2005).
Tasks in IDL, on-line or not, also require students to be self-directed, to develop
collaborative and co-operative skills through group work, and to practise presenting and
documenting information effectively in written and graphic form as required by SAICA
(2010a) and the DoE (1997).
Academic reading, writing and English language skills are a particular focus in first
year in terms of introducing students to the wider discourses of management studies,
since appropriate use of language is best learned in the context of course material where
“the content is intimately bound up with how to read write and speak about a discipline”
(Goodier and Parkinson, 2005:66). Students generally, and underprepared students
particularly, can be overwhelmed by academic reading demands and writing tasks like
summarising, analysis, synthesis and report-writing (Hesketh, 2003). However, students
can learn surprisingly quickly and well from regular, short, authentic writing exercises
resembling the “real activities that members of the discourse community engage in”
(Goodier and Parkinson, 2005:67). Examples include writing short analyses or
summaries of scenarios in preparation for discussion, or writing about how learning
relates to current news events in e-journals.
Consistent formative assessment gives regular feedback on academic progress. Central
to this process are regular, short ‘concept tests’ that test understanding of “important
main concepts, to synthesise these concisely and to relate them to existing knowledge”
making explicit the kind of conceptual understandings required and motivating in
students serious attempts to make sense of what they are learning, in relation to the big
picture (Goodier, 2005:8). Assessment of writing tasks involves focus on content,
argument structure, use of language and referencing with capacities to assess
formatively clearly dependent on tutor quality and commitment. There are therefore
implications for tutor recruitment, development and monitoring, ongoing tutorial
development processes, and of course the funding needed to ensure that there are
sufficient tutors who are properly paid for their important role in learning and
assessment processes. Where large classes appear prohibitive to formative assessment,
ways need to be found to assess in ways that drive effective learning. For example,
marks might be awarded for all or part of students’ work and students need not know
when their work will be selected for assessment or when they are called on to present
their solutions orally.
Small-group learning in tutorials, and sometimes in lectures, is organised around
carefully designed tasks, to promote the gathering and development of information and
ideas, analysis of situations in order to makes sense of them and to reach solutions or
identify appropriate treatments through collaborative teamwork. The outcome is then
presented orally to the class, defended and debated. Hence students develop required
18 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
competencies and acquire knowledge through meaning-making processes that mark
effective students, from their first year.
Clearly the extent to which students develop in these directions varies considerably but
it is interesting to see how rapidly students adapt their approaches and develop their
knowledge and skills in ways that are rewarded through assessment practices (Goodier
and Parkinson, 2005; Hesketh, 2003).
An IDL-based course has been offered to first-year Bachelor of Commerce students
from a range of socio-economic backgrounds and schooling at a South African
university over the past 14 years. The course has been developed in response to ongoing
evaluation and regular student course evaluations (Goodier, 2005; Hesketh, 2003) have
contributed to the evaluative process. Student responses have been consistently very
positive overall and the following comments are representative of the majority of
students’ opinions relevant to IDL’s academic growth strand (Goodier, 2005; Hesketh,
2003):
It gave me a chance to put my skills into practice and see how they fit in a real
world situation.
Develops fluency in communication.
I learned self-discipline.
I became independent.
I wasn’t aware of my circular arguments but now I am thanks to the weekly analysis
tests.
Forces me to write concise and structured pieces.
I’ve learnt to form my own opinions on issues and argue my point, rather than
sitting back and being quiet.
Has improved my writing skills as most of the course involves rigorous writing.
...[it] requires a vast quantity of talking and thinking on your feet. This is good as it
prepares you for the real world.
[The tasks were] very good exercises. Made me think critically, argue certain points
in my head. Mini papers – helped analyse the faults that were made during the
tests...
[The research project] seemed difficult at the beginning but was manageable.
[E-journals] were time-consuming and always took me longer than the allocated
time, but they helped me greatly in ensuring I was up to speed with current news.
[The tutorials represented] great debates. Discussions, learning!!!
JH Hesketh 19
[Developing a business proposal was] very good for a future business person; was
challenging but fun at the same time. It should never be taken out of the course.
...[it]was an interesting, fun and informative course that has taught me skills that
will last me a lifetime.
Hence students have experienced IDL as a means of developing ‘academic’ skills
closely aligned to professional skills and personal attributes set out in the competency
framework (SAICA, 2010a: 25-29).
Personal growth
The focus on personal growth reflects DoE requirements for helping students make the
transition from poor schooling and socio-economic environments (DoE, 1997) and
related lack of self-confidence, and hence improving pass rates (DoE, 2004). It is also
aligned with SAICA’s requirement that a person “treats others respectfully, courteously
and equitably; shows empathy by understanding why others have a particular
perspective on an issue; resolves conflict and differences of opinion by focusing on
issues, not personalities”; and “defers to others when more experience or greater
expertise” is needed (SAICA, 2010a:23).
Underpinning IDL is a learner-centred environment with a ‘whole-person’ learning
approach and relatively democratic teaching styles (Griffin, 2002; Hesketh, 2003;
Jarvis, 2004). This kind of teaching draws on students’ past experiences, values their
perspectives on issues and promotes respect for different viewpoints, while requiring
logical argument. Students’ perspectives are contributed through interactive, knowledge
constructing, collaborative processes organised around issues that need to be resolved or
decisions on specific situations reached. These small-group learning experiences
involve developing competencies like “seek(ing) and shar(ing) information, facts and
opinions through written and oral discussion” (SAICA, 2010a:25) since students often
need to work together to negotiate meaning and make sense of a scenario before they
can complete tasks. The learning process requires students to accommodate diverse
perspectives, to experience compromise and to understand something of the
responsibility of a decision-maker whose solutions can affect real lives. Hence the tasks
provide opportunities for students to develop emotional intelligence (Goleman, 1996) in
the form of self-worth, emotional well-being, assertiveness, self-awareness and other-
centredness, attributes required by employers and SAICA, as noted above.
Competencies like these are developed in an environment of trust and mutual respect
associated with a functional learning community. Students feel ‘safe’ enough to
brainstorm freely and to take risks in exploring their own and their peers’ thinking, in
relation to prior understandings, values, beliefs and new knowledge. Unsettling as it is
initially for students to engage in this kind of learning and to experience knowledge as
uncertain and contestable, it is personally empowering for them to see the relevance of
their own experience and the value of their views, marking an important milestone in
their learning biographies. (Hesketh, 2003).
Ways have to be found to engage shy or less participative students in the sense-making
processes. One way is to ensure all group members have specific responsibilities, and
know that they may be called on to present their group’s solution to the rest of the class.
20 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
(Creative assessment strategies like spot checks or peer assessment can be used to hold
students accountable to their tasks).
In student evaluations of the first-year IDL-based course mentioned above, comments
relating to personal growth and the learner-centred environment representative of the
majority of students’ opinions (Goodier, 2005; Hesketh, 2003) include:
It helped me deal with the culture shock and to settle into university life.
With our different backgrounds and ideas we knew different things and helped each
other.
I think we are an unbeatable family.
My classmates were like reference books.
Your spirit as a researcher and learner is raised.
The course enabled one to speak in small groups as we used to make presentation in
our tutorial groups. I wasn’t able to do such a thing before.
The fact that the class discussions are informal made me feel comfortable enough to
voice my opinions.
...at first I was classifying my ideas as inferior but because of the great need to
contribute, I got myself used to collaborating and co-operating with others.
...I can now collaborate with people of a different race without fear of being
discriminated [against]. This has boosted my confidence.
Before I used to take things at face value and look at issues from my own
perspective, but now I’ve learned to look at all sides....
Students’ voices thus indicate that IDL implemented even at first-year level helped
them develop skills, personal attributes and life-long learning attitudes required by
stakeholders including those set out by SAICA (2010a: 22-24).
Intellectual growth
The model illustrates how intellectual growth is the central strand to developing the
intellectual attitudes to knowledge and learning associated with critically reflective
students, lifelong learners and effective decision-makers, with the higher order skills
(Drinan, 1997) and advanced knowledge base (Ryan, 1997) required by stakeholders.
Crucial as it is, however, the intellectual strand clearly relies on academic and personal
growth and the underpinning learner-centred environment. Intellectual growth is
aligned closely with the requirement that a student “examines and interprets information
and ideas critically...identifies the purpose of the analysis and the information and/or
ideas and material to be considered ... identifies information that needs to be verified ...
evaluates information and ideas ... forms an opinion or reaches a conclusion that the
information does or does not fulfil the purpose of the evaluation... considers and
JH Hesketh 21
combines ideas and information from a variety of sources to ... arrive at a solution to a
problem, obtain a broader understanding of an issue ... draws conclusions / forms
opinions ... solves problems and makes decisions... identifies and diagnoses problems
and/or issues ... seeks to understand, identify and analyse the nature and context of a
problem or issue and to understand the factors contributing to the problem, before
drawing conclusions or considering potential solutions or courses of action” (SAICA,
2010a:26-27).
Critical reflection occurs when students at any level are confronted with controversial,
topical, relevant, accessible issues that are accompanied by appropriate teaching,
learning and assessment strategies. These strategies require students to think, applying
knowledge, identifying and gathering information needed for forming opinions, making
decisions or reaching conclusions, examining assumptions underpinning suggested
knowledge, strategies or treatments and posing questions in order to identify incorrect
accounting treatments – preferably in interactive learning processes.
While students often feel threatened initially by the requirement to develop new, more
intellectual ways of knowing, tasks are structured to enable them to make the shift.
Unresolved issues lacking consensual decisions live on in students’ minds, compelling
further thought in the quest for resolution (Wassermann, 1993), demonstrating students’
engagement with subject matter and sense-making processes as they make links
between issues and knowledge areas. An IDL approach gives the lecturer – or facilitator
of learning – opportunities also to relate new issues and questions to previous ones and
indeed to role-model critically reflective approaches, independent thought and
willingness to accept, and indeed enjoy the cut and thrust of developing different,
logical arguments. Role-modelling questioning, critical approaches to knowledge and
learning enhances course coherence and integrity and hence the notion of a critical
community of learners.
The reflective IDL processes involve students in making sense of a situation and related
knowledge (Hesketh, In Press; Schwandt, 2005) through both independent and
collaborative learning processes. Tasks include independent research, independent
analyses of scenarios in preparation for collaborative meaning-making processes;
preparation for and participation in class debates; and interactive analysis and decision-
making in small groups.
Complex, realistic issues and related tasks enable students to experience knowledge as
uncertain, contestable, value-laden and open to diverse interpretation or judgment.
Tasks require students to problematise knowledge, testing it in new situations in light of
their own understanding and, if necessary, to modify their thinking accordingly (Kolb,
1984; Kolb and Kolb, 2005; Slonimsky and Shalem, 2006). Students need to argue
points while reaching decisions about the scenario confronting them, thereby taking
ownership of knowledge, developing deep (rather than superficial) understandings of
new concepts (Marton and Säljö, 1976, Entwistle, 1987, 1992). Lecturers and tutors
from a range of courses commented on how first-year students exposed to the IDL
approach engaged far more actively with learning in their classes than other students
and said that the quality of their questions demonstrated a far greater depth of
understanding as did their marks (Hesketh, 2003). This is reminiscent of Barth’s (2011)
argument that students should be asking “why, why, why?”
22 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
The capacity to think independently is closely associated with abilities to deeply process
and elaborate knowledge (Coles, 1998; Drinan, 1998; Slonimsky and Shalem, 2006),
requiring students to accommodate ideas and information within their personal
conceptual frameworks to make sense of knowledge in relation to real situations.
Knowledge elaboration also requires learners to make the connections between different
areas and forms of knowledge as occurs in reality, and to see how theory links with
practice (Coles, 1998). Elaboration processes hence involve sense-making and increase
students’ capacities to recall conceptual knowledge at a later stage and to think
creatively and autonomously (Drinan, 1998). Students need critical, curious attitudes
associated with discovery learning (Dewey, 1916, 1938) in order to make sense of new
knowledge and to make it their own. To understand deeply and in relation to reality,
students need to come to grips, for instance, with why a certain treatment or approach is
appropriate to a situation and not just how to implement it. While it is required that a
candidate “identifies the purpose of the computation(s) and whether a precise
calculation, an estimate, a forecast, or a projection is required” and “identifies the
purpose of gathering information” (SAICA, 2010a:26) it has been seen above that these
competencies were reported lacking in the Canadian and South African
evaluators’/examiners’ reports (Chartered Accountants of Canada, 2007; SAICA,
2009).
Represented in the IDL model is the relation between criticality, ethics and social
awareness, or a moral orientation, as a necessary condition and disposition for
‘integrated critical thinking’ (Mason, 2000). This notion is supported by Naudé’s (2008)
argument for integrating ethics into courses in order for students to take ethics seriously
and recognise ethical matters to be integral to their field of study. Naudé (2008) argues
that intellectual skills are developed through tasks round ethics-related issues that
require students to think analytically and critically and to judge soundly. This argument
links to the need for students to develop sceptical attitudes and intellectual responses to
situations, if they are to be prepared for decision-making in uncertain, ever-changing
business contexts of the future. Students should therefore be assessed, inter alia, for
abilities to take into account the wider implications of their solutions within the broad
business environment. Hence, the intellectual growth strand of the IDL model links also
to SAICA’s (2010a:20) requirement that a student “identifies ethical dilemmas” and
“makes appropriate ethical judgements”. Clearly the moral or ethical behaviour of a
student cannot be assessed in professional examinations but awareness of ethical issues
can.
Requiring students to develop more intellectual attitudes to knowledge and approaches
to learning is also core to equipping students for lifelong learning, not only in the sense
of being interested in continuously attending new courses but, very importantly, in the
curiosity and interest in learning that they display throughout their lives.
Responses representative of the majority of students’ voices relating to their intellectual
growth through the above-mentioned IDL-based course (Goodier, 2005; Hesketh, 2003)
include:
Hearing different views helped us come to terms with our own points of view.
To draw conclusions one just needs reasoning; I learned to use logic and back up my
ideas, to formulate my own opinion.
JH Hesketh 23
We have to decide by ourselves.
It’s important as I don’t merely want to go with others’ opinions.
The topics covered were usually controversial and thus there was almost always
conflict of opinion – this led to interesting and exciting discussions in class.
I think [the value of the IDL approach] is that it forces students to start thinking
about current issues, morality, the environment, etc...it encourages students to
interact, discuss and air their views...the raising of awareness about world issues is
its strongest feature.
[It] made me very aware of certain business environments and introduced me to how
business is done and that although success and money is the main goal, it is
important to maintain morals and ethics in business.
I did not like it at the beginning and did not see the relevance it has on my life as a
future chartered accountant. But now I see that [it] is actually a great course that
opens your mind to all aspects of the world. It has enriched me and I have learnt a
lot from it.
[It] is a hectic course. It is totally different from the other courses, it requires a lot of
time and critical thinking...
[It] is a brilliant course, you learn a lot from it and these issues always apply and
you can apply them to other courses.
Hence students at first-year level express the sense that IDL challenged them to think in
new ways.
This sub-section has demonstrated how IDL can be implemented to meet the
stakeholders’ needs in terms of developing competencies and knowledge that have been
found lacking; developing research skills in students and providing learning
opportunities that can affect positively student throughput rates. There are, however a
number of obstacles to its implementation within university environments and these are
presented below with some recommendations for changes to university policies and
structures that are obstructive to implementing IDL or to achieving stakeholder
requirements.
Obstacles to implementing IDL and recommendations for change
Changing from a ‘traditional’ educational approach to implementing IDL requires a
great deal of commitment and yet efforts of this nature are often not supported by
university policies and structures. Change of this nature brings with it implications for
human and financial resources decisions, and hence for university structures and
policies that have so often come to prioritise efficiency over effectiveness (Bitzer, 2006;
Jarvis, 2001; Sunderland and Graham, 2001). Effective, broad implementation of IDL
would need institutional support (Biggs, 1996).
Policies and structures that represent obstacles to academics’ abilities to meet
stakeholder demands, specifically in accounting schools, need serious re-consideration.
24 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
The quest to increase academics’ research outputs and to improve student throughput
rates and graduate competencies, including research, will involve allocations of
financial resources to support ongoing researched improvement to teaching and learning
through, for example:
appropriate development of tutorial programmes and payment for quality tutors;
provision of appropriate teaching venues and IT facilities where necessary;
building teaching and assessment expertise in academic staff in contexts that include
large classes;
and adjustment of human resource and research policies around:
suitable payment categories for IDL tutors;
rewarding excellence in teaching;
promotion criteria.
Some recommendations are made in relation to developing tutorial programmes;
improving teaching facilities; developing teaching expertise; rewarding teaching
excellence; and broadening promotion criteria.
Developing tutorial programmes
Central to IDL is the opportunity for students to engage collaboratively and
interactively with conceptual knowledge. Much of the learning takes place in tutorials
making the tutorial programme crucial in achieving required competencies, but tutorial
effectiveness depends on first, the recruitment of potentially good tutors; second,
ongoing development of tutorials and tutors; third, adequate payment of tutors; and
fourth, workable tutor: student ratios (Skinner, 2009).
The nature of the tutorials make it ideal to limit student numbers to 20-25, though when
budgets, tutor availability or tutorial venues prohibit ideal class sizes the approach can
be applied in larger classes with relevant expertise, proper management and tutor
development. Senior students, if not postgraduates, should be used as tutors and given
appropriate opportunities to develop their learning-facilitation skills. It bears noting that
tutors implementing principle-based rather than rules-based learning, in line with IDL
principles, have found that their tutoring has helped them come to grips in new ways
with their own learning (Watson, 2010; 2011). This is particularly the case for honours
students tutoring third-years and benefits are further enhanced when third-year and
honours courses’ subject matter is synchronised making third-year tutorial content
concurrent with tutors’ own learning focus. Another important spin-off experienced
from developing students as tutors is that they have been enthused by the interaction
and approach to facilitating learning and have expressed some interest in academia as a
career option.
The tutorials should not be a re-teach of lectures but a means of engaging students in
making sense of concepts introduced in lectures. Tutors need to buy into IDL principles
and to spend time preparing for tutorials, marking tests and assignments, and
JH Hesketh 25
moderating marks, participating in tutorial/material development processes with
academic staff as far as possible. Tutorials should include weekly short ‘concept tests’
that contribute to the course mark. Formative assessment of tests and other class work
give ongoing feedback to tutors on students’ competencies and understanding and also
give rapid and regular feedback to students on their academic progress, warning them in
good time when they need to apply new learning approaches. The tests also ensure
students learn from the start of their courses. Tutorials should offer opportunities for
students to grapple with writing tasks, using appropriate language to express the logic
of their understandings and arguments, and tutors should be trained to give formative
feedback. This could include probing questions on reflective e-journals to provoke
appropriate critique or action from students. Hence payment of tutors has to reflect
fairly their effort, time and expertise (Skinner, 2009) and payment structures and
policies need to accommodate tutors’ implementation of IDL-like approaches.
The increased costs should be weighed against financial benefits to the school and
university in terms of increased student retention, throughput and pass rates and related
reputational benefits that help attract and retain high quality students and staff and
indeed support from other stakeholders.
Improving teaching facilities
With their vested interest in good learning outcomes, universities need to prioritise the
provision of facilities suited to the purpose. If teaching and learning would be enhanced,
for instance, by access to facilities like flat venues with good acoustics for tutorials,
adequate venues should be available. Similarly, if dealing with large numbers of
students indicates the need for Information Technology (IT) to support ways of
applying sound teaching, learning and assessment principles, through e-learning for
example, appropriate IT facilities should be made available.
Developing teaching expertise
Applying new approaches to teaching and learning involves first, recognising that
teaching based only on ‘craft knowledge’ may well be inadequate for the purpose of
enabling students to develop the required knowledge, skills and qualities (Scott et al.,
2007); second, taking the underlying educational principles on board both technically
and philosophically (Hesketh, 2003, 2004); and third, developing a scholarly interest in
teaching and ongoing curriculum development.
Hence teaching and learning needs to be brought explicitly into accounting academics’
conversations and debates and ways need to be found to facilitate such interaction,
through processes like: collegial work on improving courses; regular semi-formal
forums for reflection and discussion; collaboration with educational specialists in
curriculum development / research projects; presenting findings from teaching
practices; attending courses in higher education. The last would help directly in
applying ELT-based approaches like IDL to challenges that include: setting tasks that
evoke the kind of learning that is desired; assessing in ways that reward knowledge,
skills and qualities that are valued; facilitating and assessing learning in different
contexts, from large classes to small groups; and dealing with student diversity.
Educators for medical professions have found that teaching expertise was nurtured
through pro-teaching environments that accommodated pro-teaching governance
26 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
structures including curriculum offices offering strong administrative support
prestigious academic career paths (Irby, 2000; Kirkland, 2000; Whitcomb, 2000).
Rewarding teaching excellence
If teaching scholarship is to be seriously promoted, contributions to teaching and
learning need to be properly recognised and rewarded. Universities should find ways to
give status that equates with research status (Bitzer, 2007), to effective practices
informed by educational theory that is justified in terms of intended learning outcomes
or “systematic knowledge” of higher education teaching and learning processes
informed by reflection, literature and research (Scott et al., 2007). This could be
achieved by establishing an infrastructure for peer-review similar to that for research
(van Fleet and Peterson, 2005). Teaching excellence should be recognised also in
academics whose practices play a role in influencing curricula improvements and in
testing and developing educational theory. Promotion policies like those of the
University of Cape Town (UCT) should incorporate contributions to improving
teaching and learning (UCT, 1999).
Broadening promotion criteria
When research output drives promotion, universities should consider new ways of
thinking about the nature of research that incorporate both discipline-based and
education-based scholarly work.
Universities should, for instance, find ways to encourage practitioner-research,
including doctoral work, to promote progressive, theoretically-informed understandings
of teaching, learning and assessment within disciplines (Hesketh, 2003). Theorising
practice will provide new opportunities for academics to reflect critically on and
improve curricula, providing a platform for sharing findings with colleagues. At the
same time critical reflection on practices in different contexts, in relation to theory, will
push the boundaries of curriculum, specifically teaching and learning, theory and make
important contributions to the ongoing development of practical and theoretical
educational understandings.
Since good teaching clearly implies scholarship within the discipline taught, ways
should be found to draw together the cultures of teaching and research in order to
promote excellence in teaching. There will be little incentive to engage in scholarship
around teaching and learning until both discipline-based and education-based research,
or ideally an integration of both, in different, approved balances, are given the status
they deserve on promotion panels (Bitzer, 2007; van Fleet and Peterson, 2005).
Promotion rules should incorporate a notion of scholarship that is broader than, but
includes, research and these rules should be applied in ways that make sense in terms of
contributing to knowledge in different disciplines. For instance, in accounting schools,
scholarly work might be reflected through contributions to policy documents with
independent evidence of the influence of this work. Scholarship might include crucial
involvement in local or international standard setting processes and contributions to
accounting thought rather than refereed publications. Work might involve resolving
problems of principle and revising definitions underpinning structures of financial
reporting. The end product of such work will not result in a research publication but will
be subjected to intense scrutiny because of its importance to global standard setting,
JH Hesketh 27
contribution to accounting knowledge and capacity to promote defensible
improvements to financial reporting. Scholarship has been acknowledged in these kinds
of ways at the UCT (2010).
Scholarship could also be evaluated in terms of producing quality text books. Criteria
could be set for reviewing scholarship, perhaps through anonymous review processes,
in terms of effectiveness in: promoting understandings of principles or concepts rather
than rules (Barth, 2008, 2011; Watson, 2010); demonstrating theoretically-based
understandings of the discipline (Barth, 2008, 2011; Watson, 2010) and demonstrating
understandings of teaching and learning through explicit linking of intended learning
outcomes to learning theory or systematic educational knowledge informed by research
(Scott et al., 2007). Producing text-books that meet these criteria could then add weight
to scholarship-based applications for promotion.
Clearly each university and accounting school will have to work together to find ways
forward that will allow the schools to meet university and stakeholder demands.
Universities’ research-centred promotion policies can threaten accounting schools’
capacities to provide strong departmental leadership and administration, with associated
contributions to the university. For example, if school heads resign their positions
because onerous responsibilities make it difficult for them to meet promotion criteria
like research output, schools might well find it difficult to function optimally and
indeed to meet SAICA’s range of accreditation criteria. Indeed this scenario has the
potential to affect very negatively schools’ capacities to meet the stakeholder
requirements discussed above, specifically to drive the development of teaching and
learning strategies in line with SAICA’s competency framework.
CONCLUSION
The stakeholder requirements are both challenging and urgent and call for some
fundamental re-thinking on teaching, learning and assessment practices. The article has
argued that IDL offers a suitable means of addressing the requirements and has
demonstrated how it can be used to enable students to: develop the knowledge, skills
and qualities that have generally been found lacking; develop research skills in
undergraduate students to enhance learning and to equip them for honours research
projects; provide appropriate learning opportunities to promote student retention; and
offer additional research opportunities for accounting academics who are interested in
developing research-informed teaching practices.
However the article is not intended to be prescriptive on the use of IDL. The theoretical
underpinnings of ELT and IDL have therefore been made explicit in order to allow
readers to consider for themselves the relevance of the theories to their own contexts;
whether and to what extent they might consider implementing the ideas; and how they
would do so.
Aside from demonstrating how IDL works to meet stakeholder requirements, the article
has contextualised SAICA’s and the DoE’s requirements within a bigger picture of very
similar challenges facing others elsewhere. The confluence of current thought and
experience of SAICA with other accounting bodies, other professional bodies and
employers generally, has been demonstrated to add credence and weight to the
competency framework (SAICA, 2010a) requirements and to thereby confirm the
competency framework’s importance and relevance to higher education and the
28 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
accounting profession. Hence the article has argued that there is no space for
dismissive, simplistic responses to the stakeholder challenges, however tough the
requirements might appear in certain contexts. If ways forward are to be found for
implementing appropriate educational approaches, it is important that accounting
academics not only acknowledge that change is necessary but also consider how it
might be implemented.
Obstacles to implementing IDL and thereby meeting stakeholder requirements are
identified so that they can be acknowledged and addressed. They relate to financial
resources, human resources and research policies that could hinder academics’ efforts to
implement educational approaches like IDL. Hence overcoming the obstacles represents
another set of challenges. For this reason the article includes recommendations for the
kind of changes to university structures and policies that would enable academics to
respond in educationally appropriate ways to stakeholder requirements.
Since time is short for implementing new teaching, learning and assessment strategies it
is important to share ideas. This article is intended to catalyse collaborative action but at
the very least to engender critical reflection and debate on how accounting academics
can best prepare their students for the new-style professional exams from 2013.
ACKNOWLEDGEMENTS
The author wishes to express sincere appreciation to Drs Jane Skinner and Caroline
Goodier for years of collaborative work on which this article draws and for their
comments on the first draft, and to the anonymous reviewers and editor for their helpful
comments. Grateful thanks also go to Alex Hesketh for his unwaveringly cheerful
technical support.
REFERENCES
Atwater, J.B., Kannan, V.R. and Stephens, A.A. (2008). Cultivating systemic thinking
in the next generation of business leaders. Academy of Management Learning and
Education, 7 (1):9-25.
Barth, M. 2008. Global financial reporting: implications for U.S. academics. The
Accounting Review, 83 (5):1159-1179.
Barth, M. (2011). Seminar at University of Pretoria, 21 June 2011.
Beenen, G. and Pinto, J. (2009). Resisting organizational-level corruption: an interview
with Sherron Watkins. Academy of Management Learning and Education, 8 (2):275-
289.
Biggs, J. (1996). Assessing learning quality: reconciling institutional, staff and
educational demands. Assessment and Evaluation in Higher Education, 21(1):5-16.
Birkinshaw, J. (2009). Talking management for the Globe and Mail with Karl Moore
(talking to Julian Birkinshaw, professor and dean at London Business School)
http://theglobeandmail.com/report-on-business/ managing/man. Accessed 12 Oct 2009.
JH Hesketh 29
Bitzer, E.M. (2006). Restoring the status of teaching scholarship at a research-orientated
university. South African Journal of Education, 20(4):372-390.
Boud, D. and Feletti, G. (1998). Changing problem-based learning: introduction to the
second edition. In Boud, D.. and Feletti, G. (eds.). The challenge of problem-based
learning: 2nd edition, Kogan Page, London, pp.1-14.
Chartered Accountants of Canada (2007). Uniform Evaluation Report. Toronto:
Canadian Institute of Chartered Accountants.
Coles, C. (1998). Is problem-based learning the only way? In Boud, D. and Feletti, G.
(Eds.). The challenge of problem-based learning: 2nd
edition, Kogan Page, London,
pp.313-325.
Clinebell, S. K. and Clinebell, J. M. (2008). The tension in business education between
academic rigor and real-world relevance: the role of executive professors. Academy of
Management Learning & Education, 7 (1):99-107.
D’Andrea, V., Gosling, D., Scott, I. and Tyeku, S. (2002). Outline of current HE policy
goals and possible areas of capacity building in relation to teaching and learning.
Council on Higher Education’s Higher Education Quality Committee Improving
Teaching and Learning Project. Pretoria.
Department of Education. (1997). Education White Paper 3: A programme for the
transformation of higher education. Government Gazette, 386. Pretoria: Government
Printer.
Department of Education. (1998). South African Qualifications Authority. Government
Gazette, 393. Pretoria: Government Printer
Department of Education. (2007). The Higher Education Qualifications Framework.
Government Gazette, 928. Pretoria: Government Printer.
Department of Education. (2004). A new funding framework: how government grants
are allocated to public higher education institutions. Pretoria: Government Printer.
Dewey, J. (1916). Education and democracy. The Free Press, Macmillan Publishing Co:
New York.
Dewey, J. (1938). Experience and education. Macmillan: New York.
Dominicé, P. F. (2000). Learning from our Lives. Jossey-Bass, San Francisco.
Drinan, J. (1997). The limits of problem-based learning. In Boud, D. and Feletti, G.
(Eds.). The challenge of problem-based learning: 2nd edition. London: Kogan Page
pp.333-339.
Entwistle, N.J. (1987). A model of the teaching-learning process. In J.T.E. Richardson,
M.W. Eysenck, and D. Warren Piper (eds). Student Learning: Research in Education
and Cognitive Psychology. Milton Keynes: SRHE/Open University Press.
30 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Entwistle, N.J. (1992). The Impact of Teaching on Learning Outcomes in Higher
Education. Sheffield: Universities' and Colleges Staff Development Unit.
Fenton-O’Creevy, M.P., Knight, P and Margolis, J. (2006). A Practice-Centered
Approach to Management Education in a Distance Learning Context. In C. Wankel and
R. DeFillippi. (2006) New Visions of Graduate Management Education: Research in
Management Education and Development, vol. 5. Greenwich, CT: Information Age
Giacalone, R.A. and Thompson, K. R. (2006). Business ethics and social responsibility
education: shifting the worldview. Academy of Management Learning and Education,
2006. Vol 5. No. 3, 266-277.
Goleman, D. (1996). Emotional intelligence: why it can matter more than IQ. London:
Bloomsbury.
Goodier, C. (2005). Report on academic/business literary courses: IBS and IBSM.
Unpublished. Management Studies Education Unit, University of KwaZulu-Natal.
Goodier, C. and Parkinson, J. (2005). Discipline-based academic literacy in two
contexts. Journal for Language Teaching 39(1):66-79.
Gregory, J. (2002). Principles of experiential education. In P. Jarvis, (ed.). The theory
and practice of teaching, 94-107. London: Kogan Page Limited.
Grey, C. (2004). Reinventing business schools: the contribution of critical management
studies. Academy of Management Learning and Education, 3 (2):178-186.
Griesel, H and Parker, B. (2009). Graduate attributes: A baseline study on South
African graduates from the perspective of employers. Higher Education South Africa
(HESA) and South African Qualification Authority (SAQA)
Griffin, C. (2002). Didacticism: lectures and lecturing. In P. Jarvis (ed.). The theory and
practice of teaching, 94-107. London: Kogan Page Limited.
Grundy, S. (1987). Curriculum: product or praxis. London: The Falmer Press.
Henry, J. (1989). Meaning and practice in experiential learning. In Weil, S. W. and
McGill, I. (Eds). Making sense of experiential learning. Buckingham: SRHE and OUP,
pp. 25-37.
Hesketh, J. (2003). A management studies curriculum for free thought in a changing
South African context: learning from a unique experience. Doctoral thesis. Guildford:
University of Surrey.
Hesketh, J. (2004). What can we know from case study research? Journal of Education,
Kenton Special Issue, 33.
Hesketh, J. (In Press). Ch 16: Section on ‘Learning’. In Wankel, C. (ed.) Management
through Collaboration: teaming in a networked world. New York: Routledge.
JH Hesketh 31
Hilton. M. (2008). Skills for work in the 21st century: What does the research tell us?
The Academy of Management Perspectives, 22(4), November 2008: 63-78.
International Federation of Accountants (IFAC) (2010). International education
information paper: development and management of final written examinations based
on the practices of selected IFAC member bodies. http://www.ifac.org [Accessed 29
May 2010].
Independent Regulatory Board for Auditors, (2008). Public Practice Examination
November 2008: examiners’ comments.
www.irba.co.za/index.php?option=com_docman&task=doc...35, [Accessed 25 June
2010].
Irby, D. (2000). University of California, San Francisco, School of Medicine. In The
education of medical students: ten stories of curriculum change. Association of
American Medical Colleges / Milbank Memorial Fund, pp 26-40.
http://www.milbank.org/reports/americanmedical colleges/0010medical... [Accessed 23
December 2010].
Jacobs, D. (2007). Critical biography and management education. Academy of
Management Learning and Education, 6 (1):104-108.
Jarvis, P. (1999). The practitioner-researcher: Developing theory from practice. San
Francisco: Jossey Bass Publishers.
Jarvis, P. (2001). Universities and corporate universities – the higher learning industry
in global societies. London: Kogan Page.
Jarvis, P. (2002). Teaching styles and teaching methods. In P. Jarvis, P. (ed.). The
Theory and Practice of Teaching. London: Kogan Page, pp. 22-30.
Jarvis, P. (2004). Adult education and lifelong learning: theory and practice. 3rd
edition.
London: RoutledgeFalmer.
Jarvis, P., Holford, J. and Griffin, C. (1998). The theory and practice of learning.
London: Kogan Page.
Johnstone, R. and Vignaendra, S. (2003). Learning outcomes and curriculum
development in law: a report commissioned by the Australian Universities Teaching
Committee (AUTC). Higher Education Group, Department of Education, Science and
Training, Australia.
Kirkland, R.T. (2000). Baylor College of Medicine. In The education of medical
students: ten stories of curriculum change. Association of American Medical Colleges /
Milbank Memorial Fund, pp14-26.
http://www.milbank.org/reports/americanmedical colleges/0010medical...[Accessed 23
December 2010].
Kolb, D.A. (1984). Experiential learning: experience as the source of learning and
development. New Jersey: Prentice-Hall.
32 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Kolb, A.Y. and Kolb, D.A. (2005). Learning styles and learning spaces: enhancing
experiential learning in higher education. Academy of Management Learning and
Education 4 (2): 193-212.
Learmonth, M. (2007). Critical management education in action: personal tales of
management unlearning. Academy of Management Learning and Education, 6 (1):109-
113.
Liang, N. and Wang, J. (2004). Implicit mental models in teaching cases: an empirical
study of popular MBA cases in the United States and China. Academy of Management
Learning and Education, 3 (4):397-413.
Marton, F. and Säljö, R. (1976). On qualitative differences in learning – outcome and
process. British Journal of Educational Psychology, 46:4-11.
Mason, M. (2000). On the poverty of thought in contemporary education: an integrated
conception of critical thinking. Unpublished paper. Port Elizabeth: Kenton Education
Association Conference, October 27-29, 2000.
Merriam, S.B. (1993). Taking stock. Merriam, S.B (Ed.) An update on adult learning
theory. San Francisco: Jossey Bass, 105-110.
Mezirow, J. (1990). How critical reflection triggers transformative learning. In J.
Mezirow and Associates. Fostering critical reflection in adulthood – A guide to
transformative and emancipatory learning. San Francisco: Jossey-Bass, 1-20.
Mulligan, J. and Griffin, C. (eds). (1992). Empowerment through experiential learning:
explorations of good practice. London: Kogan Page.
Navarro, P. (2008). The MBA core curricula of top-ranked U.S. business schools: a
study in failure? Academy of Management Learning and Education, 7 (1):108-123.
Naudé, P. J. (2008). Ethics education in accounting: An outsider perspective. South
African Journal of Accounting Research, 22 (1):1-17.
Oberholzer, R. (2007). A note: Study abroad programs: A potential key to successful
accounting and management education in a global world. South African Journal of
Accounting Research, 21 (1):127-146.
Olivier, M. and Kleinhans, A. (2008). Developing the leaders of the future: defining and
delivering competent chartered accountants. Unpublished paper to explain the
competency framework. SAICA, Bruma Lake
Reynolds, M. (1999). Critical reflection and management education: rehabilitating less
hierarchical approaches. Journal of Management Education, 23(5) pp. 537-554.
Reynolds, M. and Vince, R. (2004). Critical management education and action-based
learning: synergies and contradictions. Academy of Management Learning and
Education, 3 (4):442-456).
JH Hesketh 33
Ryan, G. (1997). Ensuring that students develop an adequate, and well-structured,
knowledge base. In Boud, D. and Feletti, G. (Eds.). The Challenge of Problem-based
Learning – 2nd edition. Kogan Page: London, pp.125-136.
Samuelson, J. (2006). The new rigor: beyond the right answer. Academy of
Management Learning and Education, 5 (3):356-365.
Schwandt, M. (2005). When managers become philosophers: integrating learning with
sense-making. Academy of Management Learning and Education, 4 (2):176-192.
Scott, I., Yeld, N. and Hendry, J. (2007). A case for improving teaching and learning in
South African Higher Education. Research paper prepared for the Council on Higher
Education. Higher Education Monitor, 6, October 2007.
Shareef, R. (2007). Want better business theories? Maybe Karl Popper has the answer.
Academy of Management Learning and Education, 6 (2):272-280.
Skinner, J. (2005). Brief report on some major companies’ perceptions of University of
KwaZulu-Natal graduates in the workplace. Internal report commissioned by DVC:
Access, September 2005, Durban.
Skinner, J. (2009). Educational innovation in accounting tutorials and language
development, School of Accounting, University of KwaZulu-Natal. Unpublished report
to SAICA.
Slonimsky, L. and Shalem, Y. (2006). Pedagogic responsiveness for academic depth.
Journal of Education, 40:35-50.
South African Institute of Chartered Accountants (SAICA). (2005). The education
requirements of the South African Institute of Chartered Accountants for entry into Part
1 of the qualifying examination. SAICA, Bruma Lake.
South African Institute of Chartered Accountants (SAICA). (2009). Examiners’
comments Part 2 of the Qualifying Examination – Financial Management, November,
2009.
https://www.saica.co.za/Portals/0/LearnersStudents/documents/FinMan_2009_Examine
rs_Comments.pdf , [Accessed 15 July 2010].
South African Institute of Chartered Accountants (SAICA). (2010a). Competency
framework: competencies of a chartered accountant (SA) at entry point to the
profession. SAICA, Bruma Lake.
South African Institute of Chartered Accountants (SAICA). (2010b). Qualification
model: foundational premises. Internal discussion document for meeting. SAICA,
Bruma Lake, 3 March.
South African Institute of Chartered Accountants (SAICA). (2011). From discussions in
SAICA Workgroups for the development of new Part 1 and Part 2 qualifying
examinations. SAICA, Bruma Lake, 21 January; 23 February; 23 March; 11 April; 13
May; 8 June.
34 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Sunderland, N. and Graham, P. (2001). The role of academic community in higher
learning: Alternatives to a drive-thru education. [Online]. Laboratory for Comparative
Human Cognition. Available from: http://lchc.ucsd. [Accessed 5 August 2005].
University of Cape Town (UCT). (2010).
http://www.uct.ac.za/downloads/uct.ac.za/about/policies/academicstaff.pdf [acessed 15
July 2010]
University of Cape Town (UCT). (1999). CHED AD Hominem criteria and ratings.
hr.uct.ac.za/humdev/docs/CHED_adhom_critletter.doc [accessed 15 July 2010].
Van der Schyf, D.B. (2008). The essence of a university and scholarly activity in
accounting, with reference to a department of accounting at a South African university.
Meditari Accountancy Research, 16(1): 1-26.
Van Fleet, D.D and Peterson, T.O. (2005). Increasing the value of teaching in the
academic marketplace: the creation of a peer-review infrastructure for teaching.
Academy of Management Learning and Education, 4(4):506-514.
Wassermann, S. (1993). Getting down to cases: learning to teach with case studies. New
York: Teachers College Press, Columbia University.
Watson, A. (2010). Comments made by Alex Watson to delegates on education
implication of adopting IFRS. Africa Standards Setting Conference, Cape Town, 6
October 2010.
Watson, A. (2011). Personal communication at SAICA offices, Bruma Lake, 11 April
2011.
Weil, S.W. and McGill, I. (eds). (1989). Making sense of experiential learning –
diversity in theory and practice. Buckingham: SRHE and Open University Press.
Wessels, P.L. and Steenkamp, L.P. (2009). Appropriate learning styles and teaching
approaches in the economic and management sciences faculty. South African Journal of
Education, 23 (5):1039-1058.
Whitcomb, M. E. (2000). Responsive curriculum reform: continuing challenges. In The
education of medical students: ten stories of curriculum change. Association of
American Medical Colleges / Milbank Memorial Fund, pp 26-40.
http://www.milbank.org/reports/americanmedical colleges/0010medical... [Accessed 23
December 2010]
Zhu, Y. (2009). Philosophical underpinnings for equilibrium of values and implications
for management education: A response to Harold Leavitt. Academy of Management
Learning and Education, 8(2):290-296.
SJ Kruger 35
Students’ and lecturers’ perceptions of the
effect of open-book examinations on the
learning behavior of accountancy students
SJ Kruger
Department of Accounting, Stellenbosch University
Received: November 2010 SAJAR
Revised: July 2011, September 2011 Vol 25 No. 1
Accepted: September 2011 2011
pp.35 to 57
In 2003 the South African Institute of Chartered Accountants (SAICA) changed its assessment
policy from a closed-book to an open-book format. One reason for this was to encourage better
learning behaviour amongst students. This paper investigates whether students perceive this aim
to have been achieved, by means of a survey conducted among final-year accounting students
who were preparing for the qualifying examination of SAICA. Assessment in the first three
years of study was generally done by means of closed-book examinations, but in their final year,
students were assessed by means of open-book tests and examinations (OBA). A comparison
was made between the views of these potential chartered accountants and those of lecturers in
accounting departments at selected universities.
Students were generally positive about the introduction of OBA, but lecturers were less
enthusiastic. Their conclusion was that OBA has generally been a step in the right direction to
preparing students better for the workplace. Cognisance should, however, be taken of the
negative study behaviour that was encountered. The paper also proposes strategies to improve
the effectiveness of OBA.
KEY WORDS
Open-book assessment; Closed-book assessment; Learning; Professional body; Knowledge
management; Bloom’s taxonomy; Accounting education
Contact
INTRODUCTION
Most academics are familiar with the expression “assessment drives learning”.
Numerous research studies have confirmed the fact that students change their method of
learning when the method of assessment changes. What students do in preparation for
examinations depends to a large extent on what assessment tasks require of them
(Biggs, 1999). In one study, amongst medical students, it was found that students
engaged more in clinical learning activities after the examination format was changed
from a theoretical to a clinical format (Newble & Jaeger, 1983:165-171). In another
study among nursing students, it was found that better formulated questions, used in
conjunction with case scenarios, enhanced critical thinking (Leung, Mok & Wong,
36 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
2008:711-719). The move by SAICA (the South African Institute of Chartered
Accountants) to change the qualifying examination from a closed-book to an open-book
examination has led to various debates amongst accounting academics, as well as
current members of SAICA. Most parties agree that the change in the method of
assessment has resulted in a change in learning. This has been confirmed in other
studies where open-book assessment (OBA) was introduced (Koutselini-Ioannidou,
1997; Boniface, 1985; Baillie & Toohey, 1997). Whether the change has been for the
better, however, is a point of contention.
BACKGROUND AND LITERATURE REVIEW
SAICA introduced an open book assessment policy for Part 1 of the South African
Qualifying Examination (QE 1) in March 2003. The IRBA (Independent Regulatory
Board for Auditors – previously the PAAB) followed suit in November 2004 for Part 2
of the Qualifying Examination, the Public Practice Examination (PPE). Since then, the
IRBA has followed the same policy as SAICA for its examinations, with the exception
of change in 2006, when the IRBA decided to adhere to its previous policy of not
allowing any notes to be made in allowed texts.
Prior to 2003, candidates were assessed by means of a closed-book examination. From
2003 to 2005, candidates were assessed by means of a partially open-book examination
where they were allowed to bring in certain texts, but were not allowed to add notes to
the texts apart from underlining, highlighting and making limited annotations. In 2006,
this policy was adapted by SAICA to allow students to make any notes they wished in
the allowed texts (SAICA, 2006). SAICA was not the first accounting institute to
introduce open-book assessment. The Institute of Chartered Accountants of England
and Wales first investigated open-book techniques for its examinations in 1988 and
implemented OBA soon thereafter. In an article commenting on the introduction of
OBA in England and Wales (Timms, 1990:21-22), it is mentioned that OBA is more
relevant to testing the skills required of a professional accountant and that the intent is
to reduce the reliance on rote learning. However, the fear that it would be abused and
that students would be discouraged from committing essential knowledge to memory
has also been mentioned. Amongst other professional accounting bodies that have
implemented OBA are those of the USA, Australia, New Zealand, Ireland and Scotland.
The IRBA motivates its change in assessment policy in a booklet (IRBA, 2004) in
which one of the specified aims of open-book assessment is to guide students in their
learning to a point where they are able to independently make the transition from data to
information to knowledge. This is also referred to as “knowledge management”; one of
the professional competencies1 which, the IRBA believes will be enhanced by adopting
an open-book assessment policy. This aim illustrates that cognisance is being taken by
the regulatory body of the rapid rate of knowledge expansion in the accounting field.
This ‘information overload’ is placing an ever-increasing strain on students and
lecturers alike; the IRBA argues that the sheer volume of information that professionals
are expected to use meaningfully requires a different set of skills that can be better
developed by exposing students to an OBA environment. This view is echoed in a study
among medical students (Broyles, Cyr & Korsen, 2005), which reported that medical
1According to the IRBA (2004:1), professional competence is ‘the ability to apply integrated
knowledge, skills and professional values appropriate to the practice of a Registered Accountant and
Auditor at entry level.’
SJ Kruger 37
professionals experience the same pressure of a significant increase in the volume of
available information. This provided the primary motivation for the present study,
namely, to explore the effect that OBA had on accounting students. The fact that
technological advances such as the internet, portable computers, personal digital
assistants and ‘smart phones’, inter alia, have made access to up-to-date information in
any location much easier than previously, also places a question mark over the
expectation that this information should be committed to memory.
This opinion is supported by Francis (1982:15), who makes the following comment:
“… the strength of open-book examination is that it allows candidates to be tested on
their ability to find and select knowledge and information to produce organised
responses and satisfactory solutions to problems under reasonably favourable
conditions.”
In a study on OBA in legal education, Maharg (1999) states that the skills-based
learning movement has been at the forefront of challenging the rationale behind
memorising great tracts of “black letter law” (1999:231). This movement emphasises
the skill components that underpin learning, instead of the weight of memorised “black
letter law”. It is learner-centred, rather than content (“corpus”)-centred, contending that
what legal professionals do with the legal information is more important than a
substantive knowledge of law. Although memory plays a role, it is not pre-eminent. In
conclusion, Maharg (1999) states that there are deeper issues underlying the debate
about assessment, namely, how the profession views what it does, how practitioners
should be prepared and appropriate models of teaching and learning. Various
researchers have conducted studies on the relationship between theory and practice and
Maharg (1999) argues that preferences of assessment models will be affected by
whichever model of this relationship is preferred. Propagators of the technical-
rationality model (which separates doctrine and theory), who apply this model in
practice, will prefer closed-, rather than open-book examinations and blank open-book,
rather than annotated open-book examinations. If Schön’s (1987) model of the
reflective practitioner is preferred, then assessment methods that place legal doctrine
and theory into context as much as possible, in what Vygotsky (1978) defined as “the
zone of proximal development”, will be preferred. The arguments of Maharg (1999) can
be put forward for the accounting profession in general.
The ‘skills-based learning movement’ of Maharg (1999) is better known in accounting
education as ‘competency-based’ education. According to Weil, Oyelere & Rainsbury
(2004), the required competencies for entry into the accounting profession have been
the topic of increasing debate since the late 1980s.
SAICA is currently in the process of establishing a new competency framework for
chartered accountants at the entry point of the profession. It is evident from the in-
progress documentation (SAICA, 2010) that the regulator is not only concerned with
the content students need to master, but also with the ‘soft’ skills they need to develop
in order to master the content. Soft skills encompass a broad range of qualities, defined
by Weber, Finely, Crawford & Rivera (2009:356) as “interpersonal, human, people or
behavioural skills needed to apply technical skills and knowledge in the workplace”. De
Villiers (2010) conducted a study on the incorporation of soft skills into accounting
curricula. He argues that:
38 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
“All stakeholders expect solid accounting skills and strong analytical skills, supported
by a set of soft interpersonal professional skills to enhance their employability and
effectiveness in the real workplace. In the increasingly technological economy, it simply
is not enough to deliver academically strong graduates.”(2010:9)
Various studies in accounting education, focusing on teaching and assessment
interventions, have attempted to develop some of these competencies. These studies
include:
Team work (Gammie, Gammie & Cargill, 2002; Stainbank, 2009),
Case studies (Weil, Oyelere, Yeoh & Firers, 2001; Weil et al., 2004;
Boyce et al., 2001),
Audit simulations (Arens, May & Dominiak, 1970; Steenkamp &
Rudman, 2007),
Role playing (Craig & Amernic, 1994), and
Writing skills’ development programmes (Craig & McKinney, 2009;
Ashbough, Johnstone & Warfield, 2002).
Most of these interventions aim to bridge the gap between theory and practice by
presenting students with problems that closely resemble ‘real-life’ scenarios, often in a
group setting. Some of the competencies which have been found to be developed by
these interventions include interpersonal skills, meeting management skills, writing and
oral communication skills and intellectual skills, which include locating and organising
information.
One of the soft skills widely accepted as being a requirement for professional
accountants is continuous learning ability. PAAB (2004) states that one of the expected
benefits of introducing OBA is that lifelong learning will be encouraged by students
taking more responsibility for their own learning and learning to manage their
knowledge more effectively. The importance of a lifelong learning culture is supported
by Kelly, Davey & Haigh (1999:325) who state:
“While graduates need to be able to contribute to the society which they join, another
perspective suggests they should emerge as independent learners, capable of
challenging and changing that society, and most importantly, to appreciate the totality
of relationships within society. Present students need to learn a new set of values, which
will allow them to encourage the development of a potentially totally different future
world, and learn how to change the thinking of their contemporaries.”
The regulators have long encouraged the setting of examination questions that resemble
– as closely as possible – problems that will be encountered in practice. By allowing
students to access textbooks during examinations, the regulators have gone a step
further by indirectly acknowledging that the skill of using available material is one of
the professional competencies that need to be assessed.
A number of studies have compared the performances of groups of students using open-
and closed-book tests. The results of these studies show that students' performances
SJ Kruger 39
remained unchanged when higher-order learning was tested, regardless of whether or
not students had access to texts (Koutselini-Ioannidou, 1997; Kalish, 1958; Michaels &
Kieren , 1973; Krarup, Naeraa & Olsen, 1974; Jehu, Picton & Futcher, 1970;
Betteridge, 1971; Shine, Kiravu & Astley, 2004). As was expected, students did better
in questions set at lower taxonomical levels when they had access to texts, compared to
a closed-book setting (Michaels & Kieren, 1973). Certain studies conclude that students
are less anxious when writing open-book tests (Michaels & Kieren, 1973; Jehu et al.,
1970; Betteridge, 1971; Feldhusen, 1961; Theophilides & Dionysiou, 1996;
Theophilides & Koutselini, 2000). Other researchers found that some students stated
that they put in less effort preparing for open-book than closed-book tests (Boniface,
1985; Feldhusen, 1961; Tanner, 1970) and that there was a negative correlation between
marks obtained and the number of texts consulted during examinations (Koutselini-
Ioannidou, 1997; Boniface, 1985; Francis, 1982).
Another area of concern which has surfaced is the practice of adding notes to allowed
texts. When only limited annotations were allowed2, invigilation became problematic,
as texts had to be inspected to ensure a level playing field for all students. The subjects
in this study were allowed to add unlimited notes to the allowed texts, which led to
concerns among lecturers that instead of mastering the content, time was wasted on
adding notes. This perception could conceivably have played a significant role in
SAICA’s decision to revert back to the previous policy of only allowing limited
references and annotations for years after 2007. Some researchers, however, are of the
opinion that note-taking and annotation of texts is an important part of the learning
process and should be encouraged (Maharg, 1999; Eilertsen & Valdermo, 2000).
O’ Flaherty, then the PAAB chief executive remarked after the 2004 PPE examinations
that OBA was implemented to place the “emphasis on application rather than rote
learning” (Green, 2005). This widely propounded benefit of OBA runs like a golden
thread through numerous research studies (Theophilides & Dionysiou, 1996:22;
Theophilides & Koutselini, 2006; Chan, 2003; Vanderburgh, 2005). In these studies,
‘deeper’ learning, or learning at higher taxonomical levels, is emphasised over
memorisation. The taxonomical levels refer to Bloom’s taxonomy, which describes the
levels of understanding, in ascending order, as being knowledge, comprehension,
application, analysis, synthesis and evaluation. When accounting lecturers in New
Zealand designed a course fostering lifelong learning and encouraging critical thinking,
the assessment method of choice was an open-book examination (Kelly et al., 1999).
Despite the fact that OBA has been used in accounting education internationally for
some time, there is no evidence of the effect that its introduction has had on the learning
behaviour of these students. This study aims to make a contribution in this regard.
RESEARCH METHODOLOGY
Background
When the regulators changed their assessment from closed- to open-book, it was done
with certain expectations in mind. These expectations included an overall enhancement
of the learning environment, improvement in continuous learning skills and the
2When OBA was first introduced from 2003 to 2005, students were allowed access to certain texts, but
were not allowed to add notes to the texts apart from underlining, highlighting and limited annotation.
40 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
provision of a better measure of professional knowledge and the application thereof
(IRBA, 2004). Other possible effects that OBA might have on learning behaviour, as
identified in the literature and background section of this study, are:
Earlier preparation,
The practice of adding notes to texts,
The volume of work mastered, and
An increased focus on the application of knowledge, rather than on the
memorisation of information.
Research objective
The primary objective of this research study is to gain a better understanding of
accounting students’ perceptions of how OBA has affected their learning behaviour,
especially with respect to the following:
Did the introduction of OBA lead students to increase their engagement in deeper
learning activities?
Has OBA assisted students to master more work than would otherwise have been
the case?
Did students waste time by unnecessarily adding notes to the allowed texts?
Did students take more responsibility for their learning?
Did the timing of students’ preparation for the examination change due to the
introduction of OBA?
Is OBA a better way to assess professional competency?
The sub-objectives of the study are to:
Measure lecturers’ perceptions of the issues identified and compare them with
those of the students.
Determine whether there are significant differences in perceptions between a
number of homogenous groups based on gender, first language, second language,
language of tuition and ethnicity.
Determine whether perceptions varied between students of different academic
ability (based on their performance in the previous year of study).
SJ Kruger 41
Questionnaire design
The questionnaire to be administered to the students was designed to be completed
anonymously in a classroom environment. A lecturer at each of the participating
universities was identified and requested to distribute the questionnaires to students to
complete during class time. This resulted in a very good response rate being achieved
(see section regarding subjects).
The first part of the questionnaire requested demographic information in respect of
gender, first language, second language, language of course presentation, language in
which exams were written and ethnicity. Students were also asked to indicate what their
marks were in the previous year of study. They also had to indicate how many times
they had written open-book tests or examinations during the year. To determine whether
students did change their approach to learning, they were asked the following question
for which they had to indicate their answer per subject:3
Does your approach or technique in studying for and preparing for examinations with
the view to writing an exam (based on an open-book policy) differ from the approach
you would have previously followed in preparing for a traditional examination? Use a
scale of 1 to 5, where 1 is “no change in approach” and 5 is “significant change in
approach”.
In addition, specific statements were added to evaluate the perceptions of students about
whether the expectations of the regulators had been met. These statements were
formulated as follows:
Using a scale of 1 to 5, indicate whether you agree with the following statements, where
1 represents “disagree totally” and 5 “totally agree”.
My approach to learning has improved due to open-book assessment.4
The open-book approach for assessment is a more reliable way of testing your
professional knowledge and the application thereof than a traditional
examination.5
The change to open-book assessment caused my lecturers to require me to take
responsibility for my own learning and manage my own knowledge effectively.6
3Students were enrolled in four subjects, namely financial accounting, auditing, taxation and
management accounting.
4The IRBA (2004) anticipated that the teaching/learning environment would be enhanced by
introducing OBA.
5This can be seen against the background of the argument that in practice the professional accountant
will have access to textbooks and OBA will resemble this scenario better than a closed-book
examination.
6This question applies to the competency of continuous learning.
42 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
When preparing for an examination, students spend less time on memorising
information, formulas etc. and more time working through integrated questions
and in-depth study than they would have for a closed-book exam.7
To gain a better understanding of what students’ perceptions are of the other issues
identified in the literature study and background section, the following statements were
added:8
Using a scale of 1 to 5, indicate whether you agree with the following statements, where
1 represents “disagree totally” and 5 “totally agree”.
“An open-book examination is better than a closed-book examination as I master
more work than I would have if it were a closed-book exam.”9
“Students could use their preparation time for examinations better by studying the
content instead of writing notes in the permissible texts.”
“I started preparing for the examination earlier by reading and placing tags on
books.”10
Apart from the questions on the change in approach to learning per subject and earlier
preparation (see table 3) the questionnaire administered to the lecturers was the same as
the one designed for the students and space was provided for comments.
SUBJECTS
Students who are studying to become chartered accountants are, for the purposes of this
study, regarded as being in training to become professional accountants. They have
already completed an accounting degree and are enrolled in a programme to obtain a
Certificate in Theory of Accounting (CTA), which is required by SAICA in order to
gain access to the qualifying examination.11
SAICA accredits certain institutions – most
of them full-time universities – to award this qualification.
7The IRBA (2004) expects students to prepare in this way. Studies on OBA in other disciplines have
confirmed a change in emphasis to understanding, rather than memorising (Theophilides & Dionysiou,
1996; Theophilides & Koutselini, 2006; Chan, 200; Vanderburgh, 2005). 8These statements are derived from questionnaires used in other research studies on OBA (Theophilides
& Dionysiou, 1996; Theophilides & Koutselini, 2000; Chan, 2003).
9One of the strongest arguments in favour of the implementation of OBA is that the volume of
knowledge that professionals are expected to have to perform their duties is expanding rapidly (IRBA,
2004; Breton, 1999:3).
10
In a closed-book assessment environment, students typically postpone their preparation to closer to
the examination time than in an OBA environment (Theophilides & Koutselini, 2000).
11
The qualifying examination is centrally administered by SAICA. All candidates write the same
examination, which covers the same topics as those presented in the CTA programmes of the accredited
institutions. The examination can be attempted for the first time in the first half of the year following
the year in which candidates obtain the CTA.
SJ Kruger 43
The student target group of the study is CTA students enrolled at full-time universities
in South Africa during 2006. Eight full-time universities participated in the study. Four
accredited full-time universities did not participate due to students being unavailable
because classes for the year have been completed12
, or simply because they did not
respond to the request to participate. Full-time universities were targeted because their
students had, at the time the questionnaires were administered, already been exposed to
open-book tests. Part-time (correspondence) universities typically do examination-type
assessment at the end of the year only.
Most of these students had completed more than half of their CTA course, but were
still to write their final CTA examinations. If they passed these examinations, they
would be allowed to write part one of the qualifying examination of SAICA during
March of the following year.
The lecturers surveyed were all lecturers at accounting departments of seven full-time
universities.
Demographic details
The number of questionnaires processed for the students was 1 065. A total of 1 490
students were enrolled at the different universities when the questionnaire was
administered, yielding a response rate of 71%. In total, 2 185 CTA students were
enrolled at full-time universities in South Africa, which means that 49% of all full-time
students enrolled for the CTA participated in the study. The lecturers employed at the
participating universities numbered 252. The number who responded was 59, yielding a
response rate of 23%. The majority of these lecturers were qualified chartered
accountants. The demographic details of the student respondents are presented in
Table 1.
Ninety two percent of the students who participated in the study indicated that, at the
time of completing the questionnaires, they had been exposed to open-book tests and
examinations on more than four occasions in their CTA year of study. The CTA
students of 2006 had, in general, also been assessed by means of closed-book
examinations in their undergraduate studies. The students, therefore, were in a position
to compare the two approaches.
Data analysis
The responses of students and lecturers were summarised per question by indicating the
percentage of respondents who chose 1-5 on the Likert scale. Answers of 1 or 2 were
further summarised to indicate disagreement, while answers of 4 and 5 were
summarised to indicate agreement, with 3 being regarded as a neutral response. These
results are shown in Table 2. For further analysis, the aggregate mean score and
standard deviation for each of the questions was computed for both students and
lecturers. When testing for significant differences where more than two groups were
applicable, such as for first language and ethnicity, a Kruskal-Wallis test was done. In
addition, a Spearman test was done to determine whether there was a correlation
between the ability of students – measured by their prior results - and their perceptions
12
Hard copies of the questionnaires were handed out by lecturers in the classrooms at the participating
Universities. At the Universities where classes were completed for the year this was not possible.
44 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
on the different questions. Only significant statistical differences within homogenous
groups were reported. The Man-Whitney U-test was used to determine any statistical
differences between the perceptions of lecturers and students. These results are
presented in Table 3.
Table 1: Student respondent demographics
Personal profile
Male
Female
Gender 45%
55%
Language Afrikaans English Other
First language 46% 38% 16%
Second language 27% 59% 14%
Language of course presentation 42% 58%
Language of writing examinations 40% 60%
Ethnic group Black Coloured White Indian
21% 4% 69% 6%
Marks achieved in final year of study
75-100% 60-74% 50-59% 0-49% N/A
Financial accounting 13% 44% 37% 6% 0%
Auditing 11% 46% 40% 3% 0%
Taxation 17% 48% 31% 4% 0%
Management accounting 20% 42% 34% 4% 0%
Commercial law 12% 19% 19% 1% 49%
Information systems 18% 20% 10% 0% 52%
Exposure to an open-book examination policy
1-4 >4
Amount of open-book examinations/tests written in the current academic year 8% 92%
SJ Kruger 45
RESULTS AND DISCUSSION
Table 2: Summary of answers
Disagree Neutral Agree
1 2 3 4 5
Approach to learning has improved
Students 9% 17% 28% 29% 17%
26% 28% 46%
Lecturers 17% 25% 31% 23% 4%
42% 31% 27%
Spent more time working through integrated questions and in-depth study
Students 4% 3% 11% 24% 58%
7% 11% 82%
Lecturers 13% 11% 14% 42% 20%
24% 14% 62%
Mastered more work
Students 9% 9% 22% 26% 34%
18% 22% 60%
Lecturers 15% 31% 23% 18% 13%
46% 23% 31%
Take responsibility for own learning and manage own knowledge effectively
Students 6% 11% 24% 35% 24%
17% 24% 59%
Lecturers 8% 8% 21% 44% 19%
16% 21% 63%
Started preparing earlier Students 11% 12% 22% 25% 30%
23% 22% 55%
OBA is a more reliable way of testing your professional knowledge and the application thereof
Students 6% 8% 23% 28% 35%
14% 23% 63%
Lecturers 9% 24% 23% 31% 13%
33% 23% 44%
Wasted time adding notes
Students 21% 26% 24% 12% 17%
47% 24% 29%
Lecturers 7% 4% 9% 40% 40%
11% 9% 80%
Table 3: Comparison between students’ (Stud) and lecturers’ (Lect) perceptions
Stud
mean
(S.D.) Lect
mean
(S.D.) p-value
Approach to learning changed for:
1 Taxation 3.54 (1.28)
2 Auditing 3.34 (1.35)
3 Financial accounting 3.27 (1.31)
4 Management accounting 2.66 (1.49)
5 Approach to learning improved 3.34 (1.18) 2.71 (1.13) <0.01
6 Spent more time working through integrated
questions and in-depth study
4.26 (1.09) 3.45 (1.29) <0.01
7 Mastered more work 3.66 (1.30) 2.84 (1.26) <0.01
8 Take responsibility for own learning and manage
own knowledge effectively
3.66 (1.14) 3.60 (1.12) 0.70
9 Started preparing earlier 3.54 (1.32)
10 More reliable way of testing professional
knowledge and application
3.81 (1.21) 3.15 (1.19) <0.01
11 Wasted time adding notes 2.78 (1.36) 4.02 (1.15) <0.01
General change in approach to learning
Students indicated that, with the exception of management accounting (a mean of 2.66),
their approach to learning changed. Their approach to learning changed the most for
taxation (a mean of 3.54), while it changed the least for financial accounting (a mean of
3.27). During the course of analysing student responses, it became apparent that for
46 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
management accounting, some universities did not allow access to texts.13
This partly
explains the reason for the low mean score for management accounting. Another
explanation for the difference in the perceived change in approach might be found in
how much memorising had to be done in preparing for the subjects if they were to be
assessed without allowing access to texts. Taxation is arguably more rule-driven,
compared to financial and management accounting, which are more principle-driven
and therefore require more memorisation.
It is evident that students and lecturers have differing opinions about whether students’
approaches to learning have improved due to open-book assessment. Most students
(46%) agree, while most lecturers (42%) disagree. The mean for students was 3.34,
compared to a mean of 2.71 for lecturers. With a p-value smaller than 0.01, the
difference is statistically significant. Almost a third of both groups had a neutral stance.
The reasons for lecturers’ scepticism are explored in other sections of this paper.
Effect on learning activity
It is evident that students agree strongly with the fact that less rote learning takes place
in the case of OBA and that they spend more time performing learning activities at
higher taxonomical levels, rather than only on knowledge acquisition. Most lecturers
also agree with this statement, although considerably fewer than the students (62%
versus 82%). Only 7% of the students disagreed, compared to 24% of the lecturers.
This difference is also statistically significant (a mean of 4.26, versus a mean of 3.45,
with p<0.01).
Other studies have found that, in general, OBA supports a change to learning at higher
taxonomical levels (Broyles et al., 2005; Theophilides & Dionysiou, 1996;
Theophilides & Koutselini, 2000; Chan, 2003; Vanderburgh, 2005). This study
confirmed these findings, with the majority of students (82%) and lecturers (62%)
agreeing that less memorising was done by students and that their focus had shifted
towards integrating knowledge and in-depth study. Most lecturers will welcome this
result, as final-year students are often advised to spend more time working through
problems, preferably in a simulated test environment, than on memorising content.
There is some research evidence that testing information (retrieval) leads to better long-
term retention than repeated reading of material (Karpicke & Roediger, 2007), although
students are generally not aware of this benefit (Karpicke, Butler & Roediger, 2009).
This benefit of retrieval is referred to as the ‘testing effect’; there is evidence that it is
the same for open-book and closed-book tests (Agarwal, Karpicke, Kang, Roediger &
Mcdermott, 2008). It can therefore be argued that if students engage in more self-testing
exercises in an open-book environment than in a closed-book environment, it should
lead to overall better long-term retention.
Volume of work mastered
The difference in perception between lecturers and students is clear, with 60% of
students agreeing (a mean of 3.66), compared to 31% of lecturers (a mean of 2.84). The
p-value of less than 0.01 shows that the difference is statistically significant. This
13
None of the allowed texts applies specifically to management accounting, but students could add
notes applicable to management accounting in the allowed texts.
SJ Kruger 47
confirms the perception of lecturers that students’ approaches to learning have not
improved due to open-book assessment. The following comment made by a lecturer
illustrates this scepticism:
“In my opinion open-book has become a crutch, and students prepare less than five
years ago.”
This question should be seen in the light of one of the expected benefits of OBA,
namely, that it will assist students to master more concepts because they do not have to
spend as much time memorising information. It is also common practice in a closed-
book environment for lecturers to emphasise certain sections of work above others, or
even to exclude parts of the syllabus before examinations, because of an awareness of
the large amount of work students need to know. Without the pressure to memorise vast
tracts of information, students can be tested on the whole syllabus.
Enhancement of student’s responsibility for own learning
Most students and lecturers agree that OBA enhanced students’ responsibility for their
own learning (59% of students and 63% of lecturers) with mean scores of 3.66 and 3.60,
respectively. The p-value of 0.70 also shows that there was no statistically significant
difference between their perceptions. Professional accountants perform their profession
within a regulatory framework which changes constantly. The knowledge on which
students are tested in the qualifying examination soon becomes outdated. Being aware
of what changes have taken place and being able to independently assimilate new
information are crucial skills of a competent professional. It does seem that the
introduction of OBA has had a positive impact in this regard.
Change in timing of study effort
Fifty five percent of students indicated a 4 or 5 on the 5-point scale (with a mean of
3.54), demonstrating that most students did start their preparation earlier. This should be
viewed in a positive light, as it seems that when students are not required to memorise
information, they are more likely to work more consistently throughout the academic
year. This finding is consistent with findings of the Theophilides & Koutselini (2000)
study on OBA.
Appropriateness for assessing professional competence
One of the main reasons for the adoption of OBA is to assess students in a way that
matches a ‘real life’ situation as closely as possible. It is argued that in real life the
professional will have access to information when performing his/her duties (Maharg,
1999; Shine et al., 2004; Gray, 1994). Other disciplines also debate ways to reduce this
gap between theory and practice, as can be seen in research done in the training of legal
professionals (Maharg, 1999). SAICA is currently in the process of establishing a new
competency framework for chartered accountants at the entry point of the profession
and it is evident from the in-progress documentation (SAICA, 2010) that in the view
of the regulator, bringing theory and practice closer together is one of the fundamental
building blocks for an effective teaching strategy. The work of Dewey (1915) is referred
to in the document; he argues that knowledge (theory) is learned in the course of
‘focused practical experimentation’. There is a strong argument that if students are
48 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
allowed to use the ‘tools of their trade’ (i.e. textbooks and other material), it will assist
in bringing theory closer to practice.
The majority of lecturers and students agree that OBA is superior to traditional
examinations (closed-book) for testing professional14
knowledge and the application
thereof. Lecturers are, however, less positive, with 33% disagreeing, compared to only
14% of students disagreeing. The mean scores (3.81 versus 3.15) and the p-value of less
than 0.01 show that the difference is statistically significant.
Time spent making notes
More students (47%) disagreed with the statement (“Wasted time making notes”) than
those who agreed (29%), which indicates that they found the effort of making notes
worthwhile. The overwhelming majority of lecturers (80%), on the other hand, felt that
students wasted time making notes and would have been better served studying content.
The statistically significant mean scores (2.78 versus 4.02) and the p-value of less than
0.01 confirm that lecturers believe that students engage in the ‘shallow’ activity of
making notes, rather than making an effort to understand the content. Almost a third of
students (29%) did indicate that they agreed with the statement, acknowledging that
time was indeed wasted by adding notes, instead of studying the content. This might be
because students lack the necessary experience of studying in an open-book
environment; they rely on notes that will be useless without a deeper understanding of
how to apply the knowledge. If this “preparation” leads to, or is the result of, good note
taking or summarising skills, it should not necessarily be seen in a bad light. The
process of deciding which notes to add and designing the format of the notes can
involve thinking at higher taxonomical levels. It can also be compared to a tradesman
sharpening his/her tools. Textbooks may be given more meaning and made more
effective when notes are added. Maharg (1999) was critical of the regulatory body
which allowed only blank15
textbooks in an open-book examination for legal
professionals, claiming that candidates were not allowed to use the same ‘tools’ in the
examination as they would when practising law. Candidates in Maharg’s (1999) study
also indicated that the ban on notes increased anxiety. The major potential pitfall is that
students unthinkingly add notes, giving them a false sense of security when preparing
for examinations.
Further statistical analysis
Student performance
Due to the anonymity of the survey and the fact that the students’ final marks for the
year were not available at the time of the survey, the correlation between the actual
performance of the students in an OBA setting and their statements in the questionnaire
could not be tested. An attempt was made to differentiate between students with
14
‘Professional’, as opposed to ‘academic’, knowledge has the added dimension of abilities relating to
‘application of integrated knowledge, skills and professional values appropriate to the practice of a
Registered Accountant and Auditor at entry level’ (IRBA, 2004).
15
‘Blank’ in this context means that students were not allowed to add any notes to the allowed
textbooks.
SJ Kruger 49
differing ability16
by calculating an average mark based on their performance in the
previous year of study. This was done by taking the midpoint of the range of marks
attained, as indicated by students on the questionnaire. A Spearman test was done,
which showed no statistical correlation between their ability and their perceptions of
OBA. Kalish (1958) also found that ratings by students of the benefit of open-book
exams were not related to marks obtained in exams. In another OBA study, the expected
graduation grades of students did not have a significant influence on their perception of
the functions of OBA (Theophilides & Dionysiou, 1996).
Gender
The only statistically significant difference in the responses of male and female students
related to whether they started their preparation earlier. The mean of the male students
was 3.41 compared to 3.65 for the female students, with a p-value of less than 0.01.
Although female students may well have responded better to OBA than male students
as far as earlier preparation was concerned, other studies have found that, in general,
there are not significant gender differences in students’ approaches to learning
(Ballantine, Duff & Mc Court Larres, 2008). In certain studies, however, male and
female students ranked the benefits of case studies in developing core competencies
differently (Weil et al., 2004; Stainbank, 2010).
Language
Some South African studies have found that, as far as perceptions on learning and
academic performance are concerned, there might be differences between students
based on their first language (Stainbank, 2010; Baard, Steenkamp, Frick & Kidd, 2010).
In a study amongst black trainee accountants, 70% indicated that they did not regard the
fact that they could not write examinations in their first language (which usually is not
Afrikaans or English) as a disadvantage (Wiese, 2006).
In this study, 46% of students spoke Afrikaans as their first language, 38% English and
16% spoke other languages. At the time of the study, courses were only presented in
Afrikaans or English. Forty two percent of the respondents received tuition in
Afrikaans, while 58% were taught in English. The fact that most of the allowed texts
were only available in English, might have further contributed to differences in
perceptions between Afrikaans-speaking and English-speaking students on OBA. The
significant statistical differences based on first language and languages of tuition are
presented in Tables 4 and 5, respectively.
Afrikaans- (a mean of 3.22) and English-speaking (a mean of 3.54) students differed
significantly in their perceptions about a change in approach in learning towards
auditing. Afrikaans-speaking students (a mean of 3.81) differed significantly from
students with “other” first languages (a mean of 3.39) as far as mastery of more work
was concerned. With respect to spending more time working through integrated
questions and in-depth study, English students (a mean of 4.41) showed significantly
more agreement than Afrikaans students (a mean of 4.18). The same trend emerged for
language of course presentation, when the three groups were reduced to two and the
differences reported for tuition received in Afrikaans and English. This was mainly due
16
‘Ability’ in this case was based on academic performance.
50 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
to the fact that 97% of students with a first language other than Afrikaans or English
received tuition in English.
Table 4: Responses of students according to first language
Afr (S.D.) Eng (S.D.) Other (S.D.)
mean
mean
mean
Approach to learning changed for auditing 3.22a (1.35) 3.54b (1.31) 3.51ab (1.41)
Mastered more work 3.81a (1.24) 3.60ab (1.29) 3.39b (1.36)
Spent more time working through integrated questions and in-depth study
4.18a
(1.11)
4.41b
(0.97)
4.23ab
(1.11)
Means containing the same superscript (a or b) are not significantly different.
Table 5: Responses of students according to language of course presentation
Afrikaans (S.D.) English (S.D.)
P
mean
mean
Approach to learning changed for auditing 3.12 (1.35) 3.58 (1.33) <0.01
Mastered more work 3.77 (1.23) 3.58 (1.32) 0.03
Spent more time working through integrated
questions and in-depth study
4.18
(1.12)
4.33
(1.02)
0.04
When considering the statistically significant differences relating to first language and
language of tuition, it is found that the different groups all agreed, with only the level of
agreement differing. There was also no specific trend, with no grouping being more
positive about OBA in general. Further research about the effect that language might
have on perceptions of students on OBA is required to provide more meaningful
information.
Ethnicity
Due to the political history of South Africa, there are significant cultural and socio-
economic differences between ethnic groups in South Africa (Wiese, 2006). These
differences might have an impact on how the different ethnic groups perceive their
learning environment. The statistically significant differences found between ethnic
groups are presented in Table 6.
SJ Kruger 51
Table 6: Perceptions of different ethnic groups
Black Coloured White Indian
Approach to learning changed for:
Auditing 3.69a 3.44ab 3.29b 3.46ab
Taxation 3.27a 3.24ab 3.59b 3.35ab
Approach to learning has improved 3.05a 3.15ab 3.37b 3.40ab
Mastered more work 3.30a 3.68ab 3.79b 3.57ab
Means containing the same superscript (a or b) are not significantly different.
With respect to the change in approach to learning for taxation, improvement in
approach to learning and mastering more work, white students showed significantly
stronger agreement than black students. On the change in approach to learning for
auditing, black students, however, showed significantly stronger agreement than white
students. The means of Indian and coloured students were not significantly different
from each other or from those of black and white students.
As in the case of language, when the statistically significant differences between ethnic
groups are considered, it is difficult to reach any meaningful conclusions. The different
groups all agreed; the level of agreement just differed. White students may possibly be
more positive than black students, in general, about the effect of OBA. Further research
on the effect that ethnic background might have on students’ perceptions of OBA might
provide more meaningful information.
Summary
Most students will, when they are initially offered the choice between a closed-book
and an open-book examination, choose the latter (Croasdale, 1973). Students prefer to
have as many aids as possible available in examinations and having access to textbooks
and other material clearly constitutes such an aid. One could argue that this explains
students’ positive attitude towards OBA. This is, however, not necessarily the case. In a
study involving chemistry students, it was found that more students preferred a
traditional closed-book exam to an open-book exam (Betteridge, 1971). Eilertson and
Valdermo (2000) found that the culture of rote learning caused difficulty for many
students to adapt to OBA, some even preferring the traditional way of assessment that
mostly requires the reproduction of information as it appears in textbooks or handouts,
even if it meant that they would not have access to the textbooks.
One reason that might explain the more positive approach of students to OBA is that
they realise that they are being assessed in a way that is more representative of what
they will encounter in the workplace. Reed (1986) mentions that one of the problems of
traditional (closed-book) examinations is that students question the relevance thereof to
their future careers.
52 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Negative learning behaviour
It could be argued that, in addition to the perceived benefits, negative learning
behaviour has also resulted from the introduction of OBA; this is evident from the
scepticism shown in lecturers’ responses with respect to the improvement of students’
approaches to studying, the volume of work mastered and the practice of adding notes.
This is illustrated by some of the lecturers’ remarks, for example:
“I am really convinced that open-book assessment has resulted in candidates no longer
thinking about many issues. Total reliance is placed on the book's content – they often
are unable to apply the content in the books allowed during assessments.”
“Students have a false sense of security when in an open-book examination and think
they do not have to study AT ALL.”
The fact that OBA might have a negative effect on the learning behaviours of students
has also been found in other studies (Jehu et al., 1970; Feldhusen , 1961; Tanner, 1970;
Moore & Jensen, 2007). This is mainly attributable to an over-reliance on texts, which
manifests itself in time wasted by trying to look up answers and in students being
under-prepared for examinations. When evaluating the findings of other studies, it is
important that the contexts of the studies are taken into account. If students are exposed
to OBA over the short-term and examination questions are set at lower taxonomical
levels, many students will prepare less than they would have for a closed-book exam.
However, when examinations test students at higher taxonomical levels over the longer
term, students will change their study behaviour to a deeper approach (Eilertson &
Valdermo, 2000). Students generally do what is necessary to pass a course and the type
of questions that they expect in examinations is paramount in this regard (Reed, 1986).
When OBA is introduced, lecturers need to invest special effort to prepare students and
to warn them of the pitfalls of not preparing properly (Theophilides & Dionysiou, 1996;
Chan, 2003).
It is worth considering whether there has perhaps been an overreaction amongst
lecturers about the extent of the reported negative outcomes of the introduction of OBA.
In some studies, these negative behaviours were more pronounced among weaker
students (Koutselini-Ioannidou, 1997), who would conceivably have performed badly,
regardless of the format of the exam. In a study with psychology students, the relative
performance of students did not change when assessment methods were changed
(Miller, Bradbury & Lemmon, 2000). In this study, a reasonable proportion of students
(29%) acknowledged that they should have spent more time on studying content,
instead of adding notes to textbooks. The majority of students (47%), however, felt that
the process of making notes was worthwhile. As the responses in this study were
anonymous, a link between negative behaviour and performance could not be tested.
In a study by Baillie and Toohey (1997), lecturers changed their teaching approach to
focus on deeper learning, while they still assessed students by means of a traditional
closed-book examination. After analysis of the results, they concluded that the change
in teaching approach did not achieve the desired shift to deeper learning. In the
following year the method of assessment was changed to OBA, after which the
researchers found that students did change their learning behaviour to a deeper
approach. The conclusion to be drawn from this is that for a change to deeper learning
SJ Kruger 53
to take place, not only tuition methods need to be altered, but the assessment method as
well (Baillie & Toohey, 1997).
CONCLUSION
The aim of this study is to determine the perceptions of students and lecturers of the
effect of the introduction of OBA on the learning behaviour of professional accounting
students. A survey was conducted that showed that, as in other studies, there has been a
definite perceived effect. Students and lecturers were of the opinion that students do less
memorising and spend more time working through integrated questions and in-depth
study as a consequence of OBA. It was also found that OBA encouraged students to
take more responsibility for their own learning and contributed to students preparing
earlier for examinations. As far as the volume of work mastered and the practice of
adding notes to texts were concerned, students were positive and lecturers negative
about the effect of OBA. These different perceptions are topics for future research.
As in other studies, negative learning behaviour was also encountered; this consisted
mainly of an over-reliance on texts and time wasted by adding notes to texts, instead of
studying the content.
The results of this study provide a strong argument in favour of the introduction of
OBA by education regulators. Seen within the context of the broader competency
framework and curricula becoming loaded with an increasing volume of information,
removing the pressure to memorise information can assist in changing the learning
behaviour of students to a deeper approach. OBA also encourages the development of
knowledge management skills required for accountants to be effective lifelong learners.
Cognisance, however, needs to be taken of the negative learning behaviour that could
result from the introduction of OBA. Overall, however, it seems that the perceived
advantages of OBA outweigh the disadvantages and that it is a step in the right direction
towards improving the learning behaviour of accountancy students.
REFERENCES Agarwal, P.K., Karpicke, J.D., Kang, S.H.K., Roediger III, H.L. and McDermott, K.B.
(2008). Examining the testing effect with open- and closed-book tests, Applied
Cognitive Psychology, 22(7):861-876.
Arens, A.A., May, R.G. NS Dominiak, G. (1970). A simulated case for audit education,
The Accounting Review, 45(3):573-578.
Ashbough, H., Johnstone, K.M. ns Warfield ,T.D. (2002). Outcome Assessment of a
Writing-Skill Improvement Initiative: Results and Methodological Implications, Issues
in Accounting Education, 17(2):123-148.
Baard, R.S., Steenkamp, L.P., Frick, B.L. NS Kidd, M. (2010). Factors influencing
success in first-year Accounting, SA Journal of Accounting Research, 24(1):129-147.
Baillie, C. ns Toohey, S. (1997). The 'Power Test': its impact on student learning in a
materials science course for engineering students, Assessment & Evaluation in Higher
Education, 22(1):33-48.
54 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Ballantine, J.A., Duff, A. NS Mc Court Larres, P. (2008). Accounting and business
students’ approaches to learning: A longitudinal study, Journal of accounting
education, 26:188-201.
Betteridge, D. (1971). Open-book examinations, Education in Chemistry, 8(2):68-69.
Biggs, J. (1999). What the Student Does: teaching for enhanced learning, Higher
Education Research and Development, 18(1):57-75.
Boniface, D. (1985). Candidates' use of notes and textbooks during an open-book
examination, Educational Research, 27(3):201-209.
Boyce, G., Williams, S., Kelly, A. and Yee, H. (2001). Fostering deep and elaborative
learning and generic (soft) skill development: the strategic use of case studies in
accounting education. Accounting Education, 10(1):37-60.
Breton, G. (1999). Some empirical evidence on the superiority of the problem-based
learning (PBL) method, Accounting Education, 8(1):1-12.
Broyles, I.L., Cyr, P.R. and Korsen, N. (2005). Open book tests: assessment of
academic learning in clerkships, Medical Teacher, 27(5):456-462.
Chan, M. (2003). An open and shut case? Taking books into examinations - a Hong
Kong study, ASEE/WFEO International Colloquium, American Society for Engineering
Education.
Craig, R. and Amernic, J. (1994). Roleplaying in a conflict resolution setting:
description and some implications for accounting, Issues in Accounting Education,
9(1):28.
Craig, R. and McKinney, C.N. (2009). A successful competency-based writing skills
development programme: Results of an experiment, Accounting Education, 19(3):257-
278.
Croasdale, R. (1973). Open-book examinations in operational research (A personal
view), International Journal of Mathematical Education in Science and Technology,
4(2):133-135.
De Villiers, R. (2010). The incorporation of soft skills into accounting curricula:
preparing accounting graduates for their unpredictable futures, Meditari Accountancy
Research, 18(2): 1-22.
Dewey, J. (1915). The school and society. Rev. ed, Chicago: University of Chicago
Press, 10.
Eilertsen, T. V. and Valdermo, O. (2000). Open-Book Assessment: A Contribution to
Improved Learning? Studies in Educational Evaluation, 26: 91.
Feldhusen, J.F. (1961). An evaluation of college students' reactions to open-book
examinations, Educational and Psychological Measurement, 21:637-646.
SJ Kruger 55
Francis, J. (1982). A case for open-book examinations, Educational Review, 34(1):13-
26.
Gammie, B., Gammie, E. and Cargill, E. (2002). Personal skills development in the
accounting curriculum. Accounting Education, 11(1):63-78.
Gray, T.G.F. (1994). Open book examinations, Biochemical Education, 22 (3):122-125.
IRBA (2004). Open-book Assessment and the Public Practice Examination (PPE).
Jehu, D., Picton, C.J. and Futcher, S. (1970). The use of notes in examinations, British
Journal of Educational Psychology, 40:335-337.
Kalish, R.A. (1958). An experimental evaluation of the open book examination, Journal
of Educational Psychology, 40:335-337.
Karpicke, J.D. and Roediger III, H.L. (2007). Repeated retrieval during learning is the
key to long-term retention, Journal of Memory & Language, 57(2):151-162.
Karpicke, J.D., Butler, A.C. and Roediger III, H.L. (2009). Metacognitive strategies in
student learning: Do students practise retrieval when they study on their own? Memory,
17(4):471-479.
Kelly, M., Davey, H. & Haigh, N. (1999). Contemporary accounting education and
society, Accounting Education, 8(4):321-340.
Koutselini-Ioannidou, M. (1997). Testing and life-long learning: Open-book and
closed-book examination in a university course, Studies In Educational Evaluation,
23(2):131-139.
Krarup,N., Naeraa,N. and Olsen,C. (1974). Open-book tests in a University course,
Higher Education, 3:157-164.
Leung, S.F., Mok, E. and Wong, D. (2008). The impact of assessment methods on the
learning of nursing students, Nurse Education Today, 28(6):711-719.
Maharg, P. (1999). The culture of Mnemosyne: open-book assessment and the theory
and practice of legal education, International Journal of the Legal Profession, 6(2):219.
Michaels, S. and Kieren, T.R. (1973). An Investigation of Open-Book and Closed-Book
Examinations in Mathematics, Alberta Journal of Educational Research.
Miller, R., Bradbury, J. and Lemmon, G. (2000). Justifying means with ends:
assessment and academic performance, South African Journal of Higher Education,
14(1):166-173.
Moore, R. & Jensen, P.A. (2007). Do open-book examinations impede long-term
learning in introductory biology courses? Journal of College Science Teaching,
36(7):46-49.
56 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Newble, D.I. and Jaeger, K. (1983). The effect of assessments and examinations on the
learning of medical students, Medical Education, 17(3):165-171.
Reed, F.B. (1986). The case for open-book examinations, Biochemical Society
transactions, 14(5):854-856.
SAICA (2010). Competency Framework: Detailed Guidance for Academic
Programmes.
Shine, S., Kiravu, C. and Astley, J. (2004). In defence of open-book engineering degree
examinations, International Journal of Mechanical Engineering Education, 32(3):197-
211.
Stainbank, L.J. (2009). Working in teams: improving the team experience. Meditari
Accountancy Research, 17(1):69-80.
Stainbank, L.J. (2010). Students’ perceptions of the usefulness of an accounting project
in acquiring accounting knowledge and professional skills. SA Journal of Accounting
Research, 24(1):79-100.
Steenkamp, L. and Rudman, R. (2007). South African students’ perceptions of the
usefulness of an audit simulation. Meditari Accountancy Research, 15(2):23-41.
Tanner, W.F. (1970). Performance in open-book tests, Journal of Geological
Education, 18:166-167.
Theophilides, C. and Dionysiou, O. (1996). The Major Functions of the Open-Book
Examination at the University Level: A Factor Analytic Study, Studies in Educational
Evaluation, 22(2):157.
Theophilides, C. and Koutselini, M. (2000). Study behavior in the closed-book and the
open-book examination: A comparative analysis. Educational Research & Evaluation,
6(4):379.
Timms, C. (1990). Towards an open book policy. Accountancy, 105(1161):21-22.
Vanderburgh, P.M. (2005). Open-book tests and student-authored exam questions as
useful tools to increase critical thinking. Advances in Physiology Education, 29(1-
4):183-184.
Wiese, A. (2006). Transformation in the South African chartered accountancy
profession since 2001: a study of the progress and the obstacles black trainee
accountants still encounter. Meditari Accountancy Research, 14(2): 151-167.
Weber, M.R., Finely, D.A., Crawford, A. and Rivera, D.J. (2009). An exploratory study
identifying soft skill competencies in entry-level managers. Tourism and Hospitality
Research, 9(4):353-361.
Weil, S., Oyelere, P., Yeoh, J. and Firers, C. (2001). A study of students' perceptions of
the usefulness of case studies for the development of finance and accounting-related
skills and knowledge. Accounting Education, 10(2):123-146.
SJ Kruger 57
Weil, S., Oyelere, P. and Rainsbury, E. (2004). The usefulness of case studies in
developing core competencies in a professional accounting programme: a New Zealand
study. Accounting Education, 13(2):139-169.
G Sher & GDI Barr 59
Portfolio rebalancing in South Africa
G Sher
Department of Actuarial Science, University of Cape Town
GDI Barr
Department of Economics and Statistical Sciences, University of Cape Town
Received: August, 2010 SAJAR
Revised: December 2010 Vol 25 No. 1
Accepted: March, 2011 2011
pp.59 to 80
The theory of so-called ‘optimal’ portfolio rebalancing has emerged over the last decade in
journals and working papers, but this theory has several drawbacks, being based on advanced
mathematics and relying on sets of restrictive assumptions. This paper investigates rebalancing
strategies by using resampling procedures from actual South African data in a way that captures
the intrinsic high volatility nature of South African asset price movements. It considers the main
consequences of calendar and range rebalancing strategies on tracking error, transaction costs
and portfolio performance and demonstrates that range rebalancing has advantages over
calendar rebalancing in the South African financial context using the comparative portfolio
allocations of Firer, Peagram and Brunyee (2003). As such it provides a practical framework for
South African portfolio managers to make informed choices on the appropriate approach for
best-practice portfolio rebalancing.
Keywords
Portfolio Rebalancing; Range Rebalancing; Calendar Rebalancing; Volatility; South Africa
Contact
INTRODUCTION
Over time, asset allocations within a portfolio will deviate from the benchmark
allocation due to the relative performance of assets, changes in risk aversion, and cash
injections or withdrawals. The persistence of such forces causes the portfolio to become
severely overweighted in some assets and underweighted in others. For an institution,
where the benchmark allocation is presumably chosen to match the liability stream as
closely as possible, this means being exposed to a mismatch risk between the
benchmark and actual portfolios. Portfolio rebalancing is the process of reducing this
mismatch risk, through purchases and sales of assets in the portfolio that move weights
closer to the benchmark allocation. A portfolio rebalancing strategy is a rule that is
applied to manage this mismatch risk over time.
In the absence of transaction costs, the above mismatch risk can be made negligible by
choosing an arbitrarily small rebalancing interval, say sixty seconds. However,
60 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
transaction costs quickly create an optimisation problem. The decision of whether and
how much to rebalance at a point in time becomes a trade-off between incurred
transaction costs, affected risk reduction and expected future transaction costs and risk
reduction.
Purpose of investigation
In this paper we investigate the ability of simple rebalancing strategies to control a
portfolio’s tracking error in the relatively high volatility South African market, and we
investigate the effects of portfolio rebalancing strategy on transaction costs and risk-
adjusted returns. We demonstrate that the rebalancing strategy often employed to trade
at regular intervals is inferior to a simple range-based approach.
Method of investigation and assumptions
The choice of method for simulating asset returns is always controversial. In this paper,
we have used a bootstrap resampling technique for generating returns, rather than
assuming some statistical distribution of returns. The bootstrap resampling technique
has several advantages over known statistical distributions, some of which are presented
below. Sampling is based on daily data from the sample period 4 Jan 1999 – 16 Feb
2009 following the introduction of the Euro and only considering days where the South
African market traded.
We use five portfolios in this paper, to match as closely as possible the portfolios used
by Firer, Peagam and Brunyee (2003:19). These portfolios are listed in Table 1 below,
as well as the assumed costs (in basis points) of buying and selling each asset. Portfolio
B is a typical allocation benchmark for a pension fund. The cost estimates were
obtained by speaking to traders in large South African asset management firms and are
intended to be broadly representative of an institutional portfolio. Such portfolios will
be relatively large, but not too large that they incur price impact costs, which we have
assumed away in our analysis. However, every portfolio manager will face unique costs
and what really matters for the applicability of our analysis are the relative, rather than
absolute, transaction costs of each asset. We explain this further in the results section
below.
Table 1. Allocations for portfolios A - E, notional costs of trading asset classes and
the indices used as proxies for each asset class
Portfolio allocations (%) Costs (bps) Proxy index
A B C D E Buy Sell
SA Equity 90 60 50 40 20 65 40 J203T
SA Bonds 8 13 15 15 30 5 3 ALBI
SA Property 0 5 5 0 0 65 40 J873T
Foreign Equity 0 7 10 5 0 56 26 MSCI World
Foreign Bonds 0 10 15 10 0 65 40 Barclays World Bond Index
SA Cash 2 5 5 30 50 5 3 SA bankers’ 90-day acceptance rate
The assumptions made in this paper are very general, but they have some important
implications. Firstly, transaction costs may not be proportional to value traded in reality
and such price impact costs would deter larger and more frequent rebalancing trades.
Secondly, taxes are set to zero and do not differentiate between income and capital
G Sher & GDI Barr 61
gains. Differential rates of taxation should affect the rebalancing decision by attempting
to defer capital gains tax.
The bootstrap resampling method for generating asset returns is critical to the results
obtained and departs from other parametric approaches. Consequently, asset returns are
based on the sample period. There is little reason to believe that past returns are a
reliable predictor of future returns, especially in terms of market crashes. However, the
bootstrap procedure is scientifically appealing because it does not rely on a subjective
specification of asset price returns.
The principle disadvantage of straightforward bootstrap resampling as applied in Clark
(1999) is that successive days’ returns are treated as being independent. Share returns
are known to exhibit autocorrelated volatility (so-called ARCH terms, see Engle
(1982)), so that during a crisis period, say, share prices will exhibit periods of higher-
than-average volatility. This feature tends to be relatively enhanced in emerging
markets like the South African market. It is therefore necessary to modify the bootstrap
procedure so that sampled share returns exhibit a true-to-sample ARCH effect.
Structure of the paper
In section 2 below we describe the various types of rebalancing considered in this
paper. In section 3 we offer an overview of the literature on so-called ‘optimal’
portfolio rebalancing to set the context. In sections 4 and 5 we implement a
straightforward bootstrap resampling procedure on South African data to provide results
that will help the practitioner set the parameters of a rebalancing strategy in South
Africa. Once these parameters have been investigated, we apply them to investigate in
section 6 how rebalancing strategies would have performed historically in South Africa.
In section 7 we adapt the bootstrap technique to mimic the volatility typical of periods
like the 2008-9 credit crisis, thus providing a high-volatility stress test of the
comparative benefits of different rebalancing strategies under such conditions. Our
paper is concluded in section 8.
TYPES OF REBALANCING
Strategies for rebalancing portfolios are many and varied in the literature. The most
basic strategies which we consider here are pure drift, calendar rebalancing and range
rebalancing. Only these three methods are discussed in this paper; for more strategies
the reader may consult Donohue & Yip (2003:55).
Any rebalancing strategy may be described by:
1. the event that triggers rebalancing; and
2. the amount to buy or sell of each asset once rebalancing is triggered.
Pure drift
Pure drift is a rule that does not rebalance. The asset allocations are allowed to wander,
or ‘drift’. Apart from being trivial to understand and implement, it minimises
transaction costs, portfolio turnover, and typically the shortfall probability as well. The
latter effect arises because, in an efficient market, the portfolio will become dominated
62 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
by the most volatile of its assets. The clear disadvantage of pure drift is poorer
diversification resulting in greater variability than that justified by the returns earned.
Furthermore, pure drift maximises tracking error, because it makes no attempt to follow
the benchmark allocation. While it is unlikely that an institutional investor would adopt
a pure drift rebalancing strategy, it forms an important benchmark for assessing the
performance of other strategies and is a critical concept in the development of theories
about optimal portfolio rebalancing.
Calendar rebalancing
Calendar rebalancing is a rule to rebalance at regular calendar intervals, e.g. weekly,
quarterly or annually. The rebalancing ‘trigger event’ is reaching the end of the week,
quarter or year. Once rebalancing has been triggered, each asset is bought or sold in
such a quantity that the proportion invested in each asset matches the proportions in the
benchmark portfolio.
Calendar approaches have the advantage of being easy to understand and implement.
They make operational sense, because they can be administered with the accounting
processes (Chan 1998:42). Calendar methods are unbiased, in the sense that they do not
assume any momentum or mean reversion effects in relative price processes, consistent
with an efficient market. Their chief disadvantage is inefficiency in the presence of
transaction costs, because they trade when it is unnecessary to do so and they trade too
far toward the benchmark allocation.
Range rebalancing
Range rebalancing is a rule to rebalance whenever asset allocations break through a
predefined tolerance band around the benchmark allocation. These tolerance bands form
a so-called ‘no-trade region’, because no trading of assets is necessary while allocations
remain within these tolerance bands (Leland 1999:2).
Breaking through such tolerance bands forms the rebalancing trigger event. Once
rebalancing has been triggered by a particular asset, that asset is bought or sold in an
amount that brings its allocation within the tolerance band once more. One common
approach is to buy or sell the asset in such a quantity that the proportion invested in that
asset matches the proportion in the benchmark portfolio. This is known as trading ‘to
the benchmark’, or trading ‘to target’. An alternative approach is to buy or sell the asset
in such a quantity that the proportion invested in that asset falls just within the tolerance
band. The asset allocation may immediately break through the tolerance band again, in
which case it is again traded just within the tolerance band. It can be seen that
rebalancing trades may often occur in patches – there may be periods of no rebalancing
followed by many rebalancing trades in rapid succession. In section 5, we present a
graph of such batch-trading as Figure 3. The latter approach is known as trading ‘to
boundary’.
The literature has observed that, in the absence of fixed costs to trading, it is optimal to
trade to the boundary of the no-trade region rather than to the target allocation (Leland,
1999:3; Dybvig, 2005:3). This result arises because shallow trades incur substantially
less transaction costs without experiencing much higher tracking error than the deep
trades required to rebalance to target. We therefore found it appropriate in our
G Sher & GDI Barr 63
simulations below to investigate the tolerance bands of only the shallow-trading
approach.
Range methods are conceptually more complicated than the calendar and drift methods,
but they also offer substantial flexibility. For example, one could assign different
tolerance bands to each asset class and could compare range strategies that trade to-
target with range strategies that trade, say, to halfway between the target allocation and
the tolerance band, as proposed by Masters (2003).
They are, however, more difficult to implement to the extent that portfolio allocations
must be monitored continuously and short-term results are more susceptible to
manipulation. Range methods enjoy distinct performance advantages over their calendar
counterparts, maximising Sharpe ratios and controlling costs, turnover and shortfall.
They give managers more direct control over tracking errors, and can be extensively
customised.
THEORY OF OPTIMAL PORTFOLIO REBALANCING
So-called ‘optimal’ portfolio rebalancing has been researched over the last decade in
international investment journals (Leland, 1999; Donohue & Yip, 2003; Dybvig, 2005;
Sun, Fan, Chen, Schouwenaars & Albota 2006). However, in trying to implement
current state-of-the-art methodology in a South African context, portfolio managers will
encounter several difficulties including their reliance on the complex mathematics of
stochastic calculus and dynamic programming, which may cause difficulty when
designing procedures and explaining them to clients; their reliance on the existence and
specification of utility functions; their reliance on the existence and specification of a
‘coefficient of risk aversion’; their reliance on forecasts of stable correlations between
assets and their reliance on assumed lognormal distribution of asset returns.
In short, by relying on such a restrictive set of assumptions, no previous work to our
knowledge into so-called ‘optimal’ portfolio rebalancing provides the portfolio manager
with a practical evaluation of the effects of the choice of rebalancing strategy that
allows the manager to make an informed decision about what strategy to employ.
There is some consensus in international research that calendar rebalancing is inferior to
other rebalancing strategies (Sun et al., 2006:42, Masters, 2003:52, Donohue & Yip,
2003:63, Clark, 1999:19) because it ignores the closeness of the asset allocation to the
benchmark allocation and therefore trades when unnecessary. Range strategies in their
various forms are most often advocated.
Leland (1999) has investigated the problem of ‘optimal’ portfolio rebalancing with the
fewest restrictive assumptions. This author shows that trading to the boundary of the
no-trade region is more optimal than trading to somewhere further within the region, for
example trading to the benchmark allocation. The difficulty lies not in how far to trade
towards the benchmark allocation, but in how to specify the no-trade region. Thus far
we have presented the no-trade region as a tolerance band to be set around each asset’s
benchmark allocation, but this constitutes a simplification.
First note that it is plausible to set different tolerance bands around each asset in the
portfolio, so that equity may be allowed to vary within a 5% band while other assets
would be restricted to a 3% band, for example. In the case of only two risky assets,
64 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
equity and bonds, with cash being the risk-free asset, it would be possible to plot the no-
trade region as a rectangle, as shown in the first panel of Figure 1 below. Then at any
point in time the current asset allocation could be represented by a single point on this
plot. The position of this point, inside or outside the no-trade region, would determine
whether that asset should be bought or sold.
It can then be seen by simple extension that the no-trade region could be skewed, an
example of which is in the second panel of Figure 1 below. With such a no-trade region
it makes no sense to speak of ‘tolerance bands’ around each asset’s benchmark
allocation, because these bands are interdependent. Leland (1999) shows that the
‘optimal’ no-trade region can be reduced, without significant loss of information, to one
with straight sides in the form of the second panel in Figure 1 below. The exact four
points that specify the corners of the region need to be solved with stochastic calculus
and the resulting shape will depend on the correlation between the assets, see, for
example, Dybvig (2005) and Donohue & Yip (2003).
Figure 1. Example no-trade regions for stocks and bonds in the case of two risky
assets and one risk free asset. Each axis represents the active allocation, i.e. the
excess of the current allocation over the benchmark allocation for that asset.
Figures were produced by the authors using the free statistical package R (www.r-
project.org).
Few papers have been produced that compare the performance of rebalancing strategies
in a South African context. The seminal work of Firer et al. (2003) draws on an earlier
MBA research report by Peagam and Brunyee (2001), which offers a thorough literature
review for the interested reader. Swanepoel (2007) developed a rebalancing strategy for
the South African environment, but benchmarked the strategy against Masters (2003), a
study the validity of which has been called into question by, amongst others, Dybvig
(2005:3).
-5 0 5
-4-2
02
4
No-trade region for stocks and bonds
Active equity allocation (%)
Active
bo
nd
allo
ca
tio
n (
%)
sell bonds
buy equity sell equity
buy bonds
no trade
-3 -2 -1 0 1 2 3
-3-2
-10
12
3
No-trade region for stocks and bonds
Active equity allocation (%)
Active
bo
nd
allo
ca
tio
n (
%)
sell bonds
buy equity
sell equity
buy bonds
no trade
G Sher & GDI Barr 65
THE BOOTSTRAP RESAMPLING PROCEDURE AND DATA DESCRIPTION
Research that desires to simulate or theorise about the performance of assets and
portfolios has to make sometimes uncomfortable assumptions about the distribution of
asset returns. A common approach is to assume that asset returns are log-normally
distributed and to estimate the mean, variance and covariances of returns from historical
data. This approach is appealing because it can easily be reconciled with such theory as
the Capital Asset Pricing Model of Markowitz (1952) and much of the above research
takes this route (Leland, 1999; Donohue and Yip, 2003; Sun et al., 2006).
The log-normal model is an example of a parametric model, which is fully characterised
by a small set of parameters that can be assumed or estimated from historical data and
provides significant simplification in the analysis. Assuming that asset returns follow
log-normal distributions with known parameters even allows analytical, as opposed to
numerical, solutions to optimisation problems like the portfolio rebalancing decision.
However, this expositional simplification is achieved at a cost and there are several
known limitations to the log-normal model as a description of asset returns. In
particular, mean reversion and momentum effects are ignored, and log-normal models
understate the frequency of very large, clustered asset appreciations and depreciations.
Log-normal models also require the user to have knowledge about future means and
variances, and require that the covariance structure between assets is stable over time.
The criticisms of log-normal distributions therefore become criticisms of the theory or
empirical findings of research that relies on them to simplify analysis. In other words,
we cannot be sure that the conclusions drawn would be the same had alternative
distributional assumptions been made.
We have chosen to use a Bootstrap Resampling procedure to simulate asset class returns
in the South African context. This procedure is an example of a non-parametric
technique and Clark (1999) identifies some of its advantages over parametric
alternatives as follows:
The Bootstrap Resampling procedure preserves the risk-return properties of the
asset classes and preserves any contemporary dependence structure among their
returns.
The Bootstrap Resampling procedure avoids making potentially restrictive
assumptions about the return generating processes of each asset.
The Bootstrap Resampling procedure removes any trends or mean reversion
effects, in accordance with market efficiency, which may preference one
rebalancing method over others.
The Bootstrap Resampling procedure is conceptually quite simple. Instead of simulating
returns to each asset from a known probability distribution (the parametric approach),
our procedure draws a random day in history from the period 1 January 1999 to 16
February 2009. The return allocated for each asset class at one iteration of the procedure
is the return achieved on that randomly drawn day for each asset class. Thus, from a
data set of roughly ten years of daily returns we are able to sample a series of returns on
each asset class of any length (in this study we typically use a length of four years). This
series constitutes one ‘simulation’ and we are able to replicate this as many times as
66 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
necessary to achieve a distribution of performance measures. Every simulation has the
advantages listed by Clark (1999) above.
The main advantage of simulating from historical data, rather than applying rebalancing
strategies to historical data directly, is that we receive a distribution of outcomes, which
allows us to say with a degree of confidence how our results are likely to vary if the
experiment is repeated. Thus, we obtain confidence intervals around estimates rather
than bare point estimates.
We adjusted the Bootstrap Resampling procedure in section 8 to examine our results
under periods of high and autocorrelated volatility, which periods are typical of
emerging markets like South Africa. A description of the adjustments made to the
procedure is technical and can be found in the appendix.
Over the period of investigation, the daily means, standard deviations and correlation
structures are presented below.
SA
Equity SA Bonds
SA
Property
Foreign
Equity
Foreign
Bonds SA Cash
SA Equity 100.0%
SA Bonds 4.4% 100.0%
SA Property 45.8% -8.1% 100.0%
Foreign Equity 40.0% -26.0% 33.8% 100.0%
Foreign Bonds -9.0% -39.9% 15.7% 53.1% 100.0%
SA Cash 5.7% 5.8% -3.9% -0.9% -2.0% 100.0%
Mean 0.076% 0.055% 0.073% 0.016% 0.054% 0.038%
St Dev 1.335% 0.463% 1.054% 1.272% 1.189% 0.025%
RANGE REBALANCING UNDER A REGULAR BOOTSTRAP RESAMPLING PROCEDURE
When using range rebalancing to track a benchmark portfolio it will be necessary to
choose the size of the no-trade region, which affects a portfolio’s tracking error,
transaction cost and Sharpe ratio. We simulate Portfolio B invested in the South African
market using a regular bootstrap procedure under a range rebalancing strategy over a 4-
year period and recorded its tracking error relative to the benchmark. To capture a
distribution of tracking errors we simulate 1000 such 4-year periods. On the same set of
returns, the tolerance bands are progressively widened from 0% to 10%. The result is
given in Table 2 below.
The labelling in Table 2 is designed to have the results fit onto one page and can be
described as follows: ‘TE’ stands for ‘tracking error’, ‘TC’ stands for ‘transaction
costs’, ‘turnover’ stands for the proportion of the initial portfolio that is traded per year,
‘shortfall’ stands for the shortfall probability and ‘Sharpe’ stands for the ‘Sharpe ratio’
(Sharpe, 1966). For tracking error, transaction costs and the Sharpe ratio, we report the
upper and lower 2.5th
percentile of the distribution of results, represented by the
numbers 97.5% and 2.5% in the table respectively. For example, the upper 2.5th
percentile of the distribution of transaction costs is the cost that is exceeded in only
2.5% of simulations and could be interpreted as the cost that has a 2.5% probability of
being exceeded. The upper and lower 2.5th
percentiles therefore provide an estimate of
G Sher & GDI Barr 67
the 95% confidence interval around the median tracking error or transaction cost. In
Table 2, the median tracking error or transaction cost is identified by the 50% label.
Table 2. The effect of widening the tolerance bands around all asset classes under a
range rebalancing strategy that rebalances to the boundary of the no-trade region
Boundary (Tolerance band)
Uni
t 1% 2% 3% 4% 5% 6% 7% 8% 9% 10%
TE 2.5% % 0.20 0.35 0.45 0.52 0.57 0.60 0.60 0.61 0.61 0.61
TE 50% % 0.25 0.47 0.64 0.77 0.87 0.95 1.04 1.11 1.17 1.23
TE 97.5% % 0.30 0.60 0.88 1.12 1.33 1.44 1.58 1.69 1.85 1.98
TC 2.5% %2 5.79 2.34 0.93 0.28 0.07 0.00 0.00 0.00 0.00 0.00
TC 50% %2 9.00 4.62 2.93 1.97 1.41 1.05 0.80 0.61 0.43 0.24
TC 97.5% %2 15.52 9.78 7.19 6.04 5.21 4.36 3.88 3.63 3.39 3.17
# Rebalances
77.11 43.82 27.91 18.26 12.74 9.65 7.59 6.15 5.07 4.16
Turnover % 20.80 11.15 7.32 5.24 3.99 3.16 2.59 2.16 1.82 1.52
Shortfall % 49.86 49.37 49.26 49.20 49.16 49.07 49.04 49.03 49.01 49.00
Sharpe
2.5%
-
0.585
-
0.584
-
0.581
-
0.576
-
0.567
-
0.561
-
0.562
-
0.559
-
0.557
-
0.554
Avg Sharpe 0.494 0.500 0.501 0.500 0.497 0.494 0.492 0.489 0.487 0.485
Sharpe 97.5%
1.636 1.639 1.641 1.653 1.647 1.649 1.652 1.653 1.654 1.655
From Table 2 we see that tracking errors increase at a decreasing rate as the tolerance
bands are widened. When the range rebalancing strategy is applied by trading to the
boundary, median tracking errors are kept below 1.5%, although higher tracking errors
are probable when the bands become very wide.
Transaction costs drop exponentially as the no-trade region is widened around each
asset in the portfolio because fewer rebalancing trades are triggered with wider
tolerance bands. The 95% confidence band around the median transaction cost tightens
as the boundary is widened. These simulated transaction costs are proportional to the
cost assumptions in Table 1 above because the rebalancing decision is independent of
the level of transaction costs. Hence, to evaluate the impact of a doubling in the level of
costs the reader could double the cost estimates produced in Table 2 above as a suitable
approximation. The validity of the cost assumptions made therefore depends on the
relative, rather than absolute, costs of trading each asset.
Average Sharpe ratios decrease marginally as the tolerance bands are widened.
Confidence bands around Sharpe ratios are relatively wide, showing that it is perhaps
inappropriate to choose a rebalancing strategy for the purpose of maximising risk-
adjusted returns. However, the pattern of median Sharpe ratios is interesting – there is a
unique maximum around the 3% tolerance band, and Sharpe ratios are quite low near
the 0% tolerance band. (The 0% tolerance band corresponds to daily rebalancing and is
excluded from Table 2 in the interest of space, but may be found as the daily
rebalancing column in Table 3.)
The same pattern of Sharpe ratios is obtained if rebalancing is conducted to the
benchmark allocation, rather than to the boundary. However, the maximum Sharpe ratio
68 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
is achieved around the 5% tolerance band – a wider tolerance band than that needed
above. It is clear that trading to the benchmark allocation limits the active risk taken by
the portfolio more so than trading to the boundary, at parity of tolerance band.
Similar patterns of results can be observed for Portfolios A-E and the numerical output
is provided in the appendix. Tracking errors increase and become more uncertain as the
tolerance bands are widened, but transaction costs decline and become more certain as
tolerance bands are widened. Portfolios A and E have relatively low tracking errors
because they have relatively few asset classes. For each portfolio, average Sharpe ratios
peak and then decline. For Portfolios A and E, with fewer asset classes, this peak occurs
at narrow tolerance bands, whereas for Portfolios B-D this peak occurs at slightly wider
tolerance bands. In all cases, the confidence intervals around Sharpe ratios are relatively
wide and stable, reflecting a lot of variability in portfolio returns.
These are the effects on key portfolio performance measures of changing the tolerance
bands applied to all assets. The ultimate choice of tolerance band will, of course,
depend on the portfolio manager’s trade-off between tracking error and transaction
costs and on the composition of the asset portfolio.
CALENDAR REBALANCING UNDER A REGULAR BOOTSTRAP RESAMPLING PROCEDURE
Although calendar rebalancing has been shown repeatedly to be inferior to range
rebalancing, the performance differences may be outweighed by the practical
advantages of calendar rebalancing (Sun et al., 2006:41, Masters 2003:52, Donohue &
Yip 2003:63). As such, it is worth investigating the effects of changing the frequency of
rebalancing on measures of portfolio performance. We simulate Portfolio B invested in
the South African market under a calendar rebalancing strategy over a 4-year period and
recorded its tracking error relative to the benchmark. To capture a distribution of
tracking errors we simulate 1000 such 4-year periods using a regular bootstrap
procedure. On the same set of returns, the rebalancing frequency is reduced from daily
to annual rebalancing. The results are given in Table 3 below and labelling follows the
conventions discussed in the tables above.
Tracking errors relative to the benchmark increase as rebalancing frequency decreases,
but tracking errors remain below 1.5% per annum, even at the 97.5th
percentile. The
distribution of tracking errors also widens as rebalancing frequency decreases, so that
annual calendar rebalancing has a more uncertain tracking error than, say, monthly
calendar rebalancing.
Under daily rebalancing, 0.72% of the initial portfolio is on the median lost to
transaction costs each year, but it is possible to reduce this figure to 0.05% if
rebalancing is made annually. Confidence bands around transaction costs become
tighter as calendar frequency is decreased from daily to annually. Calendar rebalancing
also incurs higher transaction costs than range rebalancing. For example, a range
rebalancing strategy trading to the boundary with a tolerance band of 2% trades on
average 44 times per year, but incurs lower transaction costs than any calendar strategy
in Table 3. Once again, these transaction cost estimates could be scaled to observe the
effect of scaling the transaction cost assumptions made in Table 1.
G Sher & GDI Barr 69
Table 3. The effect of decreasing the frequency of calendar rebalancing.
Unit Daily Weekly Monthly Quarterly Semi-Annually Annually
TE 2.5% % 0.02 0.08 0.18 0.28 0.37 0.45
TE 50% % 0.03 0.10 0.22 0.38 0.54 0.75
TE 97.5% % 0.03 0.14 0.30 0.57 0.83 1.29
TC 2.5% %2 50.89 25.06 12.18 6.56 4.35 2.72
TC 50% %2 71.92 34.54 17.25 10.31 7.55 5.39
TC 97.5% %2 100.68 48.75 26.24 16.91 13.56 11.90
# Rebalances 250 50 12 4 2 1
Turnover % 172.49 83.62 41.91 25.29 18.84 14.18
Shortfall% % 100.00 57.24 51.21 50.10 49.73 49.49
Sharpe 2.5% -0.677 -0.657 -0.637 -0.625 -0.633 -0.616
Avg Sharpe 0.466 0.489 0.500 0.503 0.504 0.502
Sharpe 97.5% 1.834 1.858 1.860 1.857 1.847 1.833
As with range rebalancing, there appears to be an optimal rebalancing frequency near
quarterly and semi-annual rebalancing that maximises average risk-adjusted returns.
However, it may be inappropriate to base a rebalancing frequency decision on risk-
adjusted returns as these are quite variable, evidenced by their wide 95% confidence
bands.
Similar patterns of results are obtained for Portfolios A-E and numerical output is
provided in the appendix. Tracking errors increase and become more uncertain as the
calendar frequency is decreased from daily to annually, whereas transaction costs
decrease and become more certain as the calendar frequency is decreased from daily to
annually. Tracking errors are highest for Portfolios B-D, which have the most asset
classes. As the calendar rebalancing frequency is reduced from daily to annually, each
portfolio seems to achieve a unique maximum in their average Sharpe ratio, and this
maximum either occurs at quarterly or semi-annual rebalancing for all portfolios. We
must again stress that confidence intervals around the average Sharpe ratio are wide in
all cases and indicate substantial variability; thus decisions about the appropriate
calendar frequency for rebalancing should not be based on this measure alone.
HISTORICAL PERFORMANCE OF REBALANCING STRATEGIES
Having investigated the parameters to apply to range and calendar rebalancing
strategies, we can examine the historical performance of such strategies in South Africa.
The following analysis is based on 10 years’ daily data, as described above.
Before looking at the performance of each strategy, we consider the asset allocation
history for calendar rebalancing strategies presented in Figure 2 below. Weekly
calendar rebalancing tracks the benchmark much more closely than annual calendar
rebalancing, as expected. This explains the much lower tracking error achieved by
weekly calendar rebalancing. Similar graphs arise for other calendar frequencies.
70 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Figure 2. Proportions invested in each asset under annual and weekly calendar
rebalancing strategies. Portfolio B was used in this analysis.
We mentioned already that under range rebalancing, where trade occurs to the boundary
of the no-trade region, rebalancing trades occur in patches. This batch trading is
illustrated in Figure 3 below. Note that asset allocations never break the 3% tolerance
band around the benchmark allocation. Each time an asset would break this band, it is
traded just enough not to do so.
Figure 3. Proportions invested in each asset under a range rebalancing strategy
that trades to the boundary of the no-trade region. Rebalancing trades are
indicated by the vertical lines. The no-trade region for each asset is set at 3%
above and below the benchmark allocation.
To illustrate the relative performance of rebalancing strategies over the sample period,
we calculated the growth of an investment of R1 over the sample period and recorded
various performance measures. Two arbitrary range strategies were used as indications
only – these are not meant to be ‘best’ range strategies because such a choice would
depend on the investor’s trade-off between tracking error, transaction costs and risk-
G Sher & GDI Barr 71
adjusted return. The output is presented in Table 4 and labelling follows the
conventions discussed in the tables above.
From Table 4 it is interesting to note a difference between the two types of range
rebalancing reflected there: trading to the boundary (of the no-trade region) and trading
to the benchmark. Trading to the boundary requires on average 40 trades a year, versus
only 1 per year when trading to the benchmark. However, despite more frequent
trading, transaction costs are in fact lower on average, because each trade is much
smaller. Lower transaction costs without a significant change in portfolio standard
deviation are the primary arguments for range rebalancing to the boundary rather than
to the benchmark.
Table 4: Performance statistics for various rebalancing strategies per Rand
invested at the beginning of the sample period in Portfolio B. Range (100%) 3% is
a range rebalancing strategy where rebalancing is triggered when allocations
break the 3% tolerance bands to either side of the benchmark allocation and
trading then occurs to the boundary of the tolerance bands. Range (0%) 5% is a
range rebalancing strategy where rebalancing is triggered when allocations break
the 5% tolerance bands to either side of the benchmark allocation and trading
then occurs to the benchmark allocation
Unit
Range (100%)
3%
Trade to
boundary
Range (0%) 5%
Trade to
benchmark
Annual Quarte
rly
Weekl
y Daily
Pure
drift
Average
TE
(%) 0.78 0.87 1.32 0.76 0.19 0.03 2.67
Average
TC
(%2) 9.95 12.39 12.09 21.03 72.28 159.5
3
0.00
# Rebal. 40 1 1 4 50 250 0
Turnove
r
(%) 21.33 28.73 29.04 49.87 169.92 373.5
2
0.00
Return
pa
(%) 16.79 16.65 16.46 16.40 16.24 15.90 16.35
Volatilit
y pa
(%) 13.67 13.68 13.46 13.63 13.79 13.79 15.60
Sharpe
ratio
0.508 0.497 0.491 0.480 0.463 0.439 0.416
STRESS TESTING UNDER MORE TYPICAL SA FINANCIAL MARKET CONDITIONS
All markets exhibit periods of autocorrelated volatility. For example, the events of
September 11th
, 2001 resulted in an extended period of risk reassessment and
adjustment as the markets attempted to price. Similarly the events collectively termed
the credit crisis of middle 2008 to early 2009 were also characterised by heavily
autocorrelated volatility. Such effects tend to be exaggerated in emerging markets
which often track established markets but exhibit higher volatility. As an emerging
market, South Africa is prone to periods of high volatility and high autocorrelated
volatility and these have to be taken into account when assessing the choice of
rebalancing strategy. To reflect this reality, it is necessary to build high, auto-correlated
volatility into the resampling scheme.
To perform a stress test of our rebalancing strategies under a period of high,
autocorrelated volatility, we restrict our sample of historical returns to those days that
72 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
produced the largest 100 daily equity (absolute) returns. We allow for volatility patches
by inducing a simple first-order volatility autocorrelation in the bootstrap resampling
scheme that broadly replicates historical first-order equity volatility autocorrelation.
That is, first-order volatility autocorrelation is captured accurately, while volatility is
increased. Additionally, the revised scheme causes correlations between real and
nominal assets to drop, as observed in periods of market turmoil, increasing the need for
regular rebalancing to maintain the portfolio close to the target allocation. A
mathematical description of the scheme is provided in the appendix. Portfolio B is used
in the investigations that follow.
We simulate a portfolio using the stress test scheme under a range rebalancing strategy
that trades to the boundary of the no-trade region over a 4-year period with 1000
repetitions. On the same set of returns, the tolerance bands are progressively widened
from 1% to 10%. The result is given in Table 5 below. We also simulate a portfolio
using the stress test scheme under a calendar rebalancing strategy over a 4-year period
with 1000 repetitions. On the same set of returns, the trading frequency is decreased
from daily to annually. The results are given in Table 6 below and labelling follows the
conventions discussed in the tables above.
Table 5: The effect of widening the tolerance bands under a range rebalancing
strategy that trades to the boundary of the no-trade region, in the presence of high
and autocorrelated volatility, for Portfolio B
Boundary (Tolerance band)
Unit 1% 2% 3% 4% 5% 6% 7% 8% 9% 10%
TE 2.5% % 0.04 0.59 1.10 1.55 1.94 2.27 2.59 2.89 3.20 3.52
TE 50% % 0.05 0.63 1.21 1.78 2.32 2.83 3.33 3.82 4.29 4.75
TE 97.5% % 0.05 0.67 1.32 2.00 2.68 3.35 4.04 4.72 5.36 5.99
TC 2.5% %2 252.58 71.36 43.12 32.05 27.00 24.01 21.52 20.33 18.63 16.26
TC 50% %2 833.92 225.73 136.54 101.44 81.84 72.69 66.92 62.20 58.06 53.74
TC 97.5% %2 2905.72 811.57 524.33 388.30 320.22 271.64 244.23 235.29 230.64 216.78
# Rebalances 250 150 109 92 82 76 72 68 65 61
Turnover % 2456.59 669.06 442.94 354.57 306.23 275.07 251.93 233.04 216.22 200.54
Shortfall % 100 50 48 48 47 47 46 46 46 46
Sharpe 2.5%
0.365 0.422 0.482 0.520 0.569 0.612 0.670 0.717 0.750 0.791
Avg Sharpe
2.122 2.207 2.258 2.313 2.369 2.425 2.479 2.532 2.584 2.635
Sharpe 97.5%
4.405 4.507 4.569 4.605 4.623 4.704 4.697 4.828 4.898 4.968
As expected, the number of trades and cost of trading rises dramatically under this
stress test. Nevertheless, Sharpe ratios are boosted by the higher mean return to all asset
classes. Range rebalancing strategies seem to produce slightly better average Sharpe
ratios, but these Sharpe ratios are also more widely spread under range rebalancing (so
that at the lower 2.5th percentile calendar strategies produce better Sharpe ratios than
range strategies).
G Sher & GDI Barr 73
Table 6: The effect of decreasing the frequency of calendar rebalancing in the
presence of high and autocorrelated volatility, for Portfolio B
Unit Daily Weekly Monthly Quarterly Semi-Annually Annually
TE 2.5% % 0.04 0.71 1.51 2.38 3.15 4.35
TE 50% % 0.05 0.78 1.78 3.14 4.59 6.70
TE 97.5% % 0.05 0.85 2.09 4.07 6.51 11.17
TC 2.5% %2 281.47 122.79 55.84 35.21 30.49 28.54
TC 50% %2 810.00 340.82 166.49 109.66 94.57 95.42
TC 97.5% %2 2892.21 1230.11 651.31 458.63 439.74 496.41
# Rebalances 250 50 12 4 2 1
Turnover % 2364.82 1024.85 538.76 392.57 376.18 398.64
Shortfall% % 100 50 48 47 46 46
Sharpe 2.5% 0.456 0.509 0.514 0.588 0.662 0.854
Avg Sharpe 2.107 2.168 2.209 2.264 2.348 2.509
Sharpe 97.5% 4.299 4.394 4.452 4.538 4.599 4.769
Under volatility stress, range strategies that rebalance to the boundary of the tolerance
band require more frequent trading than calendar strategies. A priori we might therefore
expect range strategies to underperform, but from the tables we see that transaction
costs are in fact lower for range rebalancing strategies. Once again, range rebalancing
strategies that trade to the boundary of the tolerance band (i.e. no-trade region) conduct
many small trades that keep transaction costs and portfolio turnover to a minimum.
Calendar strategies on the other hand trade less frequently but more deeply; for
example, annual rebalancing makes only one trade per year, but this trade results in
398.6% of the portfolio being turned over on average, whereas 10% range rebalancing
trades on average 57 times a year and results in 185% of the portfolio being turned over
on average. These unusually large trading figures, in excess of 100% per year, arise
because the total value traded in the four-year period has been expressed as an
annualised proportion of the opening portfolio value. Range and calendar are both able
to keep median tracking errors below 6 and 7 percent respectively, but range strategies
seem to have a slight advantage here too.
CONCLUSIONS
Choice of rebalancing strategy will have implications for a portfolio’s tracking error,
risk-adjusted return and transaction costs. Much of the current literature on rebalancing
methods that has emerged over the last decade and which considers so-called ‘optimal’
portfolio rebalancing methods is limited by restrictive assumptions, often not
appreciated by portfolio managers. Their implementation may thus often result in
portfolio managers not fully understanding the risks in their rebalancing policy.
Rather than make assumptions to enable us to identify an ‘optimal’ rebalancing
strategy, we present the implications of an empirical study which uses sampling of
actual South African returns to mimic the performance of local asset classes. This
approach allows us to consider the relative merits of various calendar and range
rebalancing strategies without restrictive assumptions. The simulations using a regular
bootstrap procedure suggest that calendar strategies have the advantages of
74 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
administrative simplicity and reduced tracking errors, while range strategies provide
more flexibility and limit transaction costs.
Moreover, once we modify our bootstrap procedure to account for periods of high and
autocorrelated volatility that are typical of emerging markets, we find evidence
suggesting that range strategies are more robust than calendar strategies by virtue of
their ability to constrain transaction costs without significantly raising tracking errors. A
summary of the cost implications of calendar and range rebalancing strategies is
provided in Table 7 below, which follows the notation conventions we have used above.
The table summarises the simulation results of the four key tables above relating to the
transaction costs of Portfolio B. If we normalise the initial portfolio to R100, then the
rows labelled TC/TE can be interpreted as the cost (in Rands) per percentage point of
tracking error achieved by the rebalancing method. Scanning the rows reveals that range
rebalancing achieves each percentage point of tracking error, or reduction therein, at a
lower cost than calendar rebalancing. Similarly, the average cost per trade is lower for
range rebalancing. However, it is important to note here that these interpretations are
only illustrative, because in general the median of a ratio is not the same as the ratio of
medians. This paper thus concludes that in typical South African conditions, where
autocorrelated volatility is taken into account, range rebalancing methods could be
expected to achieve lower transaction costs than calendar rebalancing methods and to
protect the portfolio more in times of market stress.
Table 7: Summary of the median transaction costs incurred per unit tracking
error and of the median transaction costs incurred per trade, for Portfolio B, with
and without the stress-test
Boundary (Tolerance band)
Unit 0% 2% 4% 6% 8% 10%
No stress
TC/TE 2397.33 9.83 2.56 1.11 0.55 0.20
TC/Trade % 0.29 0.11 0.11 0.11 0.10 0.06
Stress test
TC/TE 16200.00 358.30 56.99 25.69 16.28 11.31
TC/Trade % 3.24 1.50 1.10 0.96 0.91 0.88
Calendar frequency
Unit Daily Weekly Monthly Quarterly Semi-
Annually
Annually
No stress
TC/TE 2397.33 345.40 78.41 27.13 13.98 7.19
TC/Trade % 0.29 0.69 1.44 2.58 3.78 5.39
Stress test
TC/TE 16200.00 436.95 93.53 34.92 20.60 14.24
TC/Trade % 3.24 6.82 13.87 27.42 47.29 95.42
G Sher & GDI Barr 75
Future work in this area should consider the role of taxes in affecting the rebalancing
decision. Non-zero taxes would affect our above analyses by increasing transaction
costs without having significant effect on the other performance measures. When
allowing for tax it would be important to distinguish between different institutional
investors and between income and capital gains tax.
REFERENCES
Brown, D.T., Ozik, G. and Scholtz, D. (2007), Rebalancing revisited: The role of
derivatives, Financial Analysts Journal, 63(5):32-44.
Clark, T. (1999). Efficient portfolio rebalancing, Dimensional Fund Advisors Inc.
Available online at http://dbpaustin.com/pwm/TrumanClarkEfficientRebalancing.pdf.
[Accessed December 2010]
Daryanani, G. (2008). Opportunistic rebalancing: A new paradigm for wealth managers,
Journal of Financial Planning, 21(1):48-61.
Donohue, C. and Yip, K. (2003). Optimal portfolio rebalancing with transaction costs,
Journal of Portfolio Management, 29(Summer):49-63.
Dybvig, P. (2005). Mean-variance portfolio rebalancing with transaction costs, Working
paper, Draft version January 2, Washington University.
Engle, R.F. (1982). Autoregressive conditional heteroscedasticity with estimates of
variance ofUnited Kingdom inflation, Econometrica, 50:987-1008.
Firer, C., Peagam, J. and Brunyee, W. (2003). Rebalancing strategies and the
performance of balanced portfolios: 1925-2001, Investment Analysts Journal, 58:17-28.
Leland, H. (1999). Optimal portfolio management with transaction costs and capital
gains taxes, Working paper, Draft version December 22, Haas School of Business,
University of California, Berkeley.
Markowitz, H.M. (1952). Portfolio selection, Journal of Finance, 7(1):77-91.
Masters, S. (2003). Rebalancing: establishing a consistent framework, Journal of
Portfolio Management, 29(Spring):52-57.
Peagam, J. and Brunyee, W. (2001). Portfolio rebalancing: The impact of rebalancing
strategies on investment portfolio risk / return profiles, MBA Research Report,
University of Cape Town Graduate School of Business.
Sharpe, W. (1966). Mutual fund performance, The Journal of Business, 39(1, part 2:
Supplement on security prices):119-138.
Sun, W., Fan, A., Chen, L., Schouwenaars, T., and Albota, M. (2006). Optimal
rebalancing for institutional portfolios, Journal of Portfolio Management, 36(3):66-68.
Swanepoel, J. (2007). The development of an alternative rebalancing strategy for a
pension fund, PhD Thesis, Northwest University, South Africa.
76 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Appendix: Description of the ARCH scheme
Let our dataset contain N daily returns from which we want to sample a sample period
of length m days (in our study we used m=1000 days, which is equivalent to 4 years).
Furthermore, we rank the daily returns in our dataset from largest to smallest on their
absolute value, ranks 1 to N. Since we have six assets in the portfolio, the ranking is
achieved by weighting the returns on any one day by their portfolio B allocation and
then considering the absolute value.
The straightforward bootstrap resampling procedure involves sampling from historical
returns with replacement and can be described as follows:
and iindex is the rank of return number i , which effectively picks one of the N
returns at random ignoring the ranking system. The modified bootstrap resampling
procedure takes the rank of the return to be a weighted average of the previous rank
and the regular bootstrap sample rank. That is:
The value of (0,1)v determines the extent of the first-order absolute return
autocorrelation, which is a proxy for the first-order volatility autocorrelation. The value
of v is set using a Method of Moments Estimator so that the scheme reflects a mean
first-order equity volatility autocorrelation for portfolio B which is as close as possible
to the value observed in the dataset. Under normal circumstances, the scheme is able to
capture first-order equity volatility autocorrelation with only a small spread about the
mean, but second- and higher-order volatility autocorrelations in the data are not
captured.
The high volatility stress test used in this paper restricts the dataset to the 100 largest
returns by absolute value. This restriction amounts to replacing N with 100 in
formulation (2) above.
G Sher & GDI Barr 77
Appendix: Range Rebalancing for Portfolios A – E
1% 2% 3% 4% 5% 6% 7% 8% 9% 10%
Portfolio A
TE 2.5% 0.11 0.17 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18
TE 50% 0.13 0.26 0.36 0.43 0.45 0.45 0.45 0.45 0.45 0.45
TE 97.5% 0.16 0.34 0.53 0.69 0.83 0.93 1.02 1.09 1.13 1.17
TC 2.5% 0.49 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
TC 50% 1.24 0.56 0.27 0.05 0.00 0.00 0.00 0.00 0.00 0.00
TC 97.5% 2.57 1.78 1.35 1.07 0.84 0.60 0.50 0.36 0.23 0.09
# Rebal. 20.75 10.42 6.14 3.43 1.69 0.81 0.43 0.28 0.19 0.14
Turnover 4.26 2.11 1.19 0.61 0.28 0.13 0.07 0.05 0.03 0.02
Shortfall% 49.83 49.71 49.58 49.58 49.59 49.59 49.60 49.60 49.60 49.60
Sharpe 2.5% -0.652 -0.650 -0.648 -0.646 -0.646 -0.647 -0.649 -0.651 -0.653 -0.655
Avg Sharpe 0.453 0.451 0.448 0.447 0.445 0.445 0.445 0.444 0.444 0.444
Sharpe 97.5% 1.691 1.690 1.686 1.689 1.686 1.684 1.683 1.683 1.683 1.683
Portfolio B
TE 2.5% 0.20 0.35 0.45 0.52 0.57 0.60 0.60 0.61 0.61 0.61
TE 50% 0.25 0.47 0.64 0.77 0.87 0.95 1.04 1.11 1.17 1.23
TE 97.5% 0.30 0.60 0.88 1.12 1.33 1.44 1.58 1.69 1.85 1.98
TC 2.5% 5.79 2.34 0.93 0.28 0.07 0.00 0.00 0.00 0.00 0.00
TC 50% 9.00 4.62 2.93 1.97 1.41 1.05 0.80 0.61 0.43 0.24
TC 97.5% 15.52 9.78 7.19 6.04 5.21 4.36 3.88 3.63 3.39 3.17
# Rebal. 77.11 43.82 27.91 18.26 12.74 9.65 7.59 6.15 5.07 4.16
Turnover 20.80 11.15 7.32 5.24 3.99 3.16 2.59 2.16 1.82 1.52
Shortfall% 49.86 49.37 49.26 49.20 49.16 49.07 49.04 49.03 49.01 49.00
Sharpe 2.5% -0.585 -0.584 -0.581 -0.576 -0.567 -0.561 -0.562 -0.559 -0.557 -0.554
Avg Sharpe 0.494 0.500 0.501 0.500 0.497 0.494 0.492 0.489 0.487 0.485
Sharpe 97.5% 1.636 1.639 1.641 1.653 1.647 1.649 1.652 1.653 1.654 1.655
Portfolio C
TE 2.5% 0.21 0.37 0.50 0.59 0.66 0.70 0.73 0.75 0.75 0.75
TE 50% 0.24 0.47 0.67 0.84 0.97 1.08 1.17 1.25 1.31 1.37
TE 97.5% 0.29 0.58 0.88 1.16 1.43 1.66 1.84 1.98 2.17 2.29
TC 2.5% 7.45 3.20 1.49 0.67 0.21 0.02 0.00 0.00 0.00 0.00
TC 50% 11.31 6.09 3.89 2.70 1.89 1.39 1.07 0.78 0.57 0.37
TC 97.5% 17.64 11.41 9.20 7.87 6.59 5.57 4.96 4.32 4.02 3.71
# Rebal. 86.43 51.45 35.07 24.75 17.34 12.52 9.50 7.49 6.07 4.94
Turnover 25.27 13.91 9.36 6.81 5.15 4.04 3.26 2.69 2.26 1.91
Shortfall% 49.85 49.34 49.13 49.05 49.02 49.02 49.00 48.97 48.95 48.93
78 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Sharpe 2.5% -0.516 -0.505 -0.495 -0.503 -0.505 -0.509 -0.510 -0.513 -0.518 -0.524
Avg Sharpe 0.503 0.512 0.516 0.516 0.515 0.512 0.509 0.507 0.504 0.502
Sharpe 97.5% 1.690 1.701 1.695 1.697 1.697 1.697 1.693 1.688 1.681 1.678
Portfolio D
TE 2.5% 0.19 0.32 0.42 0.51 0.56 0.60 0.63 0.63 0.63 0.63
TE 50% 0.23 0.43 0.58 0.70 0.81 0.91 1.00 1.10 1.18 1.26
TE 97.5% 0.29 0.58 0.84 1.03 1.19 1.35 1.50 1.63 1.75 1.88
TC 2.5% 5.86 2.32 0.96 0.39 0.15 0.00 0.00 0.00 0.00 0.00
TC 50% 8.42 4.38 2.83 2.00 1.54 1.20 0.96 0.76 0.57 0.37
TC 97.5% 12.09 7.33 5.86 5.21 4.72 4.38 4.29 4.13 4.04 3.93
# Rebal. 73.72 40.19 24.85 17.45 13.17 10.34 8.48 7.09 5.97 5.05
Turnover 18.50 9.95 6.71 5.08 4.08 3.41 2.93 2.56 2.24 1.96
Shortfall% 49.70 49.32 49.23 49.21 49.23 49.18 49.13 49.09 49.05 49.03
Sharpe 2.5% -0.523 -0.509 -0.506 -0.504 -0.509 -0.514 -0.511 -0.510 -0.507 -0.502
Avg Sharpe 0.529 0.534 0.532 0.528 0.523 0.519 0.515 0.511 0.508 0.504
Sharpe 97.5% 1.695 1.675 1.668 1.670 1.671 1.659 1.662 1.666 1.669 1.671
Portfolio E
TE 2.5% 0.12 0.22 0.29 0.33 0.34 0.35 0.35 0.35 0.35 0.35
TE 50% 0.14 0.27 0.39 0.51 0.61 0.70 0.78 0.85 0.88 0.89
TE 97.5% 0.16 0.32 0.49 0.67 0.84 1.02 1.18 1.34 1.50 1.65
TC 2.5% 2.08 0.67 0.18 0.02 0.00 0.00 0.00 0.00 0.00 0.00
TC 50% 3.31 1.71 1.13 0.82 0.60 0.43 0.26 0.10 0.00 0.00
TC 97.5% 4.91 3.32 2.94 2.85 2.82 2.79 2.75 2.67 2.57 2.50
# Rebal. 44.24 24.59 16.84 12.33 9.37 7.26 5.64 4.42 3.50 2.81
Turnover 8.82 4.87 3.42 2.64 2.12 1.75 1.46 1.23 1.04 0.88
Shortfall% 49.62 49.49 49.39 49.28 49.22 49.17 49.14 49.12 49.11 49.10
Sharpe 2.5% -0.421 -0.409 -0.397 -0.387 -0.387 -0.387 -0.389 -0.392 -0.402 -0.402
Avg Sharpe 0.709 0.707 0.701 0.693 0.686 0.679 0.672 0.667 0.662 0.659
Sharpe 97.5% 1.868 1.850 1.860 1.851 1.839 1.856 1.839 1.839 1.830 1.827
G Sher & GDI Barr 79
Appendix: Calendar Rebalancing for Portfolios A – E
Daily Weekly Monthly Quarterly
Semi-
Annually Annually
Portfolio A
TE 2.5% 0.01 0.03 0.06 0.10 0.13 0.15
TE 50% 0.01 0.04 0.08 0.14 0.20 0.27
TE 97.5% 0.01 0.05 0.11 0.20 0.31 0.49
TC 2.5% 11.03 5.31 2.62 1.39 0.78 0.39
TC 50% 17.60 8.41 4.02 2.33 1.66 1.18
TC 97.5% 28.85 13.66 7.00 4.34 3.53 2.91
# Rebal. 250 50 12 4 2 1
Turnover 59.72 28.60 14.01 8.32 6.14 4.55
Shortfall% 100.00 55.87 51.02 50.17 49.96 49.87
Sharpe 2.5% -0.616 -0.613 -0.612 -0.612 -0.617 -0.615
Avg Sharpe 0.461 0.466 0.467 0.467 0.467 0.465
Sharpe 97.5% 1.711 1.716 1.718 1.714 1.716 1.714
Portfolio B
TE 2.5% 0.02 0.08 0.18 0.28 0.37 0.45
TE 50% 0.03 0.10 0.22 0.38 0.54 0.75
TE 97.5% 0.03 0.14 0.30 0.57 0.83 1.29
TC 2.5% 50.89 25.06 12.18 6.56 4.35 2.72
TC 50% 71.92 34.54 17.25 10.31 7.55 5.39
TC 97.5% 100.68 48.75 26.24 16.91 13.56 11.90
# Rebal. 250 50 12 4 2 1
Turnover 172.49 83.62 41.91 25.29 18.84 14.18
Shortfall% 100.00 57.24 51.21 50.10 49.73 49.49
Sharpe 2.5% -0.677 -0.657 -0.637 -0.625 -0.633 -0.616
Avg Sharpe 0.466 0.489 0.500 0.503 0.504 0.502
Sharpe 97.5% 1.834 1.858 1.860 1.857 1.847 1.833
Portfolio C
TE 2.5% 0.03 0.09 0.20 0.32 0.42 0.53
TE 50% 0.03 0.11 0.25 0.44 0.62 0.86
TE 97.5% 0.04 0.16 0.35 0.65 0.95 1.41
TC 2.5% 59.97 29.02 14.13 7.86 5.08 3.31
TC 50% 80.94 39.46 19.69 11.85 8.70 6.31
TC 97.5% 107.81 53.07 27.58 18.52 15.44 13.18
# Rebal. 250 50 12 4 2 1
Turnover 191.98 93.31 46.89 28.51 21.16 15.88
Shortfall% 100.00 57.29 51.18 50.02 49.63 49.39
80 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Sharpe 2.5% -0.639 -0.614 -0.611 -0.608 -0.582 -0.611
Avg Sharpe 0.441 0.470 0.484 0.490 0.492 0.491
Sharpe 97.5% 1.550 1.587 1.599 1.590 1.607 1.604
Portfolio D
TE 2.5% 0.02 0.08 0.18 0.29 0.37 0.46
TE 50% 0.03 0.10 0.23 0.40 0.56 0.78
TE 97.5% 0.03 0.13 0.29 0.55 0.85 1.34
TC 2.5% 48.93 23.37 11.07 5.91 3.60 2.18
TC 50% 60.25 29.07 14.38 8.51 6.10 4.48
TC 97.5% 76.12 37.48 19.28 12.86 10.78 9.29
# Rebal. 250 50 12 4 2 1
Turnover 142.34 69.05 34.50 20.89 15.36 11.71
Shortfall% 100.00 56.47 51.01 50.02 49.82 49.55
Sharpe 2.5% -0.630 -0.602 -0.590 -0.586 -0.590 -0.576
Avg Sharpe 0.503 0.534 0.548 0.551 0.548 0.544
Sharpe 97.5% 1.711 1.741 1.754 1.753 1.759 1.744
Portfolio E
TE 2.5% 0.01 0.05 0.11 0.18 0.23 0.27
TE 50% 0.01 0.06 0.14 0.25 0.35 0.50
TE 97.5% 0.02 0.08 0.18 0.34 0.55 0.91
TC 2.5% 24.48 11.35 4.97 2.43 1.44 0.76
TC 50% 27.26 12.96 6.28 3.67 2.65 1.89
TC 97.5% 30.60 14.94 7.92 5.27 4.34 3.79
# Rebal. 250 50 12 4 2 1
Turnover 77.86 37.52 18.66 11.24 8.37 6.30
Shortfall% 100.00 55.24 50.85 50.12 49.77 49.55
Sharpe 2.5% -0.369 -0.348 -0.341 -0.345 -0.338 -0.359
Avg Sharpe 0.658 0.685 0.696 0.697 0.694 0.683
Sharpe 97.5% 1.771 1.807 1.815 1.816 1.818 1.763
D Coetsee 81
A comment on research frameworks applied
in accounting research
D Coetsee
University of Johannesburg
Received: August 2010 SAJAR
Revised: December 2010, February 2011 Vol 25 No. 1
Accepted: February 2011 2011
pp.81 to 102
In the social sciences three different research frameworks are used: positivistic, interpretative
and critical. These frameworks approach research from different premises, which collectively
contribute to the pool of knowledge. Accounting institutions and their activities are fashioned by
human interventions whose perceptions of the truth change over time. Accounting information
and its treatment, therefore, cannot be adequately interpreted through the deterministic or
objective lens of a positivist framework. This article argues that there is a place for each of the
different research frameworks in accounting research and in an attempt to foster debate in South
Africa, the article identifies and explains the nature and the process of each social research
framework with specific reference to the South African context. This article also comments
briefly, where possible, on the research frameworks applied in accounting accredited journals of
South Africa.
KEY WORDS
Accounting research, critical, interpretative, mainstream, positivistic
Contact
INTRODUCTION
Most accounting departments at universities in South Africa are battling to increase
their scholarly activities, including research outputs. Van der Schyf (2008) revealed the
increased academic tension in Departments of Accounting at South African universities
to achieve their academic mission of scholarly activities. He provides the reason for this
as follows:
The reason for this is the fact that that the academic training of prospective chartered
accountants has long been the main academic focus in such departments, while they
have failed to do justice to their actual academic mission, namely scholarly activity in
accounting, in line with the essence of a university (Van der Schyf, 2008:22).
Before this, West (2006:131) assessed the global position of South African accounting
research and found that the accounting output of accounting academics is lower than
that of professional accountants in South Africa and that the local contribution to
82 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
accounting literature is poor. Van der Schyf specifically concludes as follows on the
academic responsibility of Departments of Accounting:
It was found that Departments of Accounting have a vital academic responsibility in the
university context, namely scholarly activity in accounting. In this regard, thorough
knowledge of the conceptual foundations of accounting (also known as accounting
theory) and research methodology play a key role as part of accounting as a social
science (Van der Schyf, 2008:22).
In the social sciences three main research frameworks are deployed: positivistic,
interpretative and critical (Henning, Van Rensburg & Smit, 2004). Since the accounting
discipline is a social activity (Coetsee, 2010; Baker & Bettner, 1997) or a social science
as Van der Schyf identified above, in which the appropriate accounting treatment in
practice is determined by human intervention, all three of the research frameworks
should play a role in accounting research. The question is whether accounting
academics in South Africa understand the application of these frameworks sufficiently.
This article reviews the three research frameworks as applied in accounting research.
The purpose of this article is to stimulate debate on the topic in the hope that it will
create the opportunity for further debate on accounting research in South Africa. This
article focuses on the nature and research process of each research framework by
incorporating views of the application of these frameworks in accounting research.
Where reference is made to academic views outside the field of accounting, this is done
in an attempt to enhance the discussions in accounting literature. Limited comments on
the application of these frameworks in accounting accredited journals of South Africa
are included in the article to foster further debate or research in South Africa.
The article undertakes a discussion of the different research frameworks using the
following structure: (1) an accounting perspective on the different research frameworks;
(2) the nature and process of the positivistic research framework; (3) the nature and
process of the interpretative research framework; (4) the nature and process of the
critical research framework; and (5) a structured comparison of the different research
frameworks.
ACCOUNTING PERSPECTIVES ON THE DIFFERENT RESEARCH FRAMEWORKS
International accounting scholars have also identified three different research
frameworks. Ryan, Scapens and Theobald (2002) identify the following three research
frameworks in accounting and finance research: mainstream accounting research,
interpretative accounting research and critical accounting research. According to Ryan
et al. (2002:41) mainstream accounting research is primarily concerned with the
functioning of accounting. They explain mainstream accounting research as follows:
Such work starts from an objective view of society, regards individual behavior as
deterministic, uses empirical observations and a positive research methodology
(2002:41).
Because mainstream accounting research is based on a positivistic research
methodology, the assumption made in this article is that mainstream accounting
research is synonymous with a positivistic research approach. Other accounting scholars
D Coetsee 83
agree that mainstream accounting research is based on a positivistic research framework
(Williams, 2009; Gaffikin, 2006; Baker & Bettner, 1997). Many accounting scholars are
calling for the use of other research frameworks, such as interpretative and critical
research in accounting (Granof & Zeff, 2008; Gaffikin, 2006; Macintosh, 2004;
Broadbent, 2002). Baker and Bettner (1997:307), for instance, believe that accounting
has an effect on social issues such as distribution of wealth, social justice, political
ideology and environmental degradation. Baker and Bettner (1997:295) specifically
refer to the following remark by Chua (1986:601), who as far back as the 1980s called
for a wider application of research frameworks:
Mainstream accounting is grounded in a common set of philosophical assumptions
about knowledge, the empirical world, and the relationship between theory and
practice. This particular world view, with its emphasis on hypothetico-deductivism and
technical control, possesses certain strengths but has restricted the range of problems
studied and the use of research methods. By changing the set of assumptions,
fundamentally different and potential rich insights are obtained. Two alternatives
worldviews and their underlying assumptions may be elucidated – the interpretative
and the critical.
This commentary agrees with a wider application of all the research frameworks
identified in social science. Accounting is socially constructed (Baker & Bettner,
1997:304), and is regarded by some as a social science (West, 2006: 131). Accounting
researchers should apply all three of the research frameworks commonly deployed in
the social sciences: the positivistic, interpretative and critical frameworks. Each
research framework is based on specific premises, differs in nature from the others and
is applied for different means and ends. Collectively these different research
frameworks contribute to the pool of academic knowledge, and should therefore also be
applied in accounting research. It is imperative that accounting researchers understand
the different research frameworks or approaches. The article aims to build such
understanding.
The positivistic framework focuses on describing the underlying phenomena (Deegan &
Unerman, 2006:8). The aim is to record the underlying phenomena and is thus a more
objective research approach. However, the process goes further than merely describing
the observations: it also explains and predicts the phenomena – hence the fact that it is
often referred to as a positivistic research methodology (Deegan & Unerman, 2006:8).
Interpretative research goes a step further in assessing the reasoning behind the
phenomena. Certain accounting academics explain interpretative research as follows:
This research is concerned with understanding the social world, and includes work that
seeks to understand the social nature of accounting practice (Ryan et al., 2002: 42).
IAR is based on the idea of understanding accounting practices in the field (Piber,
2008:851).1
The term interpretative research reflects a methodology perspective. In a general sense,
interpretative research attempts to describe, understand and interpret the meanings that
1IAR refers to Interpretative Accounting Research.
84 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
human actors apply to the symbols and the structure of the setting in which they find
themselves (Baker & Bettner, 1997:293).
Therefore interpretative research not only records the phenomena, but also incorporates
the perceptions and feelings of people – the reasons why they act in a certain way.
Interpretative research is still objective in that it records the views of human actors by
applying research methods objectively. In contrast, critical research questions the
practice and incorporates the view of the researcher (Wolk, Dodd & Rozychi, 2008:42).
The main goal of critical research is to promote change (Ryan et al., 2002:87). This is
the reason why accounting scholars explain the difference between interpretative and
critical research as follows:
Some critical accounting researchers … object to interpretative research because it
does not seek to promote a social critique and promote radical change. Nevertheless,
interpretative work is concerned with making sense of the social character of daily life
(Ryan et al., 2002:42).
Critical research can also be interpretative, but critical research adopts a particular
point of view regarding the research question, whereas interpretative research purports
to take a ‘neutral’ stance (Baker & Bettner, 1997:293).
The movement from total objectivity (positivistic research) to incorporating the feelings
of people (interpretative research) to subjectivity incorporating the views of the
researcher (critical research) creates different perspectives on research.
POSITIVISTIC RESEARCH FRAMEWORK
Its nature
More than twenty years ago Chua (1986: 606) explained beliefs about the physical and
social reality of mainstream accounting research, with its positivistic nature, as follows:
Ontologically, mainstream accounting research is dominated by a belief in physical
realism – the claim that there is a world of objective reality that exists independently of
human beings and that has a determinate nature or essence that is knowable.
The positivistic research framework is based on the premise that only one truth exists in
physical reality and that the aim of research is to capture that truth. Kirk and Miller
(1986:14) explain the incorporation of “positivism” in research as follows:
In its strongest form, positivism … assuming not only that there is an external world,
but that the external world itself determines absolutely the one and only view that can
be taken of it, independent of the process or circumstances of viewing.
This creates the notion that truth exists independently from the actions of humans,
which is why positivistic research is generally referred to as the scientific method of
research (Baker & Bettner, 1997; Henning et al., 2004:17). The question could therefore
be asked whether a positivistic approach (the scientific method) to research could be
used for a human activity such as accounting. Baker and Bettner (1997:304) provide the
following observation:
D Coetsee 85
Accounting is a socially constructed and subjective reality. Therefore, the scientific
method – wherein relationships among naturally occurring phenomena are assumed to
be enduring, quantifiable, and objectively determinable – is an incorrect paradigm that
limits the perspectives for doing accounting research.
However, the positivistic scientific method has become the main research framework of
mainstream accounting research. Deecan (2010:257), for instance, states that in
accounting research, positivistic research “focuses on the relationship between the
various individuals involved in providing the resources to an organisation and how
accounting is used to assists in in the functioning of these relationships”. In calling for a
move away from mainstream accounting research, Armstrong (2008:871) explains the
application of positivism in social sciences as follows:
Positivism, to be clear about it, is the belief that propositions about the social world can
be unambiguously verified against an objective social reality.
Armstrong therefore confirms that positivism could be used in social disciplines like
accounting provided that the outcome can be objectively verified. Henning et al.
(2004:17), who are social scientists, agree by stating that the purpose of positivistic
science is thus “what we can observe and measure”. The positivistic framework is thus
not limited to the natural sciences only. It can be (and is being) applied in the social
sciences. In this regard Henning et al. (2004:17) describe a positivist framework as
follows:
It is about finding truth and providing it through empirical means. It is a philosophical
position that holds that the goal of knowledge is simply to describe and, in some
designs, to explain and also to predict the phenomena that we experience (whether
quantitatively or qualitatively).
Positivism is thus a research framework that objectively observes and measures
phenomena, without incorporating human feelings. The changing social world is thus
objectively captured as if it were a phenomenon studied in the natural sciences.
Therefore positivistic research is normally quantitative in nature, although Henning et
al. (2004) confirm that it could be qualitative. However, even when positivistic research
is qualitative, the research must still be limited to describing and explaining the
phenomena, and should not move into the realm of interpretation. Empirical means are
normally used to capture the data.
Notwithstanding the thorough application of mainstream positivistic research in
international accounting research, especially in the United States (Reiter & Williams,
2002), many are critical about the application of positivistic research in accounting
(Armstrong, 2008: Baker & Bettner, 1997; Chambers, 1993; Boland & Gordon, 1992).
Parker (2008:910), for instance, expresses concerns about the current status of
mainstream accounting research:
As critics of conventional wisdom and the status quo, we face increasingly difficult
challenges. The interpretative community has a strong tradition of presenting vibrant
and challenging critiques of business, profession and government policies, practices
and ideologies. This is an essential role that falls to the dust if left to a positivistic and
uncritical acceptance of current structures and processes.
86 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
In summary, the aim of the positivistic framework is to record an objective reality that
exists independently from human behaviour. As applied in social science, it is the belief
that an objective social reality can be unambiguously verified. This is done by
objectively observing and measuring the phenomena under study.
The process
A distinct and generalised process is normally followed in positivistic research. Inanga
and Schneider (2005:230) explain the process of developing positivistic theories as
follows:
Positive theories attempt to describe real world situations as they are. Research based
on positive theories involves empirical observations of the relevant phenomena from
which a problem is defined. Data relevant to the problem are then collected and
hypotheses formulated and tested by independent process. If the theory that results is an
accurate representation (description) of the empirical phenomena, such a theory can be
used for predictive purposes. Induction follows empirical observation and takes the
form: “if event Y takes place, the outcome will be Z”. The greater the number of
empirical observations, the better supported the related induction will be.
The positivistic process is not only aimed at developing theory. Theory might be
regarded as the ultimate goal of research. However, all research contributes to the
growing pool of knowledge. The process identified by Inanga and Schneider starts with
empirical observations to define a problem. The next step is to formulate the hypothesis.
The hypothesis is then tested through an independent process. To test the hypothesis the
researcher must choose the appropriate method (including the research instrument). The
following process in positivistic research has been identified by Inanga and Schneider:
Observations (data) Hypothesis Method Testing Results
However, in accounting-related research the first part of the process is usually different.
Hypotheses are developed through deductive reasoning. Deductive reasoning is the
development of arguments through a logical process (Wolk et al., 2008:28). Once the
hypothesis is developed the appropriate research method (including the research
instrument) is then chosen to accept or reject the hypothesis through the testing process.
Statistical analysis forms an integral part of presentation of the results (Deegan,
2010:260).
The hypotheses can be based on developed hypotheses or can be created through the
deductive reasoning process. The validity of positivistic research in a social activity
such as accounting is firstly assessed by the appropriateness of the development of these
hypotheses. In the development of a hypothesis, the literature review (which provides
the current state of knowledge on the research issue) plays an important role. A link
must be created between the developed hypothesis(es) and the literature review. There
are many examples of hypotheses being developed and tested in South African
accounting research (De Clercq & Venter, 2009; Stainbank, 2009; Ruddy &
Everingham, 2008).
The development of models also forms an integral part of mainstream accounting
research. These models normally establish the relationship between data variables.
Statistical means are used to assess the correlation of the data included in the models.
D Coetsee 87
The Ohlson model (Ohlson 1995), for instance, creates the relationship between share
price and accounting data. Accounting-based valuation formulae to predict equity
values are another example of such models (Barth, Beaver, Hand & Landsman 1999;
Ohlson, 2005). In all these models earnings forecasting plays a very important role.
Earning forecasting models (Dechow, Hutton & Sloan, 1999) and earning quality
models (Kenton, 2006) are further developments in these relationship or valuation
models.
Modelling is also a characteristic of capital market research. Different views on the
application of capital market research in accounting have been expressed (Meek &
Thomas, 2004; Beaver, 2002; Kothari, 2001). Kothari (2001:105), for instance, states:
“The capital market research topics of current interest to researchers include tests of
market efficiency with respect to accounting information, fundamental analysis, and
value relevance of financial reporting”. Value-relevance research assesses how well
accounting amounts reflect information used by equity investors (Barth, Beaver &
Landsman, 2001:77).
In contrast, in South African accounting journals few examples of modelling or capital
market research are seen. Firer and Stainbank (2003), for instance, have modelled and
tested the relationship between the performance of a company’s intellectual capital and
(1) profitability, (2) productivity and (3) market valuation. Bradfield (1998) measured
the selection and timing abilities of South African fund managers. Research on the
Ohlson model has also been done a few times in South African accounting research
(Swartz & Negash, 2006; Swartz, Swartz & Firer, 2006). Accounting researchers in
South Africa lag behind international mainstream accounting research in model
development. A lack of a depth of such a research culture, and investment in time
(relative to international academics) due to a focus on profession training, as confirmed
by Van der Schyf (2008), could be argued to be the main cause.
The validity of positivistic research depends not only on the appropriateness of the
development of the hypothesis, model or research question. The other aspects of the
research process as identified by Inanga and Schneider above are also important. A
valid research instrument and research data must also be chosen to test the hypothesis or
model, and this must be followed by the correct application of statistical analysis to
verify the outcome. Since measurement or verification (positive or negative) is central
to positivistic research, the appropriateness of the measurement or verification must
always be considered
Two applications of the positivist framework in South African accounting journals,
Meditari Accounting Research and South African Journal of Accounting Research,
deserve mention. The first is data or content analysis. For instance, the financial data of
companies obtained from databases is analysed to achieve certain research goals.
Examples of such research are Chivaka, Siddle, Bayne, Cairney and Shev (2009);
Ackers (2009); Marx (2009).2 Some scholars refer to this research as archival research
(Chivaka et al., 2009). Shields (1997) identifies archival research as research that uses
archival data for analysis to identify regularities in data and to test hypothesis. The
South African examples identified above analysed databases, previous studies or
financial statements as archival data without setting hypotheses. Research objectives are
2Marx also used a structured questionnaire.
88 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
set, but hypotheses is not accepted or rejected. This illustrates that positivistic research
can be performed without the validation or rejecting of hypotheses.
The second application is the use of structured questionnaires to assess the perceptions
of people on a research question. Examples are Marx (2009),3 Charlton and Marx
(2009) and Nel and Steenkamp (2008). The research process is still controlled by the
researcher in the sense that no open-ended questions are included. In the examples
above the participants had to choose yes or no answers, or options from out of a list.
Deeper understanding of the reasons for the perceptions, as required by interpretative
research, was not sought. Henning et al. (2004:3) state that the distinction between
qualitative and quantitative research “lies in the quest for understanding and for in-
depth enquiry”. They further state that:
In a quantitative study the focus will be on control of all the components in the actions
and representations of the participants – the variables will be controlled and the study
will be guided with an acute focus on how variables are related. The researcher plans
and executes this control in the way the study and its instruments are designed.
Respondents or research subjects are usually not free to express data that cannot be
captured by the predetermined instruments (Henning et al., 2004:3).
Quantitative research forms the foundation of the positivistic research (Henning et al.,
2004:18). Therefore, structured questionnaires that control the variables, and that do not
allow for further explanation by participants, objectify the underlying phenomena and
can therefore be regarded as a form of positivistic research.
Although South African accounting academics are not following the international
developments in mainstream accounting research, in the experience of the writer a
major part of accounting research in South Africa follows the positivistic framework in
its belief in an objective reality, and it does this through the use of content analysis
(archival research) and structured closed questionnaires. The actual extent of the
difference between the application of the positivistic research by South African
academics and that in international accounting research is an opportunity for further
research.
INTERPRETATIVE FRAMEWORK
Its nature
The interpretative framework is based on the premise that social actors construct the
truth that is relevant in any given timeframe. Truth in the social environment is seen as
not fixed, but subject to change over time. Ryan et al. (2002:42) confirm this view:
The starting point for interpretative (and also for critical) research is the belief that
social practices, including management accounting, are not natural phenomena; they
are socially constructed. Consequently, they can be changed by the social actors
themselves.
3Marx also used financial statements as archival research data.
D Coetsee 89
Interpretative research therefore moves away from the objective physical realism of
positivistic research towards a social realism in which human actors play a significant
role. Parker (2008:911) explains this social nature of interpretative research as follows:
Ours is a concern to capture actors’ understandings from the inside, penetrating and
capturing multiple constructed realities. Ours is a pursuit and celebration of
complexity, depth, detail, richness, texture and meaning (Ahrens and Dent, 1998). We
grapple with a complex world of culture, language, stories, perception, cognition,
social conventions, politics, ideology and power. We seek to experience, understand
and critique structures and processes of accounting and accountability. We offer to
distinguish latent from manifest, and to better understand and reconstitute what we
thought we already know.
The view of positivists that relationships between data variables can be positively
captured in models is replaced by a view that data variables change through human
activity. Interpretative research is thus bound to the timeframe in which it is performed.
Interpretative research does not always predict future behaviour, although it could be an
indication of possible future behaviour. The same study at a later stage might provide
different results due to the changing nature of social practice. However, this could also
be said about most positivistic research in accounting. For instance, if a relationship
between data variables has been positively verified, this does not imply that at a later
stage the same result will be achieved. Many of the data variables on which positivistic
research in accounting is based are created through human invention and therefore also
change over time. The different nature of the interpretative theory of knowledge is
explained as follows:
Knowledge is constructed not only by observable phenomena, but also by descriptions
of people’s intentions, beliefs, values and reasons, meaning making and self-
understanding. Interpretivist knowledge is dispersed and distributed. The researcher
has to look at different places and at different things in order to understand a
phenomenon (Henning et al., 2004:20).
Important aspects of interpretative research are identified by Henning et al.
Interpretative research not only describes the phenomena, but goes further and assesses
the human influences and reasons for the phenomena. These social truths are more
subjective, and therefore a fundamental principle of interpretative research is that the
researcher may use different means to confirm the outcome of the research. This can
take the form of questionnaires, interviews, case studies or even focus groups.
Furthermore, since interpretative research incorporates the perceptions and feelings of
people, it is normally qualitative in nature. Qualitative research forms the foundation of
interpretative research. Parker (2008:909) refers to “a corpus of qualitative,
interpretative, interdisciplinary research in accounting literature”.
Interpretative research can be used to research different aspects of accounting practice,
for instance the reasoning behind establishing accounting standards, the application of
the standards in practice and for other broader aspects of accounting practice
incorporating theories of other disciplines. Armstrong (2008:878) summarises his
thoughts as follows:
It will probably now be obvious that I think that interpretative research has a great deal
to contribute to our understanding of how accounting is actually performed. Despite
90 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
three decades of critical accounting research, we still know remarkably little about how
accountants go about production of accountings, how the performance of audit is
carried off, how standards and reporting conventions are used in practice, and so on.
How, otherwise, can Enron have come as a surprise? Answers to questions of this kind
– not perhaps in the sense of formal theory but in the sense of grasping through
indicative cases how the culture of professional accounting operates – are, I suspect,
intimately related to the recurrent crises and scandals mentioned in the first part of this
essay. Because of economic power wielded through the medium of accounting,
understanding of this kind may turn out to be the most practically consequential
knowledge of accounting we can obtain. That, I suppose, amounts to a ‘cal-for’ of my
own.
As a starting point to the current debate regarding the application of interpretative
research in accounting research, in a debate conducted in 2006, accounting researchers
discussed the appropriateness of interpretative research as applied to accounting “… to
establish a more positive identity for ourselves as scholars interested in interpretation of
accounting practices” (Ahrens; Becker; Burns; Chapman; Granlund; Habersam;
Hansen; Khalifa; Malmi; Mennicken; Mikes; Panozzo; Piber; Quattrone & Scheytt,
2008). However, this was only a debate and no definition of interpretative research was
established.
Following the debate, the response of different accounting researchers to the future of
interpretative research is worth noting. Scapens (2008:915) argues that a wider
accumulation of accounting knowledge is needed. The argument is that interpretative
research can create knowledge about accounting practice that cannot be objectively
verified through mainstream accounting research. Cooper (2008:839) states that
interpretative accounting research “can be an instrumental vehicle for the career
aspirations of academics”. Accounting academics do not need to be followers of the
practice of mainstream accounting research, but they need to be knowledgeable about
all the research frameworks to develop a broader research focus. Cooper (2008:839)
declares that interpretative accounting research “can also be an approach that
encourages critical refection about our purpose and methods”. The argument is once
again that accounting researchers must evaluate whether the methods of mainstream
accounting research are appropriate for any particular research problem.
Therefore, certain accounting academics agree that due to the changing nature of
accounting practice, interpretative accounting research should play an important role in
accounting research. Wilmott (2008:920) expresses the central question regarding the
future of interpretative research as follows:
… whether proponents of IAR can successfully carve out and defend a space between,
on the one side, mainstream researchers who dismiss their work as methodologically
unsound and, on the other side, critical researchers who regard the orientation of
interpretative researchers as insufficiently attentive to the interpenetration of
knowledge and power (in the form of subjectification as well as exploitation and
domination) and as a consequence, may gain some measure of academic or practitioner
respectability but at the cost of generating forms of accounting knowledge that are
assessed to be unacceptably facile and conservative.
Interpretative research should have a place between mainstream and critical research. It
seeks a deeper understanding of accounting practice. The role that human interventions
D Coetsee 91
play in developing accounting practice is central to this research framework.
Interpretative research includes the views, beliefs, values and reasoning of the various
role-players. For this form of research, the question is why human beings act in certain
ways. This moves research from an objective realism to a social realism, without
including the subjective view of the researchers of critical research.
The process
Because of the social realism of interpretative research, its research process is not as
structured as that for positivistic research. Since it is not easy to establish the
perceptions and feelings of people, interpretative research sometimes uses more than
one instrument to confirm the validity of the research. Henning et al. (2004:20)
therefore state that:
Interpretative knowledge is dispersed and distributed. The researcher has to look at
different places and at difference things in order to understand a phenomenon.
Incorporating the feelings and beliefs of people and looking at different places to
understand the phenomenon makes interpretative research more difficult to apply.
Mennicken (2008:848), however, believes that this could be positive by stating:
It gives us the freedom to choose and work with different qualitative methods and social
theories that highlight different aspects of accounting practice regarding its role, uses
and limits in everyday life.
The research process in interpretative research is an individual decision of each
researcher. The validity of interpretative research is based on the appropriateness with
which the process is explained and applied to create a valid research output. In order to
create this freedom and to structure interpretative research, many academics are calling
for the establishment of a conceptual framework, which will provide coherence to the
research (Baker & Bettner, 1997:305; Trafford & Leshem, 2008:840). The conceptual
framework can be seen as a supplement to the more structured approach of positivistic
researh. Miles and Huberman (1984:28) describe a conceptual framework as follows:
A conceptual framework explains, either graphically or in a narrative form, the main
dimensions to be studied – the key factors, or variables – and the presumed relationship
among them. Frameworks come in several shapes and sizes. They can be rudimentary
or elaborate, theory-driven or commonsensical, descriptive or causal.
The conceptual framework in essence explains the different sections of the research and
how each section contributes to a sound and verifiable conclusion. The data used in
interpretative research is much more subjective; therefore validity of interpretative
research is created through the design process, which is explained in the conceptual
framework. The conceptual framework links the literature review, the research
methodology, the actual research and the research results together. Regarding the
literature review Trafford and Leshen (2008:87-88) state the following:
We have argued that this corpus is the foundation from which to generate the
theoretical perspectives that you need to develop on your topic and methodology. These
perspectives represent the conceptual foundation for you to devise your research design
and, later, give conceptual focus to your conclusion.
92 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Similarly, each section of the research process contributes to the validity of the
research. The objective of explaining the conceptual framework is to bring all aspects
together. Baker and Bettner (1997:305) go further and refer to “a framework for
conceptualizing and designing interpretative and critical research”. A conceptual
framework is thus needed for both interpretative and critical research.
Different research instruments can be used to interpret the data. Research instruments,
such as questionnaires, interviews and case studies, are used in interpretative or
qualitative research (Moll, Major & Hoque, 2006). Sometimes more than one research
instrument is used, in order to create or multiple perspectives on the phenomena
(Henning et al., 2004:23). This means, for example, that both questionnaires and
interviews could be used, or use could be made of different focus groups. Henning et al.
(2004:20) summarised the methodological implications as follows:
Unstructured observations, open interviewing, idiographic descriptions and qualitative
data analysis are all ways to capture “insider” knowledge that is part of an
interpretative methodology.
In South Africa limited articles refers to interpretative research, although interpretative
methodologies are used. Steenkamp, Baard and Frick (2009) refer to their research as
“focused within an interpretative paradigm”. They specifically state that their
questionnaires contain quantitative and qualitative elements. In the qualitative section
the study focuses on students own perceptions. They further state that “The research
design and the methodology used aimed to make the study replicable and therefore
reliable” (2009:120). De Wet and Van Niekerk (2001) provided students in a first-year
course in accounting with an opportunity to make suggestions for improvements.
Specifically no limit was place on the number of suggestions a student could make.
South African interpretative researchers also created conceptual frameworks to ground
their research. Sartorius, Eitzen and Nicholson (2006) creates a conceptual framework
to assess the appropriateness of performance measurement in a case study. Wolmarans
(2001) used 10 open-ended questions to record students’ experience of flexible mode of
learning by creating a theoretical framework for and assumptions of this study.
Insider or deeper knowledge, as stated before, is more difficult to research. The essence
of such interpretative research is that each researcher must thoroughly develop and
explain the relevant research design to establish the validity of the research. This is
normally created by identifying a conceptual framework. Experience in research
methods and structuring of research is needed to appropriately apply the interpretative
research framework.
CRITICAL RESEARCH FRAMEWORK
Its nature
The critical research framework is based on the premise that social actors can change
the world and that they do not need to accept what is presented as “truth”. Truth is
relative and therefore open to criticism. This is the justification for critical research.
Critical research involves the views of the researcher and moves from relative
objectivity to subjectivity. Henning et al. (2004:22-23) describes the critical research
framework as follows:
D Coetsee 93
A critical framework is the “undoing” of the positivistic/postpositivistic, objectivist
paradigm. It is essentially a process of deconstructing of the world. Whereas
interpretivists construct our world by means of multiple perspectives, critical theory
questions the political nature of that very process, maintaining that some relationships
in the world are more powerful than others, that some “intellectual currency” is worth
more than others. Researchers using critical theory aim at promoting critical
consciousness and breaking down the institutional structures and arrangements that
reproducing oppressive ideologies, and the social inequalities that are produced,
maintained and reproduced by these social structures and ideologies.
Roslender (2006:250) agrees with Henning et al. that critical research is fostering
change in the larger content of society by stating:
Critical theory is intimately wedded to change. More specifically, it is concerned with
the promotion of a better society, one in which the prevailing social arrangements serve
the interest of the mass of people, whose ‘potentialities’ are perceived to be constrained
by those arrangements already in place.
The essence of critical research is change of the social order in which the current
practice operates. Critical research is thus more interdisciplinary, since no discipline is
totally isolated from other disciplines. This broader aspect as applied to accounting is
confirmed by the following quotes:
In broad terms, ‘critical accounting theory’ is used to refer to an approach to
accounting research that goes beyond questioning whether particular methods of
accounting should be employed, and instead focuses on the role of accounting in
sustaining the privileged position of those in control of particular resources (capital)
while undermining or restraining the voice of those without capital (Deegan, 2010:529).
… the critical accounting project challenges the manner in which accounting has
conventionally privileged technical issues and knowledges over those demonstrating
that the accounting is not created in a social vacuum and as a result of which much of it
may be highly contestable. The emergence of a complementary set of critical insights is
now widely recognised as a sign of accounting’s maturity as an academic discipline, as
well as providing a basis for developing accounting as a set of socially responsible
practices (Roslender, 2006:247).
This integrated view of the critical research framework makes this choice more
complex. The research must firstly understand the wider application of the discipline,
but must also understand the theoretical base on which the critical research is based.
Critical research has its roots in a socialist or Marxist stance (Roslender, 2004:248;
James, 2008:643 & 676). Broadbent (2002) questions the appropriateness of this stance
and calls for different stances in accounting research. Feminism, for instance, calls for a
particular stance (2002:435). Deegan (2010:552) states that much critical accounting is
grounded in political economic theory. In this regard, Deegan articulates a political
economic theory view as follows:
The view is that society, politics and economics are inseparable, and economic issues
cannot meaningful be investigated in the absence of considerations about the political,
social or institutional framework in which economic activity takes place (2010:534).
94 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
In Roslender’s remarks quoted above, after Deegan, Roslender creates the impression
that accounting researchers are mature in the application of critical insights. However,
the question is whether other frames could be developed for critical research in
accounting, such as a fair value, mixed measurement or historical cost approaches. This
view holds that critical research will prosper in accounting research only if other frames
of reference for the basis of the critique are developed in accounting research. Although
Roslender (2006:264) states that there is no limit to the application of critical research,
critical research is not easy. Apart from creating the theoretical framework in which the
critical research is based, the researcher must also be skilled in the concept of critical
writing.
The process
No descriptive layout is provided for critical research. The research process must be
created through an appropriate research design. An explanation of the conceptual frame
in which the critical research is performed is crucial and must be explained in critical
research. This will include both the critical perspective and the deductive process
followed to deliver the critique. Baker and Bettner (1997:306) make the following
comments regarding the design of interpretative and critical research studies:
The researcher is given latitude to evaluate his or her view of the social world, define
his or her epistemological stance, and select an appropriate methodological approach
consistent with his or her social views and epistemological grounding. An important
issue that often arises concerning the design of interpretative and critical research is
whether it is appropriate to use a mixture of approaches in the same study. Some argue
that certain methods are only appropriate to certain perspectives (for insight into this
debate see Burrell & Morgan, 1979). Others have advocated a blend of research
perspectives involving both qualitative and quantitative methods and interpretative and
critical approaches (Miles & Huberman, 1984; Kirk & Miller, 1986). These blendings
may merely be extensions of, and not alternatives to mainstream perspectives and
therefore should be viewed with caution (Tinker, 1995). We argue that this is important
that the researcher’s perspective be made clear so that the reader can understand and
appreciate the context in which the research is being approached. It is this frank
admission of the researcher’s perspective which is virtually absent from mainstream
accounting research.
This extract may create the impression that the research process of interpretative and
critical research is in essence the same. However, each approaches research from a
different stance: a neutral or a critical stance. Interpretative research interprets data,
therefore using certain research instruments or means to capture the meaning behind the
data. Critical research incorporates the views of the researcher and is in many instances
much more theoretical in nature. As with interpretative research, a theoretical
framework is needed to explain the basis and process of critique.
More practical, theoretical work is included in accounting research journals in South
Africa. Examples are Van Heerden (2009); Du Toit (2008) Van der Schyf (2008);
Sascho and Oberholster (2005). However, these research articles are based more on
interpreting the existing theory than providing a critique based on an underlying
theoretical perspective. Real critical research, based on identifying the critical stance, is
not seen in South African accounting journals.
D Coetsee 95
COMPARISON OF RESEARCH FRAMEWORKS
This debate illustrates that there are fundamental differences between the different
research frameworks. However, a more recent development is that the borders of the
frameworks are starting to merge, and that any combination of the research frameworks
can be applied provided that the research design or conceptual framework on which the
research is based is adequately explained in the research. Table 1 sets out the
differences between the different frameworks as a reference table to evaluate the nature
of the research.
Table 1: Comparison of research frameworks
Positivistic Interpretative Critical
Premise One truth Social actors create
truth
Change the truth
Theory Develop and test theory Develop theory Critique theory
Main focus Verification and
measurability
Understanding the
reasoning behind human
activity
Basis of critique
existing social
practice
Nature Quantitative/Objective Qualitative/Objective Critical/Subjective
Research Empirical/Statistical Empirical/Theoretical Theoretical
The premise of truth in each research framework differs. The premises move from the
objective view that only one truth exists (Chua, 1986:606; Kirk & Miller, 1986:14) – a
natural science view – to the assumption that social actors can change the truth
(Henning et al., 2004:22-23; Roslender, 2006:250) – a totally subjective view. In the
middle, interpretative thinking believes that social actors create certain truths (Ryan et
al., 2002:42) and that researchers objectively report on those truths. The application of
positivistic approaches to a social discipline, such as accounting, creates the notion that
in the changing world of a social discipline certain practical truths can still be
objectively verified (Armstrong, 2008:871).
The nature of the development of theory also differs in each research framework.
Ultimate positivistic research tests the hypothesis or models identified (Inanga &
Schneider, 2005:30), or verifies research objectives, as discussed previously in terms of
archival research or structured closed questionnaires. Any hypothesis, model or research
objective not rejected is considered to be the basis for modification of existing theory
and/or development of new theory (Deegan, 2010:260). Interpretative research through
its interpretative process contributes to the understanding of a phenomenon (Armstrong,
2008:878). In the process new knowledge is created of the existing practice, which
contributes to the pool of knowledge (Henning et al., 2004:20). Grounded theory is for
instance a methodology applied in interpretative research to create theory from data
(Glaser, 2004; Strauss & Corbin, 1990). Critical research questions the appropriateness
of current theory, with the objective of changing the existing social structure (Henning
et al., 2004:22-23; Roslender, 2006:250).
The differences between the research frameworks amount to a shift of focus. Positivistic
research focuses on the objective verification or measurement of the phenomena
researched (Armstrong, 2008:871). Gaffikin (2006) states that the use of new
information processing technologies (especially computers) and the growing availability
of large scale stock market data bases have increased the use of the positivistic
96 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
framework. Interpretative research focuses more on understanding the human reasoning
behind perceived truths (Henning et al., 2004:20) and therefore the methodology or
frame developed to interpret the research is crucial (Baker & Bettner, 1997:305;
Trafford & Leshem, 2008:840). Critical research focuses on the frame of reference used
to provide the critique to foster social change (Roslender, 2004:248; Boardbent, 2002).
Since a positivistic research goal is to objectively verify or measure the given
circumstances (Armstrong, 2008:871), the research is mainly quantitative (Henning et
al., 2004:18). However, qualitative means can also be used to describe practices
(Henning et al., 2004:18). In this instance the practice is described through qualitative
means without stepping into the realm of interpretation (Henning et al., 2004:18).
Interpretative research focuses on more human aspects and is thus much more
qualitative in nature (Parker, 2008:909). As a rule, statistical means are not used to
verify the data. This article argues that verification in interpretative research is created
through the appropriateness of the research process. Critical research is more subjective
and is based on the theoretical perspectives of the researcher (Henning et al., 2004:23).
Both positivistic research and interpretative research normally assess data through
empirical means, but capture the data in order to provide different outcomes. Positivistic
research analyses the data captured through empirical means by using statistical
analysis to either accept or reject the hypothesis, models or research objectives created
(Inanga & Schneider, 2005:230). Interpretative research also normally uses empirical
means, but uses it differently to assess deeper knowledge that is “dispensed and
distributed” (Henning et al., 2004:20). Researchers can also interpret literature and
documents to obtained deeper understanding, so interpretative research can be
theoretical. Critical research is normally theoretical due to the subjective nature of the
research; however, Roslender (2006:265) states that there is no reason why using a
critical perspective cannot include empirical work, but critical research does not require
the collection of empirical data.
CONCLUSION
This article argues that there is a place for each of the different research frameworks in
accounting research. Collectively all of these frameworks contribute to the pool of
accounting knowledge. To contribute optimally to the pool of accounting knowledge,
accounting scholars need to be trained in all three of the research frameworks. The
application of the appropriate research framework is determined by the research
question and the nature of the research issue that are to be resolved. Training in one
research framework can result in the inappropriate application of a research framework
to a research problem.
The main issue with the application of a positivistic research framework in mainstream
accounting is that it objectifies a discipline that is changing in nature. Social scholars
have confirmed that a positivistic framework can be applied in a social discipline such
as accounting provided the outcome can be objectively measured and verified. Valuable
relations between certain data variables, such as share prices and accounting data, are
created through multivariate modelling in positivistic mainstream accounting research
and should always play an important role in accounting research. Regrettably, South
African accounting journals have made only a limited contribution to mainstream
accounting research. South African academics generally deploy a positivistic
framework by using archival data analysis and structured closed questionnaires, without
D Coetsee 97
accepting and rejecting hypotheses. Further research could be done to establish how
research in South African accounting journals differs from mainstream accounting
research with its positivistic focus.
Many accounting academics are calling for a move away from mainstream accounting
research to other research frameworks to develop a greater pool of accounting
knowledge. These academics are distinguishing more and more between interpretative
and critical research. Both are research methods that could be applied to break the
stranglehold of mainstream accounting research.
Interpretative research has been developed specifically to assess a changing
environment such as that of accounting. The international movement to debate and
comment on the application of this research framework in accounting is a positive
development. Currently, sufficient academic material on the application of
interpretative research in accounting is available. The application of interpretative
research in accounting could be further enhanced through the proper training of
inexperienced researchers. The focus of such training should be on the development of
appropriate conceptual frameworks to explain the total research process and to create
academic validity for interpretative research. A question that could be research further
is whether accounting scholars in South Africa are sufficiently trained, through master’s
and doctoral methodology courses, to apply interpretative research appropriately.
Critical research questions the nature of existing practice, with the objective of fostering
change. A fundamental issue in critical research in the field of accounting is whether
sufficient critical frames of references have been developed to perform critical research
in accounting. The use of a Marxist or socialist stance is not the only critical stance.
Other stances, such as feminist and political economic theory, are also available. In
South African journals, as in many international ones, theoretical work is published.
However, theoretical frameworks creating the critical stance on which theoretical
research is based and social integration of appropriate critical research are lacking in
South African research. Most South African theoretical work is based on the views
expressed by the researcher or on a review of existing thoughts of others.
Notwithstanding a lack of proper grounding in different research frameworks,
accounting research in South Africa is in a growth phase and it is to be hoped that South
African accounting research will yet make a significant contribution to future
international research in accounting. This article provides merely an overview of the
nature and process of the different social research frameworks. Further research or
debate could challenge the shortcomings and limitations of each research framework.
Further research could also be conducted to empirically assess the nature of different
research approaches or frameworks that are applied in South African accounting
journals. International different researcher has classified the nature of research in
journals (Luft and Shields, 2003; Carmona, Gutiérrez and Cármona, 1999; Prather and
Rueschhoff, 1996; Shields, 1997), which could be applied in a South African content.
REFERENCES
Ackers, B. (2009). Corporate social responsibility assurance: how do South African
publicly listed companies compare? Meditari Accounting Research. 17(2):1-17.
98 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Ahrens, T.; Becker, A.; Burns, J.; Chapman, C.S.; Granlund, M.; Habersam, M.;
Hansen, A.; Khalifa, R.; Malmi, T.; Mennicken, A.; Mikes, A.; Panozzo, F.; Piber, M.;
Quattrone, P. and Scheytt, T. (2008). The future of interpretative research−A
polyphonic debate. Critical Perspectives on Accounting, 19:840-866.
Armstrong, P. (2008). Calling out for more: Comment on the future of interpretive
accounting research. Critical Perspectives on Accounting, 19:876-879.
Baker, C.R. and Bettner, M.S. (1997). Interpretive and critical research in accounting: A
commentary on its absence from mainstream accounting research. Critical Perspectives
on Accounting, 8:293-310.
Barth, M.E.; Beaver, W.H.; Hand, J.M and Landsman, W.R. (1999). Accruals, Cash
Flows, and Equity Values. Review of Accounting Studies, 3:205-229.
Barth, M.E.; Beaver, W.H. and Landsman, W.R. (2001). The relevance of the value
relevance literature for financial accounting standard setting: another view. Journal of
Accounting and Economics, 31:77-104.
Beaver, W.H. (2002). Perspectives on recent capital market research. The Accounting
Review, 77(2):453-474.
Boland, L.A. and Gordon, I.M. (1992). Criticizing positive accounting theory.
Contemporary Accounting Research, 9(1):142-170.
Bradfield, D.J. (1998). Measuring the selection and timing abilities of South African
fund managers. SA Journal of Accounting Research, 12(1):1998.
Broadbent, J. (2002). Critical accounting research: A view From England. Critical
Perspectives on Accounting, 13:433-449.
Carmona, S.; Gutiérrez, I. and Cámara, M. (1999). A profile of European accounting
research: evidence from leading research journals. The European Accounting Review,
8(3):463-480.
Chambers, R.J. (1993). Positive accounting theory and the PA cult. ABACUS, 29(1):1-
26.
Charlton, G. and Marx, B. (2009). An investigation into the impact of continuous
auditing on the external auditors of the four largest banks in South Africa. SA Journal of
Accounting Research, 23(1):45-65.
Chivaka, R.; Siddle, A.; Bayne, L.; Cairney, C. and Shev, J. (2009). Reasons for share
repurchases in South Africa: Theory versus practice. SA Journal of Accounting
research, 23(1):1-30
Chua, W.F. (1986). Radical Developments in Accounting Though. The Accounting
Review. 16(4): 601-632.
Coetsee, D. (2010). The role of accounting theory in the development of accounting
principles. Meditari Accounting Research, 18(1):1-16.
D Coetsee 99
Cooper, D.J. (2008). Editorial: Is there a future of interpretative accounting research?
Critical Perspectives on Accounting, 19:837-839.
De Clercq, B. and Venter, J.M.P. (2009). Factors influencing a prospective chartered
accountant’s level of financial literacy an exploratory study. Meditari Accountancy
Research, 17(2):47-60.
Deegan, C. (2010). Financial Accounting Theory. North Ryde: McGraw-Hill.
Deegan, C. and Unerman, J. (2006). Financial Accounting Theory (European edition).
Berkshire: McGraw-Hill.
Dechow, P.M.; Hutton, A.P. and Sloan, R.G. (1999). An empirical assessment of
residual income valuation model. Journal of Accounting and Economics, 26:1-34.
De Wet, J.H. and Van Niekerk (2001). An innovative approach to accounting education
at the first-year level. Meditari Accounting Research, 9:93-108.
Du Toit, E. (2008). Characteristics of companies with a higher risk of financial
statement fraud: A survey of the literature. SA Journal of Accounting Research,
22(1):19-44.
Firer, S. & Stainbank, L. (2003). Testing the relationship between intellectual capital
and a company’s performance: Evidence from South Africa. Meditari Accountancy
Research, 11:25-44.
Gaffikin, M.J.R. (2006). The Critique of Accounting Theory. Accounting & Finance
Working Paper 06/25, School of Accounting & Finance, University of Wollongong.
Available from: http://ro.uow.edu.au/accfinwp/41. (Assessed 10 March 2009).
Glaser, B.G. (2004). Remodeling Grounded Theory. Forum: Qualitative Social
Research, 5(2).
Granhof, and Zeff (2008). Research on Accounting Should Learn From the Past.
Chronicle of Higher Education, 54(28):A34.
Henning, E.; Van Rensburg, W. and Smit, B. (2004). Finding your way in qualitative
research. Van Schaik Publishers: Pretoria.
Inanga, E.L. and Schneider, W.B. (2005). The failure of accounting research to improve
accounting practice: a problem of theory and lack of communication. Critical
Perspectives on Accounting, 16:227-248.
James, K. (2008). A Critical Theory and Postmodernist approach to teaching of
accounting theory. Critical Perspectives on Accounting. 19:643-676.
Kenton, K.Y. (2006). Earning Quality and the Equity Risk Premium: A Benchmark
Model. Contemporary Accounting Research. 23(3):833-877.
Kirk, J. and Miller, M.L. (1986). Reliability and validity in qualitative research.
Beverly Hills: Sage Publications.
100 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Kothari, S.P. (2001). Capital market research in accounting. Journal of Accounting and
Economics, 31:105-231.
Luft, J. and Shields, M.D. (2003). Mapping management accounting: graphics and
guidelines for theory consistent empirical research. Accounting, Organization and
Society, 28 (2003):169-249.
Marx, B. (2009). An analysis of audit committee responsibilities and disclosure
practices at large listed companies in South Africa. SA Journal of Accounting Research,
23(1):31-44
Macintosh, N.B. (2004). A comment on “Recovering Accounting”. Critical
Perspectives on Accounting, 15:529-541.
Meek, G.K. and Thomas, W.B. 2004. A review of market-based international
accounting research, Journal of International Accounting Research. 3(1):21-41.
Mennicken, A (2008). (In Ahrens, T.; Becker, A.; Burns, J.; Chapman, C.S.; Granlund,
M.; Habersam, M.; Hansen, A.; Khalifa, R.; Malmi, T.; Mennicken, A.; Mikes, A.;
Panozzo, F.; Piber, M.; Quattrone, P. and Scheytt, T. The future of interpretative
research−A polyphonic debate. Critical Perspectives on Accounting, 19:840-866).
Miles, M.B. and Huberman, A.M. (1984). Qualitative Data Analysis: A Sourcebook of
New Methods. Beverly Hills: Sage Publications.
Moll, J.; Major, M. and Hoque, Z. (2006). The qualitative research tradition. 375-398
(In: Hoque, Z. (2006). Methodology issues in accounting research: Theories and
Method. London: Spiramus).
Nel, G.F. and Steenkamp, L.P. (2008). An exploratory study of Chartered accountants’
understanding of XBLR. Meditari Accounting Research, 16(1):79-93.
Ohlson, J.A. (1995). Earnings, Book Values and Dividends in Equity Valuations.
Contemporary Accounting Research. 11:661-687.
Ohlson, J.A. (2005). On Accounting-Based Valuation Formulae. Review of Accounting
Studies, 10:323-347.
Parker, L.D. (2008). Interpreting interpretative accounting research. Critical
Perspectives on Accounting, 19:909-914.
Piber, M. (2008) in Ahrens, T.; Becker, A.; Burns, J.; Chapman, C.S.; Granlund, M.;
Habersam, M.; Hansen, A.; Khalifa, R.; Malmi, T.; Mennicken, A.; Mikes, A.; Panozzo,
F.; Piber, M.; Quattrone, P. and Scheytt, T. (2008). The future of interpretative
research−A polyphonic debate. Critical Perspectives on Accounting, 19:840-866.
Prather, J. and Rueschhoff, N. (1996). An analysis of international accounting research
in U.S. academic accounting journals, 1980 through 1993. Accounting Horizons,
10(1):1-17.
D Coetsee 101
Reiter, S.A. and Williams, P.F. (2002). The structure and progressivity of accounting
research: The crises in the academy revisited. Accounting, Organizations and Society,
27:575-607.
Roslender, R. (2006). Critical Theory. 247-269. (In: Hoque, Z. (2006). Methodology
Issues in Accounting Research: Theories and Method. London: Spiramus).
Ruddy, T. and Everingham, G.K. (2008). Creative accounting, accounting errors, and
the ability of users to detect and adjust for them. SA Journal of Accounting Research,
22(1):45-95.
Ryan, B; Scapens, R.W. and Theobald, M. (2002). Research Method and Methodology
in Finance and Accounting. London: Thomson Learning.
Sartorius, K.; Eitzen, C. and Nicholson (2006). The appropriateness of performance
measurement in the service sector: A case study. South African Journal of Accounting
Research, 2006:27-50
Sascho, Z.Y. and Oberholster, J.G.. (2005). Should the IASB consider exercise date
accounting for outstanding employee share options? Meditari Accountancy Research,
13(2): 89-106.
Scapens, R.W. (2008). Seeking the relevance of interpretative research: A contribution
to the polyphonic debate. Critical Perspectives on Accounting, 19:915-919.
Shields, M.D. (1997) Research in Management Accounting by North Americans in the
1990s. Journal of Management Accounting Research, 9:3-61
Stainbank, L.J. (2009). Working in teams: improving the team experience. Meditari
Accountancy Research, 17(1): 69-80
Steenkamp, L.P.; Baard, R.S. and Frick, B.L. (2009). Factors influencing success in
first-year accounting at a South African university: A comparison between lecturers’
assumptions and students’ perceptions. South African Journal of Accounting Research,
23(1):113-140.
Strauss, A. and Corbin, J. (1990). Grounded theory research: Procedures, canons, and
evaluation criteria. Qualitative Sociology, 13(1):3-21.
Swartz, G.E. and Negash, M. (2006). An empirical examination of the Ohlson (1995)
model. SA Journal of Accounting Research, 20(1):67-82.
Swartz, G.E.; Swartz, N-P and Firer, S. (2006). An empirical examination of the value
relevance of intellectual capital using the Ohlso (1995) valuation model. Meditari
Accountancy Research, 14(2):67-81.
Trafford, V. and Leshem, S. (2008). Stepping Stones to Achieving your Doctorate.
Berkshire: McGraw-Hill.
102 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Van der Schyf, D.B. (2008). The essence of a university and scholarly activity in
accounting, with reference to a Department of Accounting at a South African
University. Meditari Accountancy Research, 16(1):1-26.
Van Heerden, M. (2009). CFC translation rules: is the taxpayer currently getting the
short end of the stick? Meditari Accountancy Research, 17(2):151-165.
West, A. (2006). A Commentary on the global position of South African Accounting
Research. Meditari Accountancy Research, 14(1):121-137.
Williams, P.F. (2009). Reshaping accounting research: living in the world in which we
live. Accounting Forum, 33(4):274-279.
Wilmott, H. (2008). Listening, interpreting, commending: A commentary on the future
of interpretive accounting research. Critical Perspectives on Accounting, 19:920-925.
Wolk, H.I.; Dodd, J.L. and Rozycki, J.J. (2008). Accounting Theory. Thousand Oaks,
California: Sage Publications, Inc.
Wolmarans, H.P. (2001). Postgraduate education in taxation by means of a flexible
learning model: A new education paradigm. Meditari Accounting Research, 9:333-349.
B Marx & V van Dyk 103
Sustainability reporting at large public sector
entities in South Africa
B Marx
Department of Accountancy, University of Johannesburg
V van Dyk
Department of Accountancy, University of Johannesburg
Received: March 2011 SAJAR
Revised: August 2011, September 2011 Vol 25 No. 1
Accepted: October 2011 2011
pp.103 to 127
Public sector entities are central to the delivery of sustainable development, and every aspect of
their role shapes how people live their lives. This, together with the fact that they are under
constant and growing pressure to operate in an environmentally friendly and sustainable manner,
and to show their stakeholders that they provide good value for money, highlights the need for
open, honest and reliable sustainable reporting to public sector stakeholders. The objective of
the paper is twofold: it aims, firstly, to provide a brief overview of sustainability and
sustainability reporting and how it impacts on public sector entities, and, secondly, to provide
evidence regarding the sustainability reporting practices at large public sector entities in South
Africa. This is done through a literature review of current corporate governance and
sustainability developments in the private and public sector, supported by empirical evidence
obtained from assessing, through content analysis, the sustainability reporting of the largest
public sector entities in South Africa. The study found that sustainability reporting by public
sector entities is still in its infancy, and that the reporting by the largest public sector entities in
South Africa ranges from excellent to sub-standard.
KEY WORDS
Corporate citizenship; Corporate governance; Public sector; Sustainability; Sustainability
reporting; Public sector sustainability reporting
Contact
INTRODUCTION
The development of corporate governance over the years has given prominence to the
need for organisations to do business in a responsible manner and respect the society
and environment in which they function. Add to this the events of the recent financial
crisis, environmental disasters such as the extreme droughts, earthquakes, hurricanes
and tsunamis, diseases like HIV/Aids, and the deplorable phenomenon of child labour,
and it is obvious that sustainability has become an inescapable and key component of
modern business (Ackers, 2009:1, 4-5; De Villiers, 2004; Engelbrecht, 2010; Ernst &
104 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Young 2008a, 2008b; Institute of Directors (hereafter IoD), 2009:11-14; KPMG, 2008;
Piketh, 2010; Terry, 2008:23-58; Trialogue, 2009:1-5). There has also been a shift in
emphasis over the years from profits for shareholders only to the recognition that
organisations have a responsibility to those who give them their licence to operate,
namely their stakeholders (Ackers, 2009:2; King, 2006:21; Marx, 2008:334; Mitchell,
Curtis & Davidson, 2008:67–69; Naidoo, 2009:246–250; Sustainability South Africa,
2011a: Terry, 2008:201-202). This has resulted in the predominantly financial focus of
the past expanding into a wider and more inclusive approach of doing business in the
future – in a moving away from the single bottom line (that is, profits at any cost) to a
triple bottom line that embraces the economic, environmental and social aspects of a
company’s activities (Ackers, 2009:2; Da Piedade & Thomas, 2006; Jhaveri, 1998;
King, 2006:20-22; Marx, 2008:334-336; Mitchell et al., 2008:67-69; O’Carrol, 2009;
Paris 2006:3; Pearce & Doh, 2005:30; Sustainability, 2011a; Terry, 2008:3-14).
The third King Report on Corporate Governance for South Africa (hereafter King III) is
of the opinion that “Sustainability is the primary moral and economic imperative of the
21st century and one of the most important sources of both opportunities and risks for
businesses”. It further states that companies need to be, and to be seen to be, responsible
corporate citizens – that is, protecting, enhancing and investing in the well-being of the
economy, society and the natural environment in which they do business (IoD,
2009:11–12). Against this background it is evident that there is a need for relevant,
accurate, reliable and credible stakeholder reporting by organisations not only on their
economic performance, but also as regards their social, environmental and governance
performances and achievements (Dawkins & Ngunjiri, 2008:286-289; de Villiers,
2004:21-31; Global Reporting Initiative (hereafter GRI), 2006:2-3; Mammatt, 2010:3-5;
Mammatt, Marx & Van Dyk, 2010:22-24; Mitchell et al., 2008:67-70; Naidoo,
2009:248; Tregidga & Milne, 2007:4-6). Companies that issue sustainability reports
that are transparent and provide accurate and reliable information to the users thereof
will further increase the trust and confidence of their stakeholders and the legitimacy of
their operations (IoD, 2009:13; Dawkins & Ngunjiri, 2008:286-299; Mitchell et al.,
2008:67-69).
Whereas the concept of sustainability and sustainability reporting is reasonably well
established in the private sector (Hespenheide, 2010; Social Investment Forum, 2008:2-
3; Sustainability South Africa, 2011a; Terry, 2008), sustainability and sustainability
reporting in the public sector is still in its infancy (GRI, 2010; Leeson, Ivers &
Dickinson, 2005; Prizma, 2011). The worldwide financial crisis has emphasised the
urgent need for the public sector to adopt a sustainability agenda, and the sustainability
issue has been put at the forefront at the 2009 Copenhagen Climate Conference
(PricewaterhouseCoopers, 2011; Trialogue, 2010:4). Public sector bodies are central to
the delivery of sustainable development, and every aspect of their role shapes how
people live their lives. This, together with the fact that they are under constant and
growing pressure to operate in an environmentally friendly and sustainable manner, and
to show to their stakeholders that they provide good value for money, highlights the
need for open, honest and reliable sustainable reporting to public sector stakeholders
(Aristovnik, 2009; Gosling, 2010; Madden, 2010). Jon Sibson, government and public
sector leader of PWC, summarises the need for sustainability reporting by public sector
entities as follows: “Given governments’ worldwide commitment to promoting
sustainability and meeting environmental targets, it is evident that public sector bodies
should be taking the lead on sustainability reporting” (Sibson, 2009).
B Marx & V van Dyk 105
As is evident from the above, it is imperative for public sector entities to do business
responsibly and in a sustainable manner and to respect the rights of all stakeholders.
This is even more so for public entities that provide public services and receive
government assistance – hence the need for sustainability reporting by public sector
entities. This study found, however, that sustainability reporting worldwide, as well as
in South Africa, by public sector entities is still in its infancy and is not well supported.
The remainder of the paper is organised as follows. The next section presents the
objectives, scope and limitations of the study. The sections that then follow describe the
theoretical background of the paper, the methodology applied and the empirical findings
and deductions. Recommendations drawn from the study and areas identified for future
research are then provided, and conclusions presented in the last section.
OBJECTIVES, SCOPE AND LIMITATIONS
The objective of the paper is twofold: to provide a brief overview of the development of
sustainability and sustainability reporting; and secondly to provide evidence on the
sustainability reporting practices of large public sector entities in South Africa. This is
done by way of a literature review of current sustainability and public sector
developments and practices, and is supported by empirical evidence obtained from
assessing the sustainability reporting of the largest public sector entities in South Africa,
as defined by total asset value.
The study has three specific limitations. Firstly, the assessment is limited to the largest
public sector entities, and the findings might not necessarily be representative of the
sustainability reporting practices of smaller public sector entities. Secondly, the content
analysis techniques of annual and sustainability reports might have specific limitations.
Content analysis has limitations, such as the risk of capturing an incomplete picture of
the company’s business (as noted by Unerman, 2000:667), but is also a widely
recognised and accepted research instrument (Ackers, 2009:3; April, Bosma & Deglon,
2003:165-168; Barac & Moloi, 2010:20; Brennan & Solomon, 2008:893; Dawkins &
Ngunjiri, 2008:291-292; Mirfazli, 2008:388-391). The third limitation is that the study
was performed prior to the requirements of King III becoming effective, and therefore
excludes recommendations by King III regarding sustainability disclosure.
THEORETICAL BACKGROUND
Sustainability, sustainability reporting and corporate citizenship
Sustainability has been described as the primary moral and economic imperative of the
twenty-first century and one of the most important sources of both opportunities and
risks for businesses (IoD, 2009:11). Organisations are increasingly realising that they
are members of a wider community and must therefore behave in a responsible manner
by respecting the environment and society in which they operate and exist. Today’s
organisations are integral to society, and as such they are expected to behave, and be
seen to behave, as responsible corporate citizens – that is, protecting, enhancing and
investing in the well-being of the economy, society and the natural environment in
which they do business. Sustainability also makes business sense and is directly related
to shareholder and stakeholder value (Accenture, 2010:2, 10-12; Engelbrecht, 2009:4-5;
IoD, 2009:12; Mammatt, 2008:1-4; Naidoo, 2009:249-256; PricewaterhouseCoopers,
2008:1; Van Altena 2009:42). The recent financial crisis, as well as environmental
disasters and social rights abuses, has further emphasised the need for sustainable
106 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
business practices and corporate citizenship (Engelbrecht, 2010; Mammatt, 2010;
Trialogue 2009, Terry 2008).
Sustainability and sustainable development is a complex notion that seeks to reconcile
the goals of economic development and ecological and social well-being. The concept
of sustainable development originated from the United Nations World Commission for
Environment and Development Report (hereafter WCED), Our Common Future
(commonly referred to as the Brundland Report), which made far-reaching
recommendations regarding sustainable development, and which defined sustainable
development as “development that meets the needs of the present without
compromising the ability of future generations to meet their own needs” (Livesey,
2002:315-317; Naidoo, 2009:246; Terry, 2008:3-4; WCED, 1987:43). Over the years,
various definitions, but similar in nature, have been offered for “sustainability”, with
King II introducing the concept of sustainability into corporate governance (IoD,
2002:91, section 4) and defining it as follows: “This means that each enterprise must
balance the need for long-term viability and prosperity – of the enterprise itself and the
societies and environment upon which it relies for its ability to generate economic value
– with the requirement for short-term competitiveness and financial gain”. King III, the
most recent of the South African corporate governance reports issued to date, describes
sustainability as follows: “the sustainability of a company means conducting operations
in a manner that meets existing needs without compromising the ability of future
generations to meet their needs. It means having regard to the impact that the business
operations have on the economic life of the community in which it operates.
Sustainability includes environmental, social and governance issues” (IoD, 2009:126).
Lastly, it is interesting to note that sustainability in the broader sense has also been
referred to as “people, planet and profit” and defined as “ the ability of the company to
survive, or prosper, over the long term without depleting its capital assets – whether
natural, social or financial” (Engelbrecht, 2010; Trialogue, 2009:2-6).
The concepts of stakeholders, stakeholder engagement and corporate citizenship have
also developed over the years. One of the first definitions of “stakeholder” was that of
Freeman (as quoted by Breckenridge, 2004:27), who defined stakeholders as “any
group or individual who can affect or is affected by the achievement of the
organisation’s objectives.” This definition was slightly refined by Rossouw (2003:3)
and Rossouw, Prozesky, van Heerden and Van Zyl (2007:243), who defined a
stakeholder as “a person or party who is affected by a business or who has the potential
to affect the performance of a business”. The GRI (2006:40) defines stakeholders
broadly as “those groups or individuals that can reasonably be expected to be
significantly affected by the organisation’s activities, products, and/or services; or
whose actions can reasonably be expected to affect the ability of the organisation to
successfully implement its strategies and achieve its objectives”. Mervyn King, who
chaired the King Committee, defines stakeholders as: “the licensors of the business of
the company who grants the company the licence to operate its business” (King,
2006:21). King III advocates that a stakeholder inclusive corporate governance
approach should be followed that “recognises that a company has many stakeholders
that can affect the company in the achievement of its strategy and long term sustained
growth” (IoD, 2009: 100). Stakeholder engagement is widely accepted as the
foundation of corporate social responsibility and an important mechanism for assisting
the organisation in identifying risks and opportunities on current and emerging issues. It
will also contribute to addressing stakeholders’ perceptions and lead to better
stakeholder communication and reporting (Gilman, 2009:9; IoD, 2009: 100). The
B Marx & V van Dyk 107
concept of corporate citizenship is based on the assumption that an organisation is a
person and, like an individual citizen, should act with integrity and in a socially
responsible manner (IoD, 2009:11; Maighan, Ferrell & Hult, 1999:456; Smith,
1996:11). This is described as follows in King III: “responsible corporate citizenship
implies an ethical relationship of responsibility between the company and society in
which it operates. As responsible corporate citizens of the societies in which they do
business, companies have, apart from rights, also legal and moral obligations in respect
of their economic, social and natural environments. As a responsible corporate citizen,
the company should protect, enhance and invest in the wellbeing of the economy,
society and the natural environment” (IoD, 2009:11).
Stakeholders require both financial and non-financial information regarding the
company and its operations. Financial information was traditionally provided through
the annual report, while non-financial reporting, which focuses on environmental,
social, transformation, ethical, safety and health information, was provided as part of
the annual report, or in a separate sustainability report. Sustainability reporting is
described by the GRI reporting guidelines as: “the practice of measuring, disclosing,
and being accountable for organisational performance while working towards the goal
of sustainable development”. The GRI guidelines define a sustainability report as “a
single, consolidated disclosure that provides a reasonable and balanced presentation of
performance over a fixed period” (GRI, 2006:37-40). King III requires that
sustainability reporting and disclosure be integrated with the company’s financial
reporting (principle 9.2), and defines integrated reporting as: “the holistic and integrated
representation of the company’s performance in terms of both its finances and its
sustainability” (IoD, 2009:121). The Integrated Reporting Committee (hereafter IRC) of
South Africa, headed by Mervyn King, has subsequently (25 January 2011) brought out
a discussion paper in which a framework for integrated reporting is provided (Integrated
Reporting Committee (hereafter IRC), 2011).
Sustainability reporting in the private sector, internationally as well as in South Africa,
is now an accepted norm and well established (Ernst & Young, 2010a; Hughes, 2007;
KPMG, 2008; Rea, 2009; Social Investment Forum, 2008; Sustainability South Africa,
2011a). This has also been widely researched over the years, both internationally as
well as in South Africa, while, by contrast, research on sustainability reporting by
public sector entities is limited (cf. Ackers, 2009; Booysen, 1993; Dawkings &
Ngunjiri, 2008; Deegan, Cooper & Shelly, 2006a; Deegan, Cooper & Shelly, 2006b; De
Villiers, 1996; De Villiers, 1998; De Villiers, 2004; De Villiers & Barnard, 2000; De
Villiers & Vorster, 1995; Dewar, 1994; Griffith, 2002; Hartman & Painter-Morland,
2007; Holcroft, 1999; Mammatt, 2009; Manetti & Becatti, 2009; Mitchell, 2007;
Mitchell, Hill & Stobie, 2007; Mitchell et al., 2008; Mitchell & Hill, 2009; Mitchell &
Hill, 2010; Mitchell & Quinn, 2005; Morhardt, 2009; Rea, 2009; Steyn & Vorster,
1994; Wiseman, 1982; Wheeler & Elkington, 2002).
Sustainability and sustainability reporting in the public sector
Public sector entities are unique organisations affecting the everyday lives of a
country’s citizens. As one of the largest sectors in any country, their impact is far
reaching, and their responsibilities range from being a public steward of essential
mineral resources to providing critical public services such as transport and electricity.
As public entities, they should provide their services in a sustainable and effective,
efficient, responsible and transparent manner. Accordingly, it is essential that public
108 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
sector entities behave, and are seen to behave, as responsible corporate citizens (Cisco,
2011; Dixon & Pretorius, n.d.; Hughes, 2007; Kettle, 2005; Madden, 2010;
PricewaterhouseCoopers, 2011). Public sector entities are normally controlled by, and
report to, a single shareholder (namely, the state), but are accountable to all their public
sector stakeholders. This emphasises the need for open, honest and reliable sustainable
reporting to public sector stakeholders (Aristovnik, 2009; FRAB, 2005; Gosling, 2010;
Madden, 2010; Sibson, 2009). This is even more important in South Africa, given the
high proportion of the economy that is in the hands of state-owned entities (Payne,
2010).
Sustainability and sustainability reporting in the public sector are becoming
increasingly important as governments begin to rely on sustainable factors when
defining their agendas, aspirations, and the goals to which they are held accountable.
New expectations in this regard have also been raised by major events such as the 1992
United Nations Conference on Environment and Development in Brazil in 1992, the
United Nations World Summit on Sustainable development in South Africa in 2002, the
United Nations Association of Great Britain and Northern Ireland World Summit in
2005 and the 2009 Copenhagen Climate Conference (GRI, 2005:7-8;
PricewaterhouseCoopers, 2011; Terry, 2008: 3-14; Trialogue, 2010:4 United Nations,
2005).
Sustainability reporting, both in South Africa as well as internationally, is still in its
development phase. Ball and Grubnic (2007) state that “the agenda for research and
practice in sustainability accounting and accountability has been played out in most
exclusively for-profit, corporate settings”, and they have therefore called for researchers
to help in understanding the nature of sustainable development accounting and
accountability in Public Sector Organisations whose “core tasks have to do with social
welfare and justice”. In addition, the GRI (2005:7) notes that “Public agencies have a
civic responsibility to properly manage public goods, resources and/or other facilities in
a way that supports sustainable development objectives and promotes the public
interest. Public agencies are expected to be open and transparent in their management of
public funds and assets. As significant employers, providers of services, and consumers
of resources, public agencies also have a major impact on national and global progress
towards sustainable development. Given their size and influence, public agencies are
expected to lead by example in reporting publicly and transparently on their activities to
promote sustainability”.
However, sustainability reporting by public sector entities is still lagging behind that of
the private sector. This is described as follows by the Chartered Institute of Public
Finance and Accountancy (hereafter CIPFA): “public service organisations are failing
to keep pace with the steady growth of sustainability reporting in the corporate sector”
(CIPFA, 2004). More recently, this has been supported by the research conducted by the
GRI on sustainability reporting by public sector entities worldwide, which found that
only 1.7% of GRI reports published in 2009 are from public entities (GRI, 2010).
Moreover, Ernst & Young in South Africa found that sustainability reporting by South
Africa’s largest public entities varies from excellent to sub-standard (Ernst & Young,
2009a & 2010a). This is further supported by the research done by Farneti and Guthrie
(2008 & 2009), which confirms that sustainability reporting and research thereon is
mainly limited to private sector entities.
B Marx & V van Dyk 109
Sustainability reporting standards and guidelines
Standards for general use
Various standards and guidelines exist, internationally as well as in South Africa, on
sustainability reporting and what aspects to address. A review of the literature indicates
that the following have been well established over the years. These guidelines and
standards are also recognised by Graham Terry (2008:154-170; 201-211) in his
authoritative book (Green) on sustainability and sustainability reporting practices
worldwide, and by the South African Institute of Chartered Accountants’ sustainability
website (Sustainability South Africa, 2011b).
Standard Short description
The Global Reporting Initiative
(GRI, 2002, 2005, 2006, 2010)
Formed in 1997 in the United States by Ceres and the Tellus
Institute, the GRI provides a sustainability reporting
framework which is regarded as the most important guideline
on sustainability that currently exists. More than 1 300
companies have followed the GRI Framework, and hundreds
of others produce sustainability reports using their
criteria. The GRI Guidelines (now in the third version, G3)
are widely regarded as the de facto international standard for
sustainability reporting.
King Reports (IoD, 1994, 2002,
2009)
The King Reports on Corporate Governance were developed
in South Africa but have received international acclaim. The
three Reports have presented an increasing focus on
sustainability reporting.
The United Nations Global
Compact
(UNGC, 2007)
The United Nations Global Compact (hereafter UNGC) is a
set of 10 principles for businesses to follow covering
environmental responsibility, human rights and workers
rights. The principles are a central reference point for the
GRI Guidelines. Companies that sign up for the UNGC must
issue a Communication on Progress (COP) each year that
informs stakeholders of their progress in implementing the 10
principles in their business activities.
The Organisation for Economic
Co-Operation and Development
(hereafter OECD) Guidelines on
Multi-national Enterprises
(OECD, 2000)
First developed in 1976 and revised in 2000, the Guidelines
relate to the disclosure of information, employment relations,
environmental management, bribery, competition, consumer
interests, and science and technology diffusion. Signatory
governments commit themselves to establishing National
Contact Points, which will investigate complaints relating to
the Guidelines.
The International Organization
for Standardization (ISO, 2010)
The International Organization for Standardization (hereafter
ISO) is a non-governmental network of national standards
institutes of 161 countries which co-ordinates international
standards for business and products. The ISO 14000 Series of
standards focus on corporate environmental management
systems. ISO 26000, released in 2010, covers social
responsibility.
The AccountAbility’s
Framework and Standards
(AccountAbility, 2008)
AccountAbility is a non-profit, membership organisation
which was launched in 1996 to promote accountability for
sustainability. It is a global network of business, public and
civil institutions. The AA1000 Framework, which was
published in 1999, intends to help organisations build their
accountability and social responsibility through quality social
and ethical accounting, auditing and reporting. The AA1000
110 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Series is a principle-based set of standards that provides
guidance for improving the sustainability performance of
organisations, while the AA1000 APS (AccountAbility
Principles Standard) sets out the foundation principles of
inclusivity, materiality and responsiveness. AA1000SES
(Stakeholder Engagement Standard) gives guidance to
organisations on engaging with stakeholders.
Many organisations use the AccountAbility standards in
conjunction with the GRI Guidelines.
The Prince of Wales Accounting
for Sustainability Project
(Accounting for Sustainability,
2006).
Launched by the Prince of Wales, the Accounting for
Sustainability Project promotes sustainability and develops
tools and guidelines to assist organisations implement
sustainability strategies and practices. The project believes
that it is only through the integration of environmental and
social factors into business and management reporting that
the fundamental connection between strategic direction,
financial performance and sustainability impacts will be
made clear. The Accounting for Sustainability Project has
developed two sustainability management tools for
organisations and a framework for a sustainability report.
The Carbon Disclosure Project
(CDP, 2011)
The Carbon Disclosure Project (hereafter CDP) is an
independent non-profit organisation that coordinates investor
relationships between shareholders and corporations with
respect to climate change. The CDP, started in 2000, has
become the global standard for carbon emission disclosure
methodology and processes. The CDP website provides the
largest repository of corporate greenhouse gas emissions data
in the world. This data is obtained by sending out annual
questionnaires to the chief executives of over 4 500 of the
world’s largest companies with questions relating to
awareness and management of carbon emissions, climate
change, and, more recently, water. The CDP sends out the
questionnaires on behalf of institutional investors – it is
backed by 534 investors representing US$64 trillion in assets
under management.
The Institute of Directors of
South Africa’s Sustainable
Development Forum (IoD 2009;
2011)
The Institute of Directors of South Africa established the
Sustainable Development Forum, which researches and
disseminates information on developments relating to
sustainability issues. It has released two position papers to
date.
Position Paper 1: Implementing Sustainable
Development as a Strategic Business Model (August
2009)
Position Paper 2: South African Business & Climate
Change (September 2010)
The GRI sustainability reporting guidelines are highly regarded and widely used as
guidance on what sustainability information companies should report (Hartman &
Painter-Morland, 2007; Jones & Solomon, 2010:20-22; Rea, 2009). Leeson et al. (2005)
confirm that “after GRI, the highest awareness is with the ISO 14000 series and the
work of UNEP/SustainAbility.” They also state that the GRI framework is “perceived to
represent best practice reporting”.
B Marx & V van Dyk 111
Specific public sector standards
The GRI has also issued guidance for public sector entities in the form of the GRI
Sector Supplement for Public Agencies Pilot Version 1.0 in March 2005. This
supplement states that “broadly speaking, the sum of disclosures in a report should help
users understand if the public agency’s mission has been translated into relevant policies
and actions that contribute to sustainable development in a meaningful way. The
practice of reporting provides transparency and accountability for the agency’s
performance, which is essential in sustainable development” (GRI, 2005). Furthermore,
the supplement provides guidance on what public agencies should report in accordance
with the triple bottom line by providing both qualitative and quantitative information.
The information is divided into three types (GRI, 2005):
Information on organisational performance, which can be reported through the use
of performance indicators, such as those already in the Guidelines. This type of
information illustrates the organisation’s internal policies and its role as a
consumer and employer (which can be quite substantial);
Information on external public policies and implementation measures of the
agency that relate to sustainable development and their performance; and
Information on economic, environmental, or social conditions within the agency’s
mandate or area of jurisdiction that may be the focus of government public policies
and implementation measures.
According to the GRI Sector Supplement for Public Agencies (GRI, 2005),
understanding organisational performance for a public agency draws on all three types
of information, and seeks to create a framework that enables the reporting organisation
to:
Outline its vision and strategy for sustainable development and explain the
organisation’s position in the broader national and international context of
sustainability;
Identify key internal and public policy objectives and goals for the agency;
Provide qualitative and quantitative information on operational performance; and
Outline public policies, implementation measures, and progress in terms relevant
to the goals and mission of the organisation, and the conditions that it seeks to
influence.
Part C of the GRI Sector Supplement for Public Agencies specifies the content of a
GRI-based report. The report content is organised in what GRI considers a logical order,
and reporting organisations are encouraged to follow this structure in writing their
reports. Part C of the GRI Sector Supplement for Public Agencies comprises five
sections (GRI, 2005):
1. Vision and Strategy – description of the reporting organisation’s strategy with
regard to sustainability, including a statement from the Chief Executive Officer
(hereafter CEO).
112 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
2. Profile – overview of the reporting organisation’s structure, operations and scope
of the report, including assurance provided thereon.
3. Governance Structure and Management Systems – description of organisational
structure, policies, and management systems, including stakeholder engagement
efforts.
4. GRI Content Index – a table supplied by the reporting organisation identifying
where the information listed in Part C of the GRI Sector Supplement for Public
Agencies is located within the organisation’s report.
5. Performance Indicators – measures of the impact or effect of the reporting
organisation divided into integrated, economic, environmental, and social
performance indicators.
The five sections indicated above informed the checklist against which the sustainability
reporting of the selected sample of public sector entities was analysed to determine the
degree of compliance with the GRI Sector Supplement for Public Agencies. The
findings of the analysis are discussed in the research findings and interpretation section.
METHODOLOGY
Sustainability reporting by public sector entities was empirically tested through a
content analysis of the annual and sustainability reports of the companies selected for
review.
Sample
The sample selected for the empirical study comprises the top 10 state-owned entities,
regarded as major public entities in terms of the Public Finance Management Act, based
on the total asset value as disclosed in the latest financial statements. These entities,
representing major public interest, have a responsibility to be responsible corporate
citizens and to provide open, honest and reliable reporting on their economic, social,
environmental and governance practises to their public sector stakeholders. These
companies are also those surveyed by Ernst & Young in its yearly assessments of
Excellence in Corporate Reporting and Excellence in Sustainability Reporting (Ernst &
Young, 2009a; 2009b; 2009a; 2010b), which have established themselves as credible
surveys over the years.
Content analysis of sustainability reports
The most recently available sustainability reporting of the companies as at 30 August
2010, as contained in the annual reports, sustainability reports and company websites,
was inspected. The sustainability reporting of nine of the 10 companies in the sample
was inspected and analysed (90% coverage). The annual report of one of the companies
was not available, could not be obtained and was also not available on the
organisation’s website. The names of the companies in the sample are listed in
Annexure A.
This study is limited to the assessment of sustainability reporting in annual reports,
sustainability reports and company websites, as such reports are considered important
B Marx & V van Dyk 113
corporate governance and stakeholder documents produced by companies through
which they communicate with their investors, as well as their stakeholders at large
(Abeysekera, 2007:333; Bartlett & Chandler, 1997:245; Boesso & Kumar, 2007:281-
282; De Villiers, 2004; GRI, 2002:17; GRI, 2006:6; Ponnu & Ramthandin, 2008:44;
Stainbank & Peebles, 2006:69; Wiseman, 1982:53). This is further supported by King
III, which emphasises the importance of integrated reporting as a means of increasing
the trust and confidence of corporate stakeholders and the legitimacy of an
organisation’s operations (IoD, 2009: 13-14).
The analysis was done according to a Likert Scale of 1 to 5, with 1 representing no
disclosure, 2 disclosure to a lesser extent, 3 disclosure to some extent, 4 disclosure to a
large extent and 5 representing significant disclosure. The distinctions between the
various sections of the Likert Scale were based on the quantity of information disclosed,
balanced with the quality, relevance and perceived usefulness of the information
reported on by the entities.
Research control
The research consisted of analysing the sustainability reporting of the companies in the
population. The analysis was performed according to a checklist based on the literature
review of sustainability reporting practices, against which the information of
sustainability reporting was measured. The results were tabled and confirmed by an
independent adjudicator to ensure the quality and accuracy of the results obtained.
RESEARCH FINDINGS AND INTERPRETATION
Disclosure of sustainability in vision and strategy
Objective of the analysis
The objective of this aspect of the analysis was to establish whether large South African
public sector entities included in their sustainability reporting a description of the
organisation’s strategy with regard to sustainability, including a statement from the
CEO on the entity’s commitment to sustainability, and a description of the wider
context of sustainability for the organisation.
Findings and deductions
Table 1: Disclosure of sustainability in vision and strategy
Vision and strategy
None
To a lesser
extent
To some
extent
To a
large
extent
Significant
disclosure
Total
Strategy and analysis regarding sustainability 1 (11%) 2 (22%) 4 (45%) 2 (22%) 0 (0%) 9 (100%)
Statement of commitment to sustainability from
CEO 2 (22%) 1 (11%) 2 (22%) 3 (34%) 1 (11%) 9 (100%)
Wider context of sustainability discussed 2 (22%) 1 (11%) 3 (34%) 2 (22%) 1 (11%) 9 (100%)
Incorporation into organisation’s overall strategy,
objectives and/or values 1 (11%) 1 (11%) 2 (22%) 5 (56%) 0 (0%) 9 (100%)
Sustainability linkages discussed 1 (11%) 2 (22%) 4 (45%) 2 (22%) 0 (0%) 9 (100%)
From the above findings it can be deduced that most of the large South African public
sector entities in the sample disclose information regarding sustainability in their vision
and strategy statements, although only two entities (22%) do so to a large extent, with
114 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
none having significant disclosure in this regard. Two entities (22%) do not have a
statement from the CEO of the organisation regarding the impact of sustainability issues
on the organisation, nor do they discuss the wider context of sustainability on the
organisation. This is contrary to guidelines of the GRI Sector Supplement for Public
Agencies (as discussed in the theoretical background), which requires a description of
the reporting organisation’s strategy with regard to sustainability as well as a statement
from the CEO in this regard. As far as incorporating sustainability into the
organisation’s strategy, objectives and values is concerned, five entities (56%) do so to
a large extent, with one company not doing so at all. Sustainability linkages to the
business activities are discussed to some or to a large extent by six entities (67%), with
again one entity not doing so at all.
As discussed in the theoretical background, such disclosures, as stated in the table
above, are needed to understand the entity’s mission and how it is translated into
relevant policies and actions that contribute to sustainable development and sustainable
business practices. From the above findings it is evident that sustainability disclosure at
large public sector entities varies from no disclosure at all to disclosure to a large extent
in some cases, and significant disclosure that is the exception.
Disclosure of the entity’s profile and scope of sustainability report
Objective of the analysis
The objective of this aspect of the analysis was to establish whether large South Africa
public sector entities included an overview of the organisation’s structure and
operations as well as information regarding the scope of the sustainability report.
Findings and deductions
Table 2: Disclosure of the entity’s profile and scope of sustainability report
Profile and scope of report
None
To a lesser
extent
To some
extent
To a
large
extent
Significant
disclosure
Total
Organisational profile disclosed 1(11%) 0 (0%) 5 (56%) 3 (33%) 0 (0%) 9 (100%)
Materiality used in preparing report and
identification of material items disclosed (scope and
boundary)
2 (23%) 3 (33%) 3 (33%) 1 (11%) 0 (0%) 9 (100%)
Disclosure of completeness of information reported 3 (33%) 1 (11%) 2 (23%) 3 (33%) 0 (0%) 9 (100%)
Disclosure regarding efforts to ensure accuracy of
report 4 (44%) 4 (44%) 1 (12%) 0 (0%) 0 (0%) 9 (100%)
Disclosure of independent assurance on
sustainability information reported 8 (89%) 0 (0%) 0 (0%) 0 (0%) 1(11%) 9 (100%)
It is evident from the findings that the majority of the public sector entities within the
sample (eight companies, or 89%) disclose information regarding their organisations’
profile as recommended by the GRI Sector Supplement for Public Agencies guidelines
as discussed in the theoretical background. However, regarding the scope and boundary
of the report, only four entities disclose information on materiality and the identification
of material items for reporting to some (33%) or to a large extent (11%), with two
entities (23%) not disclosing this information at all. In terms of disclosure on the
completeness of the information reported upon, three entities (33%) have no disclosure
on this issue, with three entities (34%) disclosing this information to some (23%) or to a
lesser extent (11%). A third of the organisations (33%), however, disclose this
information to a large extent. Regarding the efforts to ensure the accuracy of the
B Marx & V van Dyk 115
sustainability information contained in the report, no entities have significant disclosure
or even disclosure to a large extent. One entity (12%) has disclosure to some extent,
with four entities (44%) providing this information to a lesser extent and the remaining
four entities (44%) having no disclosure in this regard. This result is confirmed to some
degree by the results on the assurance obtained by the organisation on the sustainability
report. One entity (11%) obtained assurance and disclosed the independent assurance
report. The remaining eight entities (89%) made no effort to independently verify the
information contained in their sustainability reports.
The above finding raises questions on the relevance and usefulness of the sustainability
information reported by the large public sector entities, as sufficient detail on the
organisations’ profile, scope and boundary of reporting is needed for the stakeholders to
understand scope of operations and sustainability impact. The absence of independent
assurance on the sustainability reporting also raises concerns about the credibility and
reliability of the information reported.
Disclosure of stakeholder engagement efforts
Objective of the analysis
The objective of this aspect of the analysis was to establish whether large South African
public sector entities disclose sufficient information regarding their stakeholder
engagement efforts.
Findings and deductions
Table 3: Disclosure of stakeholder engagement efforts
Stakeholder engagement
None
To a
lesser
extent
To some
extent
To a large
extent
Significant
disclosure
Total
Disclosure of stakeholders and stakeholder
engagement policies and processes 1 (11%) 3 (33%) 3 (33%) 0 (0%) 2 (23%) 9 (100%)
Disclosure of policies 4(45%) 3 (33%) 2 (22%) 0 (0%) 0 (0%) 9 (100%)
Disclosure of management systems 4 (44%) 5 (56%) 0 (0%) 0 (0%) 0 (0%) 9 (100%)
Stakeholder engagement and reporting thereon as discussed in the theoretical
backgroundis widely accepted as the foundation of corporate social responsibility and
an important mechanism in assisting the organisation in identifying risks and
opportunities. It will also contribute to addressing stakeholders’ perceptions and lead to
better stakeholder communication and reporting. Reporting thereon is also a GRI
recommendation, as discussed in the theoretical bacgkround. Against this background
the findings of the study indicate that reporting by the large public sector entities on
their stakeholder engagement policies and processes is lacking. From the above table of
findings it is evident that stakeholder engagement policies and processes differ
significantly, as two entities (23%) provide significant disclosure in this regard, none
provide disclosure to a large extent (0%) and six provide disclosure to some extent
(33%) or to a lesser extent (33%), with one entity (11%) providing no disclosure in this
regard. It is also obvious from the findings that the disclosures of stakeholder
engagement policies and management systems by large public sector entities are either
lacking or substandard, as the majority of the entities provide either no disclosure in this
regard (44%), or to a lesser extent (56%), and none have disclosure to a large or
significant extent.
116 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Disclosure of GRI content and use of GRI guidelines
Objective of the analysis
The objective of this aspect of the analysis was to establish whether large South African
public sector entities disclose compliance with the GRI Sector Supplement for Public
Agencies, and further whether a GRI reference table in this regard is provided.
Findings and deductions
Table 4: Disclosure of use of GRI guidelines and GRI content index
GRI
None
To a lesser
extent
To some
extent
To a
large
extent
Significant
disclosure
Total
Use of GRI guidelines 4 (44%) 0 (0%) 0 (0%) 0 (0%) 5 (56%) 9 (100%)
Disclosure of GRI content index 4 (44%) 0 (0%) 2 (22%) 2 (22%) 1 (12%) 9 (100%)
It can be concluded from the results in the above table that four of the large public
sector entities (44%) make no reference to GRI in their sustainability reporting, while
five (56%) provide significant disclosure in this regard. Of the five entities that
disclosed a GRI content index, one entity (12%) reported significant compliance with
the GRI guidelines and two entities (22%) reported compliance and disclosure thereof to
a large extent, with two entities (22%) reporting compliance to some extent. The GRI
Sector Supplement for Public Agencies (as discussed in the theoretical background)
recommends that reporting organisations should supply a table identifying where the
information is located in the organisation’s report. Providing such disclosure will
improve the user friendliness of the sustainability reporting and it is therefore
concerning that 44% of the entities make no use of the GRI guidelines.
Disclosure of performance indicators
Objective of the analysis
The objective of this aspect of the analysis was to establish whether large South African
public sector entities included performance data of their organisations with reference to
integrated, economic, environmental and social performance indicators.
Findings and deductions
Table 5: Disclosure of performance indicators
Vision and strategy
None
To a lesser
extent
To some
extent
To a
large
extent
Significant
disclosure
Total
Information compared to prior years 2 (22%) 0 (0%) 6 (67%) 1 (11%) 0 (0%) 9 (100%)
Performance quantified 2 (22%) 2 (22%) 4 (45%) 1 (11%) 0 (0%) 9 (100%)
Disclosure of economic performance 1 (11%) 3 (33%) 5 (56%) 0 (0%) 0 (0%) 9 (100%)
Disclosure of environmental performance 2 (22%) 1 (11%) 4 (45%) 2 (22%) 0 (0%) 9 (100%)
Disclosure of health issues 1 (11%) 2 (22%) 5 (56%) 1 (11%) 0 (0%) 9 (100%)
Disclosure of product responsibility 1 (11%) 1 (11%) 3 (33%) 4 (45%) 0 (0%) 9 (100%)
Disclosure of ethical practices 1 (11%) 3 (33%) 3 (33%) 2 (23%) 0 (0%) 9 (100%)
Disclosure of impact on society and
transformation 0 (0%) 3 (33%) 4 (45%) 2 (22%) 0 (0%) 9 (100%)
Disclosure of human capital issues 1 (11%) 2 (22%) 4 (45%) 2 (22%) 0 (0%) 9 (100%)
B Marx & V van Dyk 117
The results show that most of the large public entities report on social, economic and
environmental issues, as, at most, only two entities (22%) do not disclose information
on environmental performance and only one entity (11%) makes no disclosure of
economic and social performance. Mostly, the large public sector entities disclose
information to a large extent or less, as none of the entities in the sample provided
significant disclosure about social, economic and environmental issues. In instances
where information is provided, only six entities (67%) compare information to prior
years, and only four entities (45%) quantify their performance in these areas. It was
observed that the majority of the large public sector entities presented their economic
information in the form of a value-added statement, a document strongly supported by
Stainbank (2009). It is concerning that none of the large public sector entities provide
significant information in their reports on performance data regarding social, economic
and environmental issues and only a limited number provide information thereon to a
large extent.
The above disclosures are recommended by the GRI Sector Supplement for Public
Agencies and contrarily to public sector entities, accepted, established and well reported
practice for private sector entities (as discussed in the theoretical background As far as
the large public sector entities are concerned it is encouraging to see from the findings
that the majority of the entities include some information in this regard, but it is noted
that more comprehensive and detailed disclosure is required.
Summative findings
From the research results outlined above, it is evident that few of the large South
African public sector organisations have embraced the recommendations of the GRI
Sector Supplement for Public Agencies and included the recommended information on
sustainability reporting to a large or significant extent in their sustainability reports.
Most of the entities report on sustainability issues in their vision and strategy to some
extent. The organisations also disclose information on the organisational profile of the
organisation, but with respect to the scope and boundary of their reporting as well as
efforts to ensure accuracy, most public sector organisations include very limited
information. This result is echoed in the information provided on governance structure
and management, GRI content index and compliance, and performance indicators with
reference to economic, environmental and social aspects. These findings substantiate the
literature, as discussed in the theoretical background, that sustainability reporting in the
public sector is still in a development phase, is lagging behind that of the private sector
and varies considerably.
The fact that public sector entities are central to the delivery of sustainable
development, have a responsibility to provide their services in an environmentally
friendly and sustainable manner, and that their activities affect the lives of many and
diverse stakeholder groupings, emphasises the need for open, honest and reliable
sustainability reporting. Sustainability and sustainability reporting in the public sector is
also becoming increasingly important as governments begin to rely on sustainable
factors when defining their agendas, aspirations, and the goals to which they are held
accountable. Against this background the findings of the study provide evidence that
sustainability reporting at the large public sector entities in South Africa can improve
significantly, and is an aspect that should receive urgent attention.
118 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
CONCLUSION AND AREAS FOR FUTURE RESEACH
With the increased focus on sustainability and a society centred on corporate
accountability, public sector entities should conduct their business in a responsible and
sustainable manner, and should accordingly report to their stakeholders in this regard.
The study found that whereas the concept of sustainability and sustainability reporting is
reasonably well established in the private sector, it is not yet being widely applied in the
public sector. The content analysis of the sustainability reporting of the largest public
sector entities in South Africa also found that the standard of such reporting ranges from
excellent to sub-standard.
King III, which applies to all organisations, including public sector entities, became
effective for year-ends beginning after 1 March 2010. As the annual reports analysed
did not fall under the requirements of King III, it is recommended that an analysis
similar to this one is undertaken once King III becomes effective. This should be done
to assess the impact of King III on the sustainability reporting of public sector entities. It
is also recommended that future research should focus on sustainability reporting at
smaller public sector entities, as well as municipal entities.
REFERENCES
Abeysekera, I. (2007). Intellectual capital reporting between a developing and
developed nation. Journal of Intellectual Capital, 8(2):329-345.
Accenture. (2010). A New Era of Sustainability: UN Global Compact-Accenture CEO
Study 2010. Brochure. Europe.
AccountAbility. (2008). Setting the Standard for Corporate Responsibility and
Sustainable Development. Available from:
http://www.accountability.org/about-us/publications/aa1000.html (Accessed 15
February 2011).
Accounting for Sustainability. (2006). The Prince’s Accounting for Sustainability
Project. Available from: http://www.accountingforsustainability.org/output/page146.asp
(Accessed 19 February 2011).
Ackers, B. (2009). Corporate social responsibility assurance: how do South African
publicly listed companies compare? Meditari Accountancy Reseach, 17(2):1-17.
April, K.A., Bosma, P. and Deglon, D.A. (2003). IC measurement and reporting:
establishing a practice in SA mining. Journal of Intellectual Capital, 4(2):165-180.
Aristovnik, A. (2009). Performance measurement in the public sector at national and
local levels – some applications to Slovenia. Faculty of Administration, University of
Ljubljana. Ljubljana.
Ball, A. and Grubnic, S. (2007). Sustainability accounting and accountability in the
public sector. In Unerman, J., Bebbington, J. And O’Dwyer, B. (Eds). Sustainability
Accounting and Accountability. Routledge. London. Thirst edition
B Marx & V van Dyk 119
Barac, K. and Moloi, T. (2010). Assessment of corporate governance reporting in the
annual reports of South African listed companies. The South African Journal of
Accountability and Auditing Research, 10(1):19-28.
Bartlett, S.A. and Chandler, R.A. (1997). The corporate report and the private
shareholder: Lee and Tweedie twenty years on. British Accounting Review, 29(3):245-
264.
Boesso, G. and Kumar, K. (2007). Drivers of corporate voluntarily disclosure: A
framework and empirical evidence from Italy and the United States. Accounting,
Auditing and Accountability Journal, 20(2):269-296.
Booysen, S.F. (1993). ’n Raamwerk vir finansiële verslagdoening aan werknemers in
die Republiek van Suid Afrika. D.Com thesis. University of Pretoria.
Breckenridge, R.G. (2004). Modelling ethical corporate governance in terms of the
King Report. Project report submitted to the Faculty of Commerce, Law and
Management in partial fulfilment for the degree of Master of Business Administration.
University of the Witwatersrand.
Brennan, N. and Solomon, J. (2008). Corporate governance, accountability and
mechanisms of accountability: an overview. Accounting, Auditing & Accountability,
21(7):885-906.
Carbon Disclosure Project. (CDP). (2011). Carbon Disclosure Project – Global climate
change reporting system. Available from:
https://www.cdproject.net/en-US/WhatWeDo/Pages/overview.aspx (Accessed 17
February 2011).
Chartered Institute of Public Finance and Accounting. (CIPFA). (2004). Advancing
sustainability accounting and reporting: An agenda for Public Service Organisations –
a discussion paper. London. United Kingdom.
Cisco. (2011). Environmental Sustainability in the Public Sector. Sysco Systems.
Available from: http://www.cisco.com/web/strategy/docs/education/schl-green-wp.pdf
(Accessed 16 February 2011).
Da Piedade, L. and Thomas, A. (2006). The case for corporate responsibility: an
exploratory study. SA Journal of Human Resource Management, 4(2):65-74.
Dawkins, C. and Ngunjiri, F.W. (2008). Corporate Social Responsibility Reporting in
South Africa. Journal of Business Communication, 45(3):286-307.
De Villiers, C.J. (1996). The awareness levels of different stakeholder groups and their
willingness to support environmental reporting in South Africa. D.Com thesis.
University of Pretoria.
De Villiers, C.J. (1998). The Willingness of South Africans to Support More Green
reporting. South African Journal of Economic and Management Sciences, 1(1):145-167.
120 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
De Villiers, C.J. (2004). Ethical arguments regarding corporate environmental reporting.
Meditari Accountancy Reseach, 12(1):21–38.
De Villiers, C.J. and Barnard, P. (2000). Environmental reporting in South Africa form,
1994 to 1999. Meditari Accounting Research, 8:15-23.
De Villiers, C.J. and Vorster, Q. (1995). More corporate environmental reporting in
South Africa. Meditari Accounting Research, 3:44-66.
Deegan, C., Cooper, B.J. and Shelly, M. (2006a). An investigation of TBL report
assurance statements: UK and European evidence. Managerial Auditing Journal,
21(4):330-369.
Deegan, C., Cooper, B.J. and Shelly, M. (2006b). An Investigation of TBL Report
Assurance Statements: Australian Evidence. Australian Accounting Review, 16(39):2-
18.
Dewar, N. (1994). An analysis of the quality of environmental disclosures in annual
financial statements of selected South African companies and a suggested
environmental reporting model. Master’s dissertation. University of Pretoria.
Dixon, Y. and Pretorius, J.G. (n.d.). Sustainable Development: A South African
Perspective. Available from: http://www.helsinki.fi/imua2001/Pretoruis.pdf (accessed
15 February 2011).
Engelbrecht, L. (2009). All the King’s horses and all the King’s men. Directorship,
April/May/June: 4-5.
Engelbrecht, L. (2010). The business imperative of sustainability. Excellence in
Sustainability Reporting, 2010. Ernst & Young. South Africa.
Ernst & Young. (2008a). Ready or not, here comes sustainability. EYGM Limited.
DC0038.
Ernst & Young. (2008b). Business and sustainability. EYGM Limited. SCORE retrieval
file 080626.
Ernst & Young. (2009a). Excellence in Corporate Reporting 2009. A survey of the
annual reports from South Africa’s top 100 companies and top 10 state-owned entities.
South Africa.
Ernst & Young. (2009b). Excellence in Sustainability Reporting 2009: What will you
leave behind? A survey of the annual reports from South Africa’s top companies and
top state-owned entities. South Africa.
Ernst & Young. (2010a). Excellence in Corporate Reporting 2010. A survey of the
annual reports from South Africa’s top 100 companies and top 10 state-owned entities.
South Africa.
B Marx & V van Dyk 121
Ernst & Young. (2010b). Excellence in Sustainability Reporting 2010: We become what
we value. A survey of the annual reports from South Africa’s top companies and top
state-owned entities. South Africa.
Farneti, F. and Guthrie, J. (2008). Sustainability reporting by Australian Public Sector
Organisations: Why they report. 5th
International Congress on Accounting, Auditing
and Management in Public Sector reforms. 3-5 September. Amsterdam. The
Nederlands.
Farneti, F. and Guthrie, J. (2009). Sustainability reporting by Australian Public Sector
Organisations: Why they report. Accounting Forum, 33(2):89-98.
FRAB. (2005). Sustainability Reporting: Financial Reporting Advisory Board Paper.
HM Treasury. Available from:
http://www.hm-teasury.gov.ulk/d/FRAB(64)03-SUSTAINABILITY-REPORTING.pdf
(Accessed 17 February 2011).
Gilman, K. (2009). Stakeholder Engagement: Vital in business strategy and approach.
Excellence in Sustainability Reporting, 2009. Ernst & Young. South Africa.
Global Reporting Initiative. (GRI). (2002). Sustainability Reporting Guidelines (Version
2), GRI, Boston. USA.
Global Reporting Initiative. (GRI). (2005). Sector Supplement for Public Agencies. Pilot
Version 1. GRI. Amsterdam. The Nederlands.
Global Reporting Initiative. (GRI). (2006). Sustainability Reporting Guidelines (Version
3), GRI. Boston. USA.
Global Reporting Initiative. (GRI). (2010). GRI Reporting in Government Agencies.
Amsterdam. The Nederlands.
Gosling, P. (2010). Sustainability in the public sector: Accounting & Business.
Available from: http://www.paulgosling.net/2010/09/sustaianibilty-in-the-public-sector-
accounting-business (Accessed 18 February 2011).
Griffith, K. (2002). Corporate social reporting in South Africa: a test of legitimacy
theory. Master’s Dissertation. University of Cape Town.
Hartman, L.P. and Painter-Morland, M. (2007). Exploring the global reporting initiative
(GRI) guidelines as a model for triple bottom line reporting. African Journal of
Business, Ethics, 2(1):45-57.
Hespenheide, E. (2010). What is Corporate Responsibility & Sustainability reporting
and why is it important. Available from:
http://globalblogs.deloitte.com/deloitteperspectives/2010/07/what-is-corporate-
responssibili
(Accessed 14 February 2011).
Holcroft, N.M. (1999). Corporate environmental reporting: a review of national and
international trends in structures and systems necessary for the production of effective
122 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
corporate environmental reports. Master’s dissertation. University of Natal.
Pietermaritzburg.
Hughes, P. (2007). The Case for Sustainability Reporting by the Public Sector. The
Centre for Public Agency Sustainability Reporting. Australia.
Institute of Directors. (IoD). (1994). King I Report on Corporate Governance. Institute
of Directors in Southern Africa. Johannesburg.
Institute of Directors. (IoD). (2002). King II Report on Corporate Governance. Institute
of Directors in Southern Africa. Johannesburg.
Institute of Directors. (IoD). (2009). King Report on Corporate Governance for South
Africa. Institute of Directors in Southern Africa. Johannesburg.
Institute of Directors. (2011). Sustainable Development Forum. Available from:
http://www.iodsa.co.za/membership.asp?CatID=145 (Accessed 11 February 2011).
Integrated Reporting Committee of South Africa. (IRC). (2011). Framework for
Integrated Reporting and the Integrated Report. Integrated Reporting Committee. South
Africa. Available from: www.sustainabilitysa.org.
International Organization for Standardization (ISO). (2010). Management and
leadership standards. Available from:
http://www.iso.org/iso/iso_catalogue/management_and_leadership_standards.htm
(Accessed 13 February 2011).
Jhaveri, N.J. (1998). Corporate Governance: The Concept and its Implications in the
Indian Context. ASCPJournal of Management. Available from:
http://www.asci.org.in/publications/journal1998.asp (Accessed 21 November 2002).
Jones, M.J. and Solomon, J.F. (2010). Social and environmental report assurance: Some
interview evidence. Accounting Forum, 34:20-31.
Kettle, J. (2005). Sustainability must be in the hearts and minds of government.
Available from: http://wwww.edie.net/news/news-story.asp?id=9401 (Accessed 11
February 2011).
King, M.E. (2006). The Corporate Citizen. Johannesburg. Penguin Books (South
Africa) (Pty) Ltd. Third edition.
KPMG. (2008). Climate Changes Your Business. KPMG Global Sustainability Services.
Amstelveen. The Nederlands.
Leeson, R., Ivers, J. and Dickson, D. (2005). Sustainability reporting by the public
sector: practice, uptake and form. GRI G3. Available from:
http://www.globalreporting.org/NR/rdonlyres/FAFD9A06-702A-4AA8-988C-
979DBCCBC948/0/LeesonEtAISustReortingByPublicSector.pdf (Accessed 19
February 2011).
B Marx & V van Dyk 123
Livesey, C. (2002). The Discourse of the middle ground: Citizen Shell commits to
sustainable development. Management Communication Quarterly, 15:313-348.
Madden, P. (2010). Public sector leadership on sustainability. Available from:
http://sd.defra.gov.uk/2010/07/public-sector-leadership-on-sustainability (Accessed 14
February 2011).
Maighan, I, Ferrell, O.C. and Hult, G.T.M. (1999). Corporate Citizenship: Cultural
Antecedents and Business Benefits. Journal of the Academy of Marketing, 27(4):455-
469.
Mammatt, J. (2008). Business and sustainability. Ernst & Young. EYGM Limited.
Mammatt, J. (2009). Sustainability reports: Increasingly necessary, but only as good as
the assurance. In Touch. Ernst & Young. South Africa. January.
Mammatt, J. (2010). Integrated sustainability reporting and assurance. Conference
proceedings of the CIS Corporate Governance Conference held in Johannesburg.
Conducted by Ernst & Young Inc. Johannesburg. September.
Mammatt, J., Marx, B. and van Dyk, V. (2010). Sustainability reporting and assurance:
the way of the future. Accountancy SA (asa), January: 22-24.
Manetti, G. and Becatti, L. (2009). Assurance services for sustainability reports:
standards and empirical evidence. Journal of Business Ethics, 87:289-298.
Marx, B. (2008). An analysis of the development, status and functioning of audit
committees at large listed companies in South Africa. Unpublished thesis submitted in
fulfilment of the requirements for the degree Doctor Commercii in the Faculty of
Economic and Financial Sciences. University of Johannesburg.
Mirfazli, E. (2008). Evaluate corporate social responsibility disclosure at Annual Report
Companies in multifarious group of industry members of Jakarta Stock Exchange
(JSX), Indonesia. Social Responsibility Journal, 4(3):388-406.
Mitchell, M. (2007). Towards improved corporate environmental and social reporting
in South Africa. PhD dissertation. University of Natal-Pietermaritzburg.
Mitchell, M., Curtis, A. and Davidson, P. (2008). Evaluating the process of triple
bottom line reporting: Increasing the potential for change. Local Environment, 13(2):67-
87.
Mitchell, M., Hill, T. R., and Stobie, B.S. (2005). Corporate social and environmental
reporting and the impact of international environmental policy in South Africa.
Southern African Journal of Accountability and Auditing Research, 6:67-74.
Mitchell, M. and Hill, T. R. (2009). Corporate social and environmental reporting and
the impact of international environmental policy in South Africa. Corporate Social
Responsibility and Environmental Management, 13(2):17-33.
124 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Mitchell, M. and Hill, T. R. (2010). An exploratory analysis of stakeholders’
expectations and perceptions of corporate social and environmental reporting in South
Africa. SA Journal of Accounting Research, 24(1):49-78.
Mitchell, M. and Quinn, N.W. (2005). Environmental reporting Disclosure in South
Africa: A comparative study of the expectations of selected groups of preparers and
users. Meditari Accountancy Reseach, 13(2):17–33.
Morhardt, J.E. (2009). Corporate social responsibility and sustainability reporting on the
Internet. Business Strategy and the Environment. DOI: 10.1002/bse.657. Available
from:
http://0-onlinelibrary.wiley.com.ujlink.uj.ac.za/doi/10.1002/bse.657/pdf (Accessed 13
September 2010).
Naidoo, R. (2009). Corporate Governance – An Essential Guide for South African
Companies. Durban: LexisNexis. Thirst edition
O’Carroll, M. (2009). The natural environment as an integral part of the triple bottom
line. Available from: http://ujdigispace.uj.ac.za:8080/dspace/handle/10210/2030
(Accessed 12 September 2010).
Organisation for Economic Co-Operation and Development. (OECD). (2000). OECD
Guidelines for Multinational Enterprises. Available from:
http://www.oecd.org/dataoecd/56/36/1922428.pdf (Accessed 11 February 2011).
Paris, T.M. (2006). Corporate Sustainability reporting. Environment, June: 3.
Payne, N. (2010). Governance of State Owned Entities. Directorship, April:30-32.
Pearce, J.A. and Doh, J.P. (2005). The highest impact of collaborative social initiatives.
MIT Sloan Management Review, 46(3):30-39.
Piketh, S.J. (2010). SA’s reality in terms of climate change and sustainable
development. Accountancy SA (asa), January: 16.
Ponnu, C.H. and Ramthandin, S. (2008). Governance and performance: publically listed
companies in Malaysia. Journal of Business Systems, Governance and Ethics, 3(1):35-
53.
PricewaterhouseCoopers. 2008. The sustainability agenda: Industry perspectives.
(Brochure). Johannesburg. PricewaterhouseCoopers.
PricewaterhouseCoopers. 2011. The government & public sector sustainability agenda.
PwC Sustainability industry – Government & public sector. Available from:
http://www.pwc.com/gx/en/sustainability/industries/goverment-and-public-sector.jhtml
(Accessed 15 February 2011).
Prizma. (2011). GRI Sustainability Reporting in the Public Sector. Available from:
http://prizmablog.com/2010/02/13/gri-sustainability-reporting-in-the-public-sector
(Accessed 19 February 2011).
B Marx & V van Dyk 125
Rea, H. (2009). King III and GRI + 10: A 2009 review of sustainability reporting in
South Africa. SustainabilityServices. Johannesburg, South Africa.
Rossouw, D. (2003). The governance of Ethical Risk. Financial Mail supplement.
BDFM.13 June: 3-8.
Rossouw, D., Prozesky, M., Van Heerden, B. and Van Zyl, M. (2007). Ethics for
Accountants and Auditors. New York: Oxford University Press. Thirst edition.
Sibson, J. (2009). Sustainability reporting in the public sector. BPTA. Available from:
http://www.bptawards.com/post_awards_brochure_sustainability_ps_09.pdf (Accessed
19 February 2011).
Smith, N.C. (1996). Corporate Citizens and Their Critics. New York Times, 9 August:
11.
Social Investment Forum. (2008). Sustainability Reporting in Emerging Markets: An
analysis of the sustainability reporting in selected sectors of seven emerging market
countries. Available from:
http://www.socialinvest.org/resources/research/documents/SIF-SIRAN-
KLDReportforEMTransparency2008.pdf (Accessed 16 February 2011).
Stainbank, L.J. (2009). The value added statement: does it add any value? Meditari
Accountancy Reseach, 17(2):137.
Stainbank, L. and Peebles, C. (2006). The usefulness of corporate annual reports in
South Africa: perceptions of preparers and users. Meditari Accountancy Reseach,
14(1):69-80.
Steyn, J.O. and Vorster, Q. 1994. Green reporting in the Republic of South Africa.
Second Edition. Department of Accounting. University of Pretoria.
Sustainability South Africa. (2011a). Sustainability reporting around the world.
Available from:
http://www.sustainabilitysa.org/sustainabilityreporting/Wowwellarecompaniesdoingrep
o. (Accessed 19 February 2011).
Sustainability South Africa. (2011b). Standards for sustainability reporting. Available
from:
http://www.sustainabilitysa.org/sustainabilityreporting/StandardsandGuidelines.aspx
(Accessed 20 March 2011).
Terry, G. (2008). Green. The South African Institute of Chartered Accountants.
Kengray. South Africa. Thirst edition.
Tregidga, H. and Milne, M.J. (2007). From Sustainable Management to Sustainable
Development: A Longitudinal Analysis of External Communication by a Leading
Environmental Reporter. Available from:
http://eprints.otago.ac.nz/4/1/FROM_SUSTAINABLE_MANAGEMENT_TO_SUSTA
INABLE_DEVELOPMENT.pdf. (Accessed 20 November 2009).
126 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Trialogue. (2009). The Sustainability Handbook. July. Sixth edition. Kenilworth. Cape
Town.
Trialogue. (2010). What drives sustainable business today? Issue 1. Sixth edition. April.
Cape Town.
United Nations. (2005). Sustainability Reporting: Public Policy Trends Workshop.
United Nations Environment Programme. Available from:
www.unep.fr/scp/gri/pdf/SectorMeet.pdf (Accessed 17 February 2011).
United Nations Global Compact. (UNGC). (2007). The Global Compact Leaders
Summit 20076. Facing Realities: Getting Down to Business. Geneva. Available from:
http://www.unglobalcompact.org/docs/news_events/8.1/GC_Summit_Report_07.pdf
(Accessed 28 February 2011).
Unerman, J. (2000). Methodological issues – Reflections on quantification in corporate
social reporting content analysis. Accounting, Auditing & Accountability Journal,
13(5):667-681.
Van Altena, R. (2009). Sustainability, corporate social responsibility through an audit
committee lens. IA Adviser, July: 42.
Wheeler, D. and Elkington, J. (2002). The recent history of environmental and social
reporting. SustainAbility. Available from:
http://www.sustainability.com/programs/engaging/history-reporting.asp. (Accessed on
21 November 2002).
Wiseman, J. (1982). An evaluation of environmental disclosure made in corporate
annual reports. Journal of Accounting, Organisation and Society, 7(1):53-63.
World Commission for Environment and Development. (WCED). (1987). Our Common
Future [The Brundland Report]. Oxford. UK: Oxford University Press.
B Marx & V van Dyk 127
ANNEXURE A:
NAMES OF PUBLIC SECTOR ENTITIES IN THE POPULATION WHOSE
SUSTAINABILITY REPORTS WERE INSPECTED
Airports Company of South Africa
Central Energy Fund (CEF)
Development of South Africa
Denel
Eskom
Industrial Development Corporation
Landbank
Post Office
Trans-Caledon Tunnel Authority
Transnet
WD Esterhuysen & M Ward 129
The information content of the Financial
Mail’s “Top Companies” announcements
WD Esterhuysen
Gordon Institute of Business Science, University of Pretoria, South Africa
M Ward
Gordon Institute of Business Science, University of Pretoria, South Africa
Received: February 2011 SAJAR
Revised: August 2011 Vol 25 No. 1
Accepted: October 2011 2011
pp.129 to 143
Investors look beyond accounting and financial data and incorporate factors relating to morality,
society, the environment and corporate governance (inter alia) in making their investment
choices. This study examines the share price performance of South African companies which
best comply with the financial and qualitative criteria as prescribed in the Financial Mail’s “Top
Companies” publication.
Using event-study methodology, the abnormal and cumulative abnormal returns of companies
recommended by analysts as “Top Companies” were examined. Positive, significant excess
cumulative returns were observed for new entrants to the “Top Companies” sample after the
publication date. Thereafter, negative returns were observed for the long-term post-publication
holding period of up to 200 trading days.
The results suggest that any new information related to the criteria in the FM “Top Companies”
publication is of value, but only to short-term traders with low transaction costs. Long-term
investors who buy these shares based on the recommendations of the FM analysts generally
receive below market rates of return, suggesting that once companies have made it into the list,
the value is overstated.
KEY WORDS
Social responsible investment; Information content; Event study
Contact
INTRODUCTION
The market incorporates any perceived price-sensitive information into the estimation
of share prices. This includes historical accounting and financial data, as well as
qualitative criteria which may affect the future performance of share returns.
130 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Included in the qualitative measures, are so-called Responsible Investment (RI)1
criteria. These define an investment strategy that balances financial and social
objectives, which Viviers, Bosch, Smit and Buijs (2008) show is gradually becoming a
mainstream consideration in developed markets. According to Herringer, Firer and
Viviers (2009) broader social investment objectives comprise three main considerations
namely:
Environmental;
Social; and
Corporate governance.
An additional consideration in South Africa would be the promotion of Broad Based
Black Economic Empowerment (BBBEE) initiatives.
Many countries have adopted social and RI initiatives and developed their own indices
that incorporate such principles, such as the FTSE4GOOD, the Dow Jones
Sustainability indices and the South African FTSE/JSE Socially Responsible
Investment (SRI) Index. In the same genre is the Financial Mail (FM) magazine’s “Top
Companies” list. The FM is a weekly financial publication aimed at business decision-
makers and its primary function is to analyse the week's top business stories. However,
once a year the FM publishes a ranked list of 20 companies which meet specific
accounting, financial and qualitative criteria as their “Top Companies”.
Financial criteria used by the FM in their ranking, which account for 40% of the total
score, include the Return on Equity (ROE), internal rate of return (IRR) and compound
growth in earnings per share (EPS). The remaining 60% of the score is based on a
largely qualitative assessment of how "investable" a company is (Williams, 2009).
These criteria include:
how the company is managed;
its corporate governance procedures and culture;
its black empowerment status;
the quality of communication with shareholders and stakeholders;
the prospects for growth in the sector(s) in which the company operates;
contextual issues such as regulatory uncertainties and tax regimes; and
whether the share is reasonably liquid and offers value that the “herd” may have
missed. (Williams, 2009).
The aim of this research is to test the significance of the information content in the
FM’s Top Company list. Since the historical financial information included in the FM’s
analysis is likely to be widely known and already compounded into share prices, our
underlying assumption is that any new information from the FM ranking is likely to be
derived from the inclusion of the qualitative data.
THEORY AND LITERATURE REVIEW
Fama’s (1965 and 1969) “Efficient Market Hypothesis” (EMH) defined three levels of
market efficiency: Strong form EMH; Semi-strong form EMH, and Weak form EMH.
1Also referred to as Socially Responsible Investment (SRI)
WD Esterhuysen & M Ward 131
In a strong-form efficient market, prices will accurately reflect all private and public
information, and investors are unable to create excess returns in the long-term. In a
semi-strong form efficient market, prices adjust rapidly and without bias to new
information that is made available to the public. In weak form efficient markets, prices
do not reflect all public information, and astute investors can generate excess returns by
identifying and exploiting un-priced information.
Thompson and Ward (1995), in a meta-study of accumulated empirical evidence on the
efficiency of the Johannesburg Stock Exchange (JSE) based on studies between 1974
and 1993, concluded that the evidence was at best mixed, particularly regarding weak
and semi-strong form efficiency. They did however, argue that the JSE is operationally
efficient and that it would be reasonable to expect that as statistical techniques became
more sophisticated and powerful, some systematic inefficiencies were likely to be
uncovered, even in a relatively efficient market.
Bhana (1995, 1997, 1998, 1999a, 1999b, 2002, 2003, 2005) tested the share price
reaction on JSE-listed companies to various types of announcements (earnings, special
dividends, potential take-overs, equity financing, foreign listings, layoffs, key executive
dismissals and management buy-outs). In all instances Bhana found evidence of market
inefficiency, suggesting that investors could out-perform by following appropriate
trading strategies, based on new information.
Henn and Smit (1997) found that news events resulted in a 0.006% to 4% movement in
share prices on the JSE. Similarly, in a study conducted on the influence of political
news events on share market activity, van der Merwe and Smit (1997) found South
African political news events explained 1% to 23% of movement in share prices. These
findings were consistent with the findings of Mlambo and Biekpe (2007), who tested
the EMH and found inefficiencies using evidence from ten African stock markets.
With regard to the information content of RI, Herringer, Firer and Viviers (2009) note
that there is a growing body of evidence suggesting that the risk adjusted performance
of RI funds are on par with conventional funds.
Abdo and Fisher (2007) constructed a governance disclosure scorecard, denoted as a “G
Score”, to measure the level of corporate governance disclosure in JSE listed
companies. Over the period 30 June 2003 to 30 June 2006 they found the G Score to be
positively correlated with share price returns. An investment strategy that purchased
shares in the highest G-Score companies for each JSE sector outperformed the index for
the sector. Similarly an investment strategy that purchased shares in the lowest G-Score
companies underperformed the index in terms of annual average return over the 3 year
period.
In a similar study to ours, Mathur and Waheed (1995) tested the stock price reaction to
securities recommended in Business Week’s “Inside Wall Street”. Although these
recommendations were not specifically related to social or RI criteria, they found the
excess returns on the days following the publication were sufficiently large enough to
indicate that institutional traders would gain positive, excess returns (net of transaction
costs) if they were to purchase the newly mentioned shares on the publication date and
sell them 10 trading days later. Thereafter, the excess returns were negative for holding
periods up to 200 days.
132 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Following the literature review, the following null hypothesis was constructed for this
study:
Shareholders earn no cumulative abnormal returns around the announcement dates of
the FM’s “Top Companies”, implying that the information content of the new data is
insignificant.
The alternative hypothesis states that the shareholders do earn significant positive
cumulative abnormal returns around the announcement dates, implying that the
information contained in the data has significant value for investors:
H0: ACARAD = 0
Ha: ACARAD > 0
Here, ACARAD represents the average cumulative abnormal returns around the
publication date.
METHODOLOGY
The population of relevance consisted of all shares listed on the JSE over the period 1
January 2003 to 31 December 2009 that were considered by the FM for their “Top
Companies” list.
Two main criteria are used for including companies in the FM evaluation (Williams,
2009). Their first criterion is that the company should have a market capitalisation of at
least R1bn, so that the investors can be confident that an operation is sustainable and
has critical mass. Their second criterion is a constant track record of internal rate of
return and compound growth in earnings per share over the previous five years.
From this population, a score is derived where 40% of the score is based on historical
financial performance and the remaining 60% is based on a qualitative assessment of
how attractive a company is to invest in (Williams, 2009). The companies with the
highest ranked scores constitute the top 20 companies. Williams (2009) indicates that
corporate governance constitutes 9% and empowerment commitment 12% of the total
score respectively. The attractiveness of a company to invest in constitutes 12%2, and is
explained by Theobald (2003) as “volumes traded and value buy at current prices”. The
rest of the categories of the qualitative components of the score relate to an assessment
of future financial performance and the factors which influence that. It is not clear what
actual weighting can be finally attributed to specifically social or RI criteria, but we
assume that the FM’s attempt to move away from purely financial criteria will focus
investor’s minds on the information content of the qualitative criteria.
Three samples were determined for the study.
2A comparison with Theobald (2003) indicates that the weightings have shifted from when first
implemented in 2003 to the weightings in 2009. Investor communication (2003, 6%) is not listed as a
criteria in 2009. Industry profit prospect was 15% in 2003 is now 9%. Empowerment commitment was
not a criterion in 2003, but now constitutes 12%. It is not clear what impact these changes in weights
have on the study.
WD Esterhuysen & M Ward 133
1. The “full sample” refers to the complete list of the “Top 20 Companies” published
in the FM between the period 2003 and 2009. This sample contains 140
observations.
2. The “new entries” sample comprises all companies that entered the list for the first
time. This sample contained 83 observations.
3. The “repeated entries” sample comprises all companies that featured more than
once in the list since 2003. This sample contained 57 observations.
For the analysis of the abnormal returns, daily share price data was collected for each of
the companies mentioned in the list from the McGregor’s BFA database over the period
from 20 trading days prior to the announcement to 200 trading days following the
announcement. Betas for each share were estimated against the all share (J203) index
returns, using five years of prior share return data.
The share price reaction to inclusion in FM’s “Top Companies” list was tested utilising
the standard event study methodology developed by Fama, Fisher, Jensen and Roll
(1969). McWilliams and Siegel (1997) suggest that event study methodology has
become popular because it does not rely on analysing accounting-based measures of
profit, which have been criticised as not reflecting the true performance of firms. They
also assert that the event study framework provides a true measure of the financial
impact of an event only if a set of assumptions are valid and the research design is
properly executed. The assumptions are: firstly, that markets are efficient; secondly, that
the event was unanticipated; and thirdly that there were no confounding effects during
the event window.
McWilliams and Siegel (1997) further illustrate how an event study is implemented
using 10 steps. Steps one to five focus on defining and isolating the event. Step six
refers to the measurement of the price adjustment and steps seven to ten relate to the
statistical testing of the price adjustment for significance.
Price adjustments are measured as abnormal returns. Abnormal returns can be measured
in the short-term or long-term where the abnormal return or “residual” represents the
share price return after subtracting the expected return of that share. While the exact
definition of long-term is arbitrary, it generally applies to event windows of 1 year or
more (Khotari and Warner, 2006).
The Capital Asset Pricing Model (CAPM) has most frequently been used to calculate
expected returns. The CAPM has been criticised widely over the last two decades on the
grounds that a single factor beta model provides little explanation of the cross-section of
expected share returns.
Fama and French’s (1996) three-factor model, which they claim explains expected
returns more accurately than a single parameter CAPM, assumes share return sensitivity
to three factors:
the excess return on the broad market;
the difference between the return on a portfolio of small capitalisation stocks and
the return on a portfolio of large stocks, and
134 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
the difference between the return on a portfolio of high-book-to-market stocks and
the return on a portfolio of low-book-to-market stocks.
Ward and Muller (2010) also note the inadequacy of a market or single parameter
CAPM as a benchmark against which abnormal returns are estimated. They indicate
that this is due to the inability of the CAPM to account for expected returns related to:
company size, growth versus value stocks and in the South African context, a further
consideration of “resource” versus “non-resource” shares. Ward and Muller (2010)
make use of a 12 parameter “style” model to estimate expected share returns. After
allocating all JSE listed shares into one of twelve control portfolios, they calculate the
alpha and beta coefficients of each share against each of the control portfolios in a
multiple regression equation, updated quarterly. These parameters are then used to
measure the expected return of each share, and hence the abnormal returns. The average
abnormal return across the sample can then be used for the event analysis.
Following Ward and Muller (2010) both the control portfolio model and CAPM were
used in this event study to estimate abnormal returns.
The event date for the purpose of this study is regarded as the day on which the “Top
Companies” section of the FM is published. This date is denoted as “t0”.
The impact of the announcement was measured in daily returns on shares for each of
the included companies, over a period of 221 days; from the published date t0 backward
for 20 days to t-20, and forwards for 200 days to t200.
The daily share price return for each share was calculated in terms of Equation 1.
Rit = ln [Pit / Pit-1] … (1)
where:
Rit = the rate of return on share i on day t, and
Pit = the price of share i at the end of day t.
For the CAPM, the abnormal return for share i on day t, ARit, was estimated as:
ARit = Rit – (αi + βiRmt) … (2)
where:
αi and βi = the (daily) return estimates for the market model parameters for share i,
and
Rmt = the return on the JSE all share index (ASI) for day t.
Using the control portfolio model the abnormal return for share i on day t, was
estimated as:
WD Esterhuysen & M Ward 135
ARit = Rit – (αit + βi,1SGNt + βi,2SGRt + βi,3SVNt + βi,4SVRt +
βi,5MGNt + βi,6MGRt + βi,7MVNt + + βi,8MVRt +
βi,9LGNt + βi,10LGRt + βi,11LVNt + βi,12LVRt) … (3)
where:
αit = the (daily) alpha intercept term of share i on day t, and
βi,1…..βi,12 = the beta coefficients on each control portfolio return and
SGNt…LVRt are the log-function share price returns on each of the twelve control
portfolios explained in Table 1 on day t.
Table 1: Control portfolios
Control Portfolio Company size Value or growth
company
Resource or non-
resource company
SGN Small Growth Non-resource
SGR Small Growth Resource
SVN Small Value Non-resource
SVR Small Value Resource
MGN Medium Growth Non-resource
MGR Medium Growth Resource
MVN Medium Value Non-resource
MVR Medium Value Resource
LGN Large Growth Non-resource
LGR Large Growth Resource
LVN Large Value Non-resource
LVR Large Value Resource
To test the performance on a specific date the average abnormal return, AARt, is
calculated as:
AARt =
n
i
itARn 1
1 … (4)
where:
AARt = the average abnormal return for all shares on day t, and
n = the number of companies.
To test the performance of a share for each event window, the abnormal returns were
accumulated to obtain the cumulative abnormal return (CAR).
CARi =
d
t
itAR20
… (5)
where:
136 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
CARi = the cumulative abnormal returns for share i for the period from t = –d to t = d
d = the number of days before and after the event window period
Once all the cumulative abnormal returns (CARs) for the full sample were calculated,
the average cumulative abnormal return (ACAR) is calculated as the simple average
CAR of the shares in the sample.
ACAR =
n
i
iCARn 1
1 … (6)
where:
ACAR = the average cumulative abnormal return for all shares in the sample for the
period from t = -d to t = d,
and
n = the number of companies
A two tailed t-test was performed at the 5% confidence level to determine whether the
ACAR was significantly different from zero around the publication date. In addition to
the t-test, a boot-strapping process was used to test the significance of 10 day ACARs.
From the sample selection process, some data integrity issues arose:
Companies that were de-listed during the event period were excluded from the
analysis.
Shares with a daily return in excess of |30%| in the event window were
investigated for data errors (which were corrected), otherwise the share was
retained.
Companies with missing or insufficient share price data were excluded from
the analysis.
RESULTS
A summary of the sample is presented in Table 2.
WD Esterhuysen & M Ward 137
Table 2: Summary of the Top 20 companies list included in this study
Population size (companies (including repeats) which met the criterion of
market cap>R1bn) 1056
Sample Size 140
Number of companies in list by year 140
2003 20
2004 20
2005 20
2006 20
2007 20
2008 20
2009 20
JSE Sectors
Number of different sectors on the JSE 9
Number of different sub-sectors on the JSE 45
Frequency of number sectors 140
Basic Materials 38
Consumer Goods 7
Consumer Services 32
Financials 17
Health Care 7
Industrials 23
Oil and Gas 3
Technology 3
Telecommunications 10
Number of companies repeatedly in list sum product = 57
7 times 1
6 times 1
5 times 0
4 times 3
3 times 9
2 times 19
New entries / Repeated entries / Total - in list
2003 20 / 0 / 20
2004 13 / 7 / 20
2005 10 / 10 / 20
2006 12 / 8 / 20
2007 12 / 8 / 20
2008 4 / 16 / 20
2009 12 / 8 / 20
Totals 83 / 57 / 140
138 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
As mentioned above, both the CAPM and Control Portfolio Model (CP) were used to
calculate ARs. The ACARs across the 220 day event window resulting from each of the
two models are shown in Figure 1.
Figure 1: ACARs for the CAPM and control portfolio models
From figure one, a significant, positive trend in the data is observable post the event
date in the ARs derived from the CAPM benchmark. Since this trend is not evident in
the CP based data, it was concluded that the CAPM approach did not fully account for
style related returns in the AR generating process. For this reason it was decided to use
only the CP model’s ARs for further analysis.
Over the duration of the 221-day event window, the “new entries” sample initially
performs well. In the first nine days following the announcement, the ACARs reach
2.05%, before steadily declining to a low of -2.46% on day 144, and ending at -0.28%
200 trading days after the announcement. The “repeat entries” sample shows an erratic
ACAR, which is never more than |0.74%| from the benchmark. The ACAR for “repeat
entries” sample ends on -0.14% on day t200.
Figure 2 presents a more detailed view of the 21-day ACARs over the event window for
each of the three samples. The window commences on t-10, (10 days before the
announcement date), and ends on t+10, (10 days after the announcement date).
WD Esterhuysen & M Ward 139
Figure 2: Average Cumulative Abnormal Returns [-10, +10]
Figure 2 shows that the ACARs are positive 10 days after the announcement for new
entries into the FM Top Companies list. To test for significance, both the usual t-test
and a boot-strap3 based test were used. The results are presented in Table 3.
Table 3: Statistically significant CARs for the event window [-10, +10]
Full list New entries Repeated entries
10 day CAR
on t10
0.30% 0.87% -0.53%
T-test 0.648 1.559 -0.680
T-critical (5%) 1.980 1.994 -2.012
H0: μ=0 (5%) Fail to reject Fail to reject Fail to reject
Bootstrap
upper bound 0.83% 0.83% -1.06%
H0: μ=0 (5%) Fail to reject Reject Fail to reject
3The boot-strapping method of significance testing is superior to the t-test in that no assumption is
made of normality. The Boot-strap distributions were constructed for 10 day ACARs, against which the
10 day ACAR in the event period was tested for significance (Ward and Muller, 2010).
140 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
For the 10 day ACARs evaluated in the 21 day event window, a significant positive
ACAR of 0.87% was found for the “new entries” sample on day t10 using the bootstrap
approach. Since the bootstrap distribution is the more appropriate distribution of 10 day
ACARs, we find this result compelling, and further note that the t-test would also
support this at a 10% level.
For the hypothesis test of the ACARs over the full event window, only the t-test value
on the last day of the event window (t=200) was evaluated for each sample, and in all
cases, the t-test results were insignificant.
For the 221 day event window, the study found a peak ACAR of 2.05% nine trading
days after the publication date, in the “new entries” sample. This is higher than the peak
of 0.96%, 11 days after the publication date, for the full sample and 0.60%, 80 days
after the publication date, for the “repeated entries” sample. For the full 200 trading
days following the publication date, the “new entries” sample performed the worst, with
an ACAR of -2.28%, compared to the “repeated entries” sample’s ACAR of -0.14%,
although none of the results at t200 were significant.
Mathur and Waheed (1995) found surprisingly similar results when testing the stock
price reaction to securities recommended by analyst’s in Business Week’s “Inside Wall
Street”. The excess returns on the 10 days following the publication were sufficiently
large to indicate a successful short-term trading rule. They also noted that thereafter the
long-term performance deteriorated, for holding periods of up to 200 trading days after
the publication date.
In their study, on whether a great company can be a great investment, Anderson and
Smith (2006) tested the “classic mistake” of confusing great companies with great
investments. Their initial presumption, that a company’s well-known virtues are already
factored into the price of the company’s shares, was proven incorrect, clearly
contradicting the EMH. We report similar results, but only to the extent that the “new
entries” sample outperformed the market by a statistically significant margin in the
short term.
There is no compelling explanation why the “new entries” sample outperformed the
“repeated entries” sample, other than the fact that inclusion into the FM list of “Top
Companies” has information content for new entrants which is thereafter priced by
investors. Once included in the list, no further value is noted for those companies
remaining in the list.
CONCLUSION
The results of this study on the market reaction to the publication of the “Top
Companies” issue of the FM indicate significant, positive excess (abnormal) returns on
the first 10 trading days subsequent to the publication date, for companies entering the
list for the first time.
Two methodologies were used in this study to estimate abnormal returns, namely the
CAPM and the Control Portfolio model. A persistent upward trend in the ACARs
indicated a bias in the CAPM approach, in that the effects of the market were not fully
controlled for in the CAPM. The CP approach was preferred, this result confirming
similar findings relating to the event-study methodology in Ward and Muller (2010).
WD Esterhuysen & M Ward 141
A key assumption in this study was that any new information contained in the FM “Top
Companies” announcement was primarily related to factors other than financial
indicators. The premise (stated above) was that financial information would already be
fully priced into share prices, and that any new price-sensitive information would
therefore relate to the 60% weighting placed by the FM on the qualitative factors
relating to investability, management, growth, communication and corporate
governance (inter alia). The similarity between our results and those of Mathur and
Waheed (1995) raises some questions as to the extent to which the financial data
influences the analysis. Mathur and Waheed’s (1995) findings were based on
BusinessWeek analyst’s recommendations, and made no claim to factor in qualitative
issues. Since it is impossible to distinguish between the input factors (despite the
weightings used), it is possible that our results are confounded by the financial data, and
may not be a pure reflection of the value from the more qualitative criteria. The fact that
our analysis only finds significant positive returns for companies entering the FM list
for the first time, does however indicate that investors do value the assessment of the
FM process.
Whilst extensive international research relating to social, RI and other qualitative
criteria exists, this is a relatively new field of study for the Johannesburg Stock
Exchange. Further studies in this field, which focus specifically on South African
market related factors (environmental, social, corporate governance and Black
Economic Empowerment) will add to the current body of knowledge.
REFERENCES
Abdo, A. and Fisher, G. (2007). The impact of reported corporate governance
disclosure on the financial performance of companies listed on the JSE. Investment
Analysts Journal, 66, 43-55.
Anderson, J. and Smith, G. (2006). A great company can be a great investment.
Financial Analysts Journal, 62 (4), 86-93.
Bhana, N. (1995). The share market reaction to earnings announcements - a test of the
efficiency of the Johannesburg Stock Exchange. Investment Analysts Journal, 42, 45-
57.
Bhana, N. (1997). Price adjustments on the JSE for announcements of share (stock)
dividends. Investment Analysts Journal, 46, 35-44.
Bhana, N. (1998). The share price reaction on the Johannesburg Stock Exchange for
special (extra) dividend announcements. Investment Analysts Journal, 47, 5-15.
Bhana, N. (1999a). Share price reaction to announcements of equity financing by
companies listed on the Johannesburg Stock Exchange. Investment Analysts Journal,
48, 33-42.
Bhana, N. (1999b). The impact of public news regarding potential take-overs on the
share price behaviour of target companies. Investment Analysts Journal, 50, 29-41.
142 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Bhana, N. (2002). Layoff announcements, share price reaction and long-term financial
performance on the JSE Securities Exchange. Investment Analysts Journal, 56, 41-56.
Bhana, N. (2003). The share price reaction to announcements of key executive
dismissals by companies listed on the JSE Securities Exchange. Investment Analysts
Journal, 58, 29-50.
Bhana, N. (2005). The share price reaction to management buyout announcements of
companies listed on the JSE Securities Exchange. Investment Analysts Journal, 62, 19-
30.
Fama, E. F. (1965). The behavior of stock-market prices. The Journal of Business, 38
(1), 34-105.
Fama, E. F. (1969). Efficient Capital Markets: A review of theory and empirical work.
The Journal of Finance, 25 (2), 383-417.
Fama, E. F., Fisher, L., Jensen, M. C. and Roll, R. (1969). The adjustment of stock
prices to new information. International Economic Review, 10 (1), 1-21.
Fama, E.F. & French K.R. (1996). Multifactor explanations of asset pricing
anomalies. The Journal of Finance, 51 (1), 55–84.
Henn, J. and Smit, E. (1997). The influence of economic news events on share market
activity in South Africa. Investment Analysts Journal, 46, 23-34.
Herringer, A., Firer, C. and Viviers, S. (2009). Key challenges facing the socially
responsible investment (SRI) sector in South Africa. Investment Analysts Journal, 70,
11-26.
Khotari, S. P. and Warner, J. B. (2006). “Econometrics of Event Studies”. In Handbook
of Corporate Finance: Empirical Corporate Finance (Vol. A, CH1). Dartmouth:
Elsevier / North-Holland.
Mathur, I. and Waheed, A. (1995). Stock price reactions to securities recommended in
Business Week’s “Inside Wall Street”. The Financial Review, 30 (3), 583-604.
McWilliams, A. and Siegel, D. (1997). Event studies in management research:
Theoretical and empirical issues. Acadamy of Management Journal, 40 (3), 626-657.
Mlambo, C. and Biekpe, N. (2007). The efficient market hypothesis: Evidence from ten
African stock markets. Investment Analysts Journal, 66, 5-18.
Theobald, S. (2003). Top Companies Survey 2003. Financial Mail, 27th
June.
http://www.topcompanies.co.za/topcos2003/stories/tc1.htm. Accessed 23 August 2011.
Thompson, A. R. and Ward, M.J. (1995). The Johannesburg Stock Exchange as an
efficient market: A review. Journal for Studies in Economics and Econometrics, 19 (3),
33-63.
WD Esterhuysen & M Ward 143
Van der Merwe, R. and Smit, E vd M. (1997). The influence of political news events on
share market activity in South Africa. Investment Analysts Journal, 44, 11-23.
Viviers, S., Bosch, J. K., Smit, E. vd M. and Buijs, A. (2008). The risk-adjusted
performance of responsible investment funds in South Africa. Investment Analysts
Journal, 68, 39-56.
Ward, M. and Muller, C. (2010). The long-term share price reaction to Black Economic
Empowerment announcements on the JSE. Investment Analysts Journal, 71, 27-36.
Williams, D. (2009, June 26). An act worth emulating. Financial Mail , 10-12.
E van Wyk 145
A Note: The SAICA Part I Qualifying
Examinations: Factors that may influence
candidates’ success
E van Wyk
Department of Accountancy, Stellenbosch University
Received: October 2010 SAJAR
Revised: June 2011, July 2011, September 2011 Vol 25 No. 1
Accepted: September 2011 2011
pp.145 to 174
The study investigates certain factors characterising successful and unsuccessful candidates in
the 2009 and 2010 SAICA Part I Qualifying Examination (QE I) with the purpose of making
recommendations to candidates, educators, training centres and to SAICA. The study also aims
to serve as basis for future studies as to how the success rate of candidates in the QE I might be
improved.
Successful candidates tend to attempt the QE I at a young age, preferably prior to getting
married and starting a family. Candidates’ chances of passing decrease with the number of exam
attempts. Candidates with a first language of English or Afrikaans are more likely to be
successful when writing the QE I. Finally, attending a Board Course prior to writing the
examination might improve candidates’ success rate.
SAICA, in collaboration with relevant RTO/ATO training centres, may consider
providing/obtaining financial assistance to/for candidates who may not be able to attend Board
Courses. The effect on candidates of having to study and attempt the QE I in a language other
than their first language should be investigated and addressed. One way of addressing the issue
may be that SAICA may consider making the QE I papers available in the two main black
languages isiXhosa and isiZulu.
Factors affecting the academic performance of candidates deserve more research. These include
their scholastic background including the resource base of schools, candidates’ home
environment, socio-economic background and possible inferior high-school education.
KEYWORDS
SAICA QE I SAICA Part I Qualifying Examination
Qualifying examination First language
Examination success Examination attempts
South African Institute of Chartered Accountants Examination time management
Board Course attendance Stress under exam conditions
Academic workload
Contact
146 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
INTRODUCTION
Several studies have been published on factors that may affect the success of candidates
in undergraduate, post-graduate or professional accounting examinations. The majority
of these studies, such as those by Alfan and Othman (2005); Black and Duhon (2003);
De Lange, Waldmann and Wyatt (1997); Du Plessis, Müller and Prinsloo (2005);
Gierlaskinski (2003); Gracia and Jenkins (2002); Koh and Koh (1999); Moses (1987);
Toh (1994) and Tyson (1989), investigate factors influencing undergraduate student
performance. Boone, Legoria, Seifert and Stammerjohan (2006); Brahmasrene and
Witten (2001); Roos (2009); Sadler and Erasmus (2005); and Titard and Russell (1989)
document factors affecting the success of candidates in post-graduate and professional
accounting examinations.
The success of candidates in the professional examinations set by the South African
Institute of Chartered Accountants (SAICA) has become a topical issue due to the
shortage of chartered accountants (CA(SA)s) currently experienced in South Africa
(SAICA, 2010a). In 2008 South Africa’s financial management and auditing sector was
in need of approximately 22 030 accountants (Mail & Guardian Online, 2008). As the
global financial crisis heightened the demand for accountants’ services increased. In
addition, approximately 1.4% of CAs leave South Africa every year (Innocenti, 2009).
SAICA is responsible for the training of competent, entry-level CAs suitable for the
South African and the international workforce. Consequently, it also plays a significant
role in both South Africa’s business sector and in the development of the economy as a
whole (SAICA, 2010a). However, for a large number of candidates, qualifying as a
CA(SA) is a long and difficult process which poses some obstacles. One of these
obstacles is the successful completion of the Part I Qualifying Examination (QE I).
The aim of the study is:
to identify underlying factors, by investigating the relevant literature, that may
affect the success of candidates in accounting examinations in general;
to collect and statistically analyse data, regarding these underlying factors from
candidates in the 2009 and 2010 Part I Qualifying Examinations (2009 QE I and
2010 QE I respectively; or QE I when referred to from a general perspective);
to ascertain which factors characterise candidates who were successful and
unsuccessful in the 2009 QE I and 2010 QE I examinations;
to construct a set of characteristics, and the possible underlying factors which
cause them; and
to make recommendations to candidates, educators, training centres and to SAICA
to serve as basis for future studies as to how the success rate of candidates in the
QE I might be improved.
Prior to the identification of the underlying factors and analysis of the data obtained,
background is supplied on why students choose the CA(SA) qualification as a career
path. Academic and training requirements pertaining to CA(SA) candidates are also
discussed.
E van Wyk 147
BACKGROUND Benefits of the CA(SA) qualification
Candidates pursue a career as a CA(SA) due to various reasons. Myburgh (2005) lists
availability of employment, prestige, employment security, lifestyle and the social
status of the profession as the most important benefits of attaining the CA(SA)
qualification. Other studies by Inman, Wenzler and Wickert (1989) and Chan and Ho
(2000) list career mobility, development, career advancement and partnership
opportunity as important factors affecting candidates’ decision to pursue a career in
accounting.
In addition, studies by Tan and Laswad (2006); Mauldin, Crain and Mounce (2000);
Auyeung and Sands (1997) and Felton, Burh and Northey (1994) have shown that
accounting students’ discipline choice has traditionally been heavily influenced by
earnings potential and job market conditions and opportunities. Notwithstanding the
fact that a career as a CA(SA) provides some or all of the aforementioned benefits,
candidates have to undergo extensive education and training in order to attain this
qualification.
Academic and training requirements of CA(SA) candidates
The education and training requirements (on a full-time basis) to become a CA(SA) are
as follows:
a relevant four-year degree at a SAICA accredited university;
completion of a 36-month training contract with an accredited training centre:
either an Approved Training Organisation (ATO) or a Registered Training Office
(RTO);
completion of a post-graduate diploma or specialised course in auditing or
financial management;
successful completion of the two parts of the Qualifying Examination (QE) set by
SAICA and the Independent Regulatory Board for Auditors (IRBA); and
registration with SAICA after passing both qualifying examinations and
completing a training contract (SAICA, 2010b).
In certain instances, candidates complete a 60-month training contract (as opposed to a
36-month training contract) with either an ATO or a RTO. Such candidates complete
their studies on a part-time basis simultaneously while completing their aforementioned
training.
The following sections provide a more detailed discussion of the academic and training
requirements.
148 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Academic requirements
Candidates have to successfully complete both the QE I, and the QE II or the
Professional Practice Examination (PPE). The QE I is a written, standard-setting
examination which tests the core competence of candidates. This examination covers
the four main disciplines of financial accounting, auditing, taxation and management
accounting and finance (SAICA, 2010b).
The QE II (Financial Management specialism) and PPE (Auditing specialism) are
written after a minimum of 18 months of the candidate’s training contract and after the
candidate’s successful completion of a selected programme in Financial Management or
Auditing (SAICA, 2010b).
Training requirements: Auditing specialism
Candidates who wish to specialise in auditing must complete their training contracts
with a RTO. Upon completion of their training contracts and all other qualification
requirements, candidates are able to perform the attest function and sign an audit report
(SAICA, 2010b). This specialism was previously termed training in public practice
(TIPP).
Training requirements: Financial management specialism
Candidates who wish to specialise in financial management must complete a training
contract with an ATO (SAICA, 2010b). This specialism was previously termed training
outside public practice (TOPP).
Literature review
Several studies have been performed on factors associated with academic performance
of candidates in examinations at university and professional level. Educators strive to
encourage and support increased participation in higher education. In order to do so
successfully, information on factors affecting the success of candidates must be
acquired and potential problems addressed (Byrne & Flood, 2008).
While the findings of earlier studies have not been wholly consistent, certain dominant
factors have emerged. These include, inter alia, prior academic achievement and gender
(Byrne & Flood, 2008). Roos (2009) mentions a number of personal characteristics of
candidates that may have an influence on their success rates in professional accounting
examinations. These are age, gender, available study time, the candidates’ first language
and whether candidates hold other advanced qualifications. As will emerge in the
methodology section of this paper, the underlying factors causing such characteristics
are investigated along with other factors which may be more relevant to the CA(SA)
qualification process. A literature review assists in elucidating the relevance of the
factors investigated.
Age
Existing studies have provided conflicting outcomes regarding the age of candidates
and the effect this has on candidates’ success in examinations. Brahmasrene and
Whitten (2001) found that older candidates were more likely to pass the United States
E van Wyk 149
Certified Public Accountancy (US CPA) examination, which is a professional
accounting examination in the United States of America (USA).
However, the study by Koh and Koh (1999) found that younger students performed
better than their older counterparts and that mature students had difficulty settling into a
study routine and examinations. In addition, while older students might be more
motivated and disciplined, they tend to have families and therefore less time to focus on
academic matters (Gierlaskinski, 2003).
Du Plessis et al. (2005) found that South African students younger than 30 years were
more likely to pass a first-year accounting distance education course than students older
than 30 years. In a study of the factors affecting success in Chartered Institute of
Management Accountants (CIMA) examinations, Roos (2009) found that younger
candidates were more successful than older candidates.
Certain underlying factors which may be associated with the age of candidates may
affect their performance. Older students may perform better due to their work
experience and maturity, while younger students may perform better as they are ‘fresh
out of school’ (Koh & Koh, 1999). Considering this fact for candidates attempting the
QE I, it is more likely that younger students might complete their full-time studies
immediately prior to the QE I, while older students complete their studies part-time
and/or do not attempt the QE I immediately after completing their undergraduate and
post-graduate studies for various possible reasons. These reasons could be financial
constraints and/or other personal matters. In addition, the age of older students might
indicate lack of effective study skills causing them to take longer to complete their
studies.
Gender
Byrne and Flood (2008) found that, while the impact of gender on academic
performance has been investigated in several studies, gender is not significantly
associated with first year academic performance. On the other hand, Brahmasrene and
Whitten (2001) found that male students were more likely to pass the US CPA
examinations than females. Gierlaskinski (2003) also found males to perform better in
accounting courses than females, as did Koh and Koh (1999). However, research by
Alfan and Othman (2005), Mutchler, Turner and Williams (1987), Tho (1994) and
Tyson (1989) found statistically significant better performances by female students.
As in the case of the age factor, different studies produced different results. De Lange et
al. (1997) found males to outperform females in undergraduate accounting courses,
while Gammie, Paver, Gammie and Duncan (2003) found no significant difference
between the performance of male and female students. Furthermore, Gammie et al.
(2003) state that constructed gender, or gender identity, rather than biological gender,
may be a greater determinant of academic success. Gender identity is the way in which
an individual identifies with a gender category, for example as being either female or
male, or in some cases being neither. In most societies there is a basic division between
male and female genders, that are understood to be determined by biological sex, but in
all societies some individuals do not identify with the gender that is otherwise
associated with their biological sex (Wikipedia, 2011). Alfan and Othman (2005) and
Gierlaskinski (2003) state that further research is required as an increasing number of
females are entering accounting courses.
150 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
A more recent study of the Southern African region by Roos (2009) found that females
performed better than males in the CIMA examinations. However, it was also noted that
the result was not statistically significant.
Available study time
The availability of study time is determined by several factors, such as family
commitments (indicated by marital status and children) and employment (Sadler &
Erasmus, 2005). Older candidates are more likely to be married and may also have
children, factors which decrease study time availability (Gierlaskinski, 2003).
In a study of CIMA candidates, Roos (2009) used the following factors as determinants
of study time availability:
the number of papers written at a CIMA examination sitting;
marital status; and
the number of children.
The first factor, the number of papers written, is not applicable in this study as
candidates for the QE I do not have the option of writing fewer or more examination
papers, as is the case in the CIMA examinations. Much earlier research showed that
marriage does not appear to be a major factor in scholastic achievement (Samenfink &
Milliken, 1961). In addition, Roos (2009) found that marital status and the number of
children a candidate had did not have a significant effect on the success of the candidate
in CIMA examinations.
However, having children might cause candidates to have greater family commitments
and pose difficulties in settling into an effective study routine. Craig and Bittman
(2005) emphasise that, when having children, the social problem of balancing work and
family is more about significant time constraints than about financial constraints. The
calculation of the difference in time commitment between families with no children and
families with different numbers of children gives a measure of heavy time pressure on
the latter (Craig & Bittman, 2005).
Employment
The factor of employment must be considered from two perspectives: the potentially
adverse effect it may have on study time availability, and the potentially beneficial
effect employment experience may have on examination success.
Moses (1987) found employment experience to be a contributing factor in attaining
success in accounting examinations in the USA. In addition, another study in the USA
found that, using age as a proxy for work experience older students were more likely to
pass US university business examinations (Black & Duhon, 2003). The relevance of
work experience was confirmed by Brahmasrene and Whitten (2001), who indicated
that accounting experience added to the success of students, compared to those
candidates with no accounting experience.
Research by Titard and Russell (1989) showed that public accounting experience helped
candidates in certain aspects of the USA CPA examinations. Higher pass rates were
E van Wyk 151
achieved in Auditing and Practice, while lower pass rates were achieved in Law and
Theory.
Relevant studies on the effect of employment on pass rates in Accountancy
examinations have not been done in South Africa. Furthermore, a recent Southern
African study by Roos (2009) found that the results regarding employment could not be
analysed meaningfully. In that particular study 94% of respondents were employed.
Owing to the large percentage of respondent’s only representing one category, an
analysis of the relationship between employment and examination results could not be
performed in that particular study (Roos, 2009).
When one considers the effect of employment on the success rate in the QE I, it is worth
noting that the majority of candidates who attempt the QE I are employed full-time
under a training contract with a RTO/ATO organisation. For most candidates, this
training period commences immediately after completing their relevant studies at a
SAICA accredited university. This makes them ‘fresh out of school’ as suggested by
Koh and Koh (1999).
Language
Several studies have found that candidates who were not examined in their first
language performed more poorly than those who were granted the opportunity to
attempt the examination in their first language. Gierlaskinski (2003) found that students
in the USA with English as a second language, while working hard at their studies,
found language to be an obstacle in accounting courses. This obstacle proved to be
significant by the statistical results obtained from the study (Gierlaskinski, 2003).
The CIMA examinations are United Kingdom-based examinations and are only
available in English (Roos, 2009). Roos (2009) found that English second-language
candidates for the CIMA examinations were not more likely to fail their examinations
in comparison with students with English as a first language. Another corroborative
study found that the factor of first language does not have a significant impact on the
likelihood of passing or failing a professional examination such as university
Management Accounting (Jackling & Anderson, 1998).
However, a recent South African study amongst CA trainees found that 33% of black
candidates preferred to study and write examinations in their first language if given the
choice. In addition, these students were of the opinion that students who wrote
examinations in their first language had an advantage in expressing themselves more
easily (Wiese, 2006). Confirming this, Gamaroff (1995) found that using English as a
medium of instruction to black students limits their ability to do an analytical task.
Prior academic achievement
Prior academic achievement was found to be a highly significant predictor of success in
first-year accounting courses (Byrne & Flood, 2008). Furthermore, Koh and Koh (1999)
found that prior academic achievement was a significant determinant of success
throughout students’ three years of university level study.
When one considers the effect of prior academic achievement on the success rate of
candidates in the QE I, certain underlying factors need to be taken into account. These
152 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
include candidates’ scholastic background and also whether they completed their
undergraduate and/or post-graduate studies by means of distance-learning (and
therefore most probably on a part-time basis), whether they completed these on a full-
time basis, and the quality of the secondary and tertiary institution(s) they attended.
Phurutse (2005) suggests that several factors affect the performance of learners in South
Africa’s public schools. These include the resource base (funding) of schools, class size
and the formal contact hours of educators and learners. In addition, it was found that
schools in the Eastern Cape, Limpopo and the rural areas of Kwazulu-Natal were the
poorest and had the highest number of learners in their schools. The study found these
schools to be predominantly black schools. While this study suggests that poor
communities are characterised by poor educational contexts, causality should not
merely be assumed as improving the quality of education depends on a host of different
factors (Phurutse, 2005).
The quality of the tertiary institution (university) was also found to be a strong predictor
of students’ results. A high quality tertiary institution may compensate for other
potential negative factors that a student may bring to tertiary study. These factors
possibly include an unsupportive home environment, low socio-economic background
and inferior high-school education (Parker, 2010).
Prior academic achievement could also be evaluated in terms of candidates already
having attained other advanced degrees. Titard and Russell (1985) demonstrated that
candidates with other advanced degrees achieved higher success rates in CPA
examinations compared to those without such degrees. This finding is further supported
by studies by Boone et al. (2006) and Roos (2009).
Examination attempts
Research suggests that candidates attempting undergraduate accounting courses for a
second time are more likely to fail compared to first-time candidates (De Lange et al.,
1997). This was confirmed by Roos (2009), who found that CIMA candidates tend to
have higher success rates at their first attempt in writing the examinations compared to
repeat candidates.
The number of examination attempts is positively related to the age of candidates who
are unsuccessful in previous attempts. This might indicate that the underlying factors
causing older candidates to more likely fail the QE I, could also be relevant when
evaluating the impact of the number of examination attempts on success in the QE I. As
mentioned previously, these underlying factors include having family commitments for
example being married and having children, as well as poor study skills causing them to
take longer to complete their studies (Gierlaskinski, 2003).
Board course attendance
Certain South African universities and a number of private institutions provide ‘Board
Courses’ to candidates planning to write the QE I. Board Courses are similar to the type
of ‘CPA exam coaching courses’. These ‘coaching courses’ are normally presented to
candidates prior to writing the examination and are found to contribute to candidates’
success in professional accounting examinations (Journal of Accountancy, 1983).
E van Wyk 153
Titard and Russell (1989) found that formal supplementary study courses contributed
positively to examination success. The study by Roos (2009) confirmed this, finding
that candidates who attended part-time tuition classes aimed specifically at CIMA
examinations outperformed their peers.
The two major Board Courses on offer to candidates in the QE I are the Cape Town
Board Course presented by the University of Cape Town (UCT) and the Gauteng Board
Course presented by the University of Johannesburg (UJ). Accessibility and financial
means might be reasons for candidates electing to attend or not to attend Board Courses.
Candidates who contracted to complete their training periods with certain of the larger
RTO/ATO training centres, are encouraged and financially assisted to attend the UCT
and UJ Board Courses (PWC, 2011; Ernst & Young, 2011).
Other factors
Boud (1988) states that candidates with a high level of personal responsibility may be
more involved in the subject studied, which is regarded as an attribute of effective
learning. In addition, Gracia and Jenkins (2002) found that candidates tend to fail
accounting courses as a result of poor memory, lack of adequate revision, non-
attendance (of lectures), lack of motivation, encouragement, interest in the subject and
poor time-management.
Research by Macan, Shahani, Diphoye and Phillips (1990) found that females scored
significantly higher on the overall time management scale than males. On the other
hand, it was found that males were perceived to be more in control of their time than
females. Perceived control of time was also found to be related to lower levels of stress.
This indicates that stress may be related to the abovementioned factors, either as a
causal factor or as a result of issues such as academic coursework overload (Macan et
al., 1990). Furthermore, Anderson and Cole (2001) stated that several factors may cause
stress, which in turn may cause adverse academic performance and cause candidates to
suffer exhaustion. Of these factors, the following were rated highest:
intimate relationships;
parental expectations; and
finances.
After considering the possible factors affecting success in the QE I, the research
methodology is provided next.
RESEARCH METHODOLOGY
The research methodology applied in this study is provided below.
Ambit of the study
The study focuses on the QE I examination set by SAICA. A web-based questionnaire
was sent to the total population of candidates who attempted the 2009 and 2010 QE I
examinations.
The study aims, inter alia, to explore factors that characterise candidates who were
successful and unsuccessful in the 2009 QE I and 2010 QE I examinations. These
154 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
characteristics might serve as a basis for further research into the possible factors that
cause them. Consequently, this study does not endeavour to address these underlying
factors, but to make recommendations to candidates, educators, training centres and to
SAICA in order to improve the success rate of candidates.
Factors investigated and data collected
The following table summarises the factors investigated and the order in which the
results are reported.
Table 1: Factors investigated
No. Factor
1 Age
2 Gender
3 Available study time: Marital status
4 Available study time: Children
5 Employment
6 Language: First language
7 Language: Examination language
8 Prior academic achievement: Undergraduate
9 Prior academic achievement: Post-graduate
10 Prior academic achievement: Other advanced degrees
11 Examination attempts
12 Board Course attendance
13 Other factors
In addition to the above, and for completeness purposes, candidates were also asked
about the nature of their employment, whether they were:
employed full-time;
employed part-time;
employed on a casual basis;
retired;
unemployed;
other.
Furthermore, candidates had to indicate whether they were, or had previously been,
employed at an ATO / TOPP training office or at a TRO / TIPP training office. The
QE I paper is set in English and Afrikaans. The survey required candidates to indicate in
which language they attempted the paper.
Results were obtained from the 2010 QE I candidates only with respect to the following
two factors:
Prior academic achievement: Undergraduate performance;
Prior academic achievement: Post-graduate performance.
Candidates were requested to indicate which Board Course they attended (if any), from
the following:
E van Wyk 155
UCT;
UJ;
Other.
Data collection methodology
The email contact details of the target group were obtained from SAICA, and a
questionnaire was sent to all 2009 and 2010 QE 1 candidates. The questionnaire was
anonymous as research indicates that anonymity leads to respondents being more honest
and confident in their responses (Parkin’s Lot, 2005).
A web-based questionnaire was selected as the instrument to collect responses from the
target population. It was a self-administered, structured questionnaire, which reduced
the possibility of interviewer bias and made the research cost-effective, easy to
administer, less intrusive than telephone or face-to-face surveys and convenient, as
respondents could complete the survey in their own time (University of Johannesburg,
2005).
Although there are some disadvantages to using a web-based survey, a response rate of
34% for the 2009 QE I candidates and 29% for the 2010 QE I candidates was obtained.
This high response rate may be due to the fact that many people will respond to an
email invitation to take a web survey on the first day they receive the invitation, and
most will do so within a few days (Survey System, 2010). Respondents also had the
opportunity to contact the researcher in the case of any uncertainties or difficulties in
completing the survey.
Survey quality assurance and limitations
Prior to sending the questionnaire to the target group of respondents, the questionnaire
was approved by a statistician both in terms of content and design. It was then pre-
tested among a small group of former students at Stellenbosch University, who had
written the QE I at least once. The purpose of the pre-test was to ascertain that the
content of the survey was clear and understandable, and that it did not take respondents
too long to complete. Consequently, the test group was required to answer the
questionnaire in its entirety, to record the length of time it took for them to do so and to
indicate whether the questions were concise, clear and unambiguous. The assistance of
a statistician was utilised in extracting and reviewing the preliminary results obtained
from the pre-test group, as well as from the total population of respondents.
The survey had the following limitations, which could have affected respondents’
perceptions regarding the study:
The pre-test group was not a diverse group and mainly consisted of candidates
with Afrikaans or English as their first language;
The pre-test group did not represent the entire target population and its possible
diversity;
The survey was only done in English.
156 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Consequently, the study might not completely be generalised to the entire target
population. However, it might still serve as an inductive investigation to provide some
insight into the characteristics of successful and unsuccessful candidates.
RESULTS
The factors were assigned numbers and statistically evaluated to determine their
significance in relation to the candidates’ QE I results, whether they were “pass” or
“fail”. All the factors, with the exception of age and qualification, were analysed using
chi-square tests. A Mann-Whitney test was performed on the ordinal age variable for
which a normal distribution could be assumed. A Mann-Whitney test was also
conducted on the qualification variable where a normal distribution could not be
assumed.
In some instances certain factors were combined to determine their relevance to one
another, as well as their combined significance in relation to the result of “pass” or
“fail”. An example of this is the factor of “first language” which was reported together
with other factors to provide results for the different “first language” groups such as
English, Afrikaans or Other. The ‘Other’ language group primarily represents the
remaining nine official languages of South Africa.
Of the 21 universities in South Africa, 16 use English as the language of tuition. In the
other five institutions, English-medium tuition is rapidly increasing alongside
Afrikaans-medium tuition. This trend is largely due to the demographic shift in the
student population at all South African educational institutions. Increasing numbers of
black language speakers are enrolling at formerly white and formerly coloured or Indian
institutions (CHE, 2001). This, and the fact that the SAICA QE I papers are set in
English and Afrikaans only, served as primary reason for classifying the language
groups into English, Afrikaans or Other.
The results of the analyses are discussed below.
Age
As illustrated in Figure 1, the mean age of the respondents was 25 years. A significant
difference was found between the average age of the successful candidates and that of
the unsuccessful candidates (p<0.01).
The average age of successful candidates was 24.2 years while unsuccessful candidates’
average age was 26.4 years. This result is in line with the study by Roos (2009), but
contradicts the findings by Brahmasrene and Whitten (2001). In addition, candidates
aged 23.5 years or younger had an 80% success rate, while only 50% of candidates aged
older than 23.5 years passed the examination.
E van Wyk 157
Figure 1: Age
The result showing older candidates to be less successful in the QE I, might indicate the
presence of one of the following underlying factors:
Older candidates are not ‘fresh out of school’ as might be the case of younger
students (Koh & Koh, 1999);
Mature (older) candidates might find it difficult to settle into a study and
examination routine (Gierlaskinski, 2003);
Older candidates might have family commitments, such as having children
(Gierlaskinski, 2003).
Gender
Of the respondents to the study, 49% were male and 51% were female. Results show
that 64% of males passed the exams while 63% of females passed. There was no
significant difference between the gender groups, with p=0.56. This result confirms the
findings of Gammie et al. (2003), Alfan and Othman (2008) and Gierlaskinski (2003).
Available study time
Marital status
Candidates who were married amounted to 15%, while unmarried candidates
constituted 85% of the respondents. The results showed that only 45% of married
candidates passed the exam, while 67% of unmarried candidates passed. This result is
viewed as significant (p<0.01). This result contradicts the findings of Samenfink &
Milliken (1961) discussed earlier. When analysed together with the particular first
language of candidates, Figure 2 shows that the first language groups consisted of
exactly the same combination of married versus unmarried candidates.
Underlying factors that may cause married candidates to be less successful in the QE I,
might be the fact that they have children and/or other family commitments.
Figure 1: Age
Mann-Whitney : p<0.01
Vertical bars denote 0.95 conf idence interv als
Passed Failed
Exam success
23.5
24.0
24.5
25.0
25.5
26.0
26.5
27.0
27.5
Ag
e
158 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Figure 2: Marital status
Children
Figure 3 shows the effect of having children on candidates’ exam success. Of the
candidates, 91% had no children at the time of writing the QE I, while 5% had one child
and 3% of candidates had two children. Candidates with no children had a success rate
of 66% of passing the exam, while candidates with children had only a 37% success
rate, possibly indicating lack of study time, increased family commitments and
responsibilities and difficulty settling into a study routine.
This confirms the findings of Craig and Bittman (2005) that the difference in time
commitment between families with no children and families with different numbers of
children gives a measure of heavy time pressure on the latter (Craig & Bittman, 2005).
In addition Figure 3 shows that candidates with a first language other than Afrikaans or
English were more likely to have had children at the time of writing the QE I.
Figure 3: Children
Figure 2: Marital status
Chi-square test: p=0.93N
um
ber
of
respo
nd
en
ts
first language 3 groups: Afrikaans
15%
85%
Married Unmarried
Marital status
0
100
200
300
400
500
600
700
800
first language 3 groups: English
15%
85%
Married Unmarried
Marital status
first language 3 groups: other
15%
85%
Married Unmarried
Marital status
0
100
200
300
400
500
600
700
800
15%
85%
15%
85%
15%
85%
Af rikaans English
Other
Figure 3: Children
Chi-square test: p=0.00
Nu
mb
er
of
respo
nd
en
ts
first language 3 groups: Afrikaans
97%
3%
no yes
Have children
0
100
200
300
400
500
600
700
800
first language 3 groups: English
93%
7%
no yes
Have children
first language 3 groups: other
77%
23%
no yes
Have children
0
100
200
300
400
500
600
700
800
97%
3%
93%
7%
77%
23%
Af rikaans English
Other
E van Wyk 159
Employment
This study found 81% of the candidates to be employed full-time, of which 75% were
employed as trainees at appropriate training centres, whether RTO / TIPP or ATO /
TOPP employers. Only 6% of the respondents had never been RTO / TIPP or ATO /
TOPP employed.
A difference was noted between the success rates of candidates who were not employed
full-time (78% pass rate) compared to full-time employed candidates (60% pass rate).
This confirms the fact that students who are appointed under a training contract with a
RTO/ATO centre are generally ‘fresh out of school’ (Koh & koh, 1999) and also tend to
be younger candidates, factors which might affect the success rate positively. A slight
difference was reported between the success rate of RTO / TIPP candidates (63% pass
rate) and ATO / TOPP candidates (59% pass rate).
Language
First language
Of the respondents, 48% of candidates regarded English as their first language, 33%
regarded Afrikaans as their first language, and the rest of the respondents were spread
relatively equally among IsiZulu, IsiXhosa, Setswana, Sepedi, SeSotho, Tshivenda,
Xitsonga and IsiNdebele. Candidates with these languages as first language are termed
under the language group ‘Other’.
Figure 4 shows that candidates with English or Afrikaans as their first language had a
success rate in excess of 65%, while candidates with a first language other than English
or Afrikaans had a pass rate of 53%.
Figure 4: First language
This result confirms prior research by Wiese (2006) on the language preferences of, and
language barriers to, black candidates. In addition, the poor scholastic background of
some black candidates could also be a contributing factor in this regard.
Figure 4: First language
Chi-square test: p=0.00
Nu
mb
er
of
respo
nd
en
ts
first language 3 groups: Afrikaans
66%
34%
Passed Failed
Exam success
0
100
200
300
400
500
600
first language 3 groups: English
66%
34%
Passed Failed
Exam success
first language 3 groups: other
53% 47%
Passed Failed
Exam success
0
100
200
300
400
500
600
66%
34%
66%
34%
53% 47%
Af rikaans English
Other
160 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Examination language
As the QE I paper is set only in English and Afrikaans, respondents were asked in
which language they attempted the paper. Seventy-seven percent of the respondents
answered the paper in English and 23% answered the paper in Afrikaans. As shown in
Figure 5, candidates with a first language other than Afrikaans or English were more
likely to fail the exam.
Figure 5: Examination language
Prior academic achievement
Prior academic achievement was only tested for the population of 2010 QE I candidates.
Undergraduate performance
Respondents for the 2010 QE I were asked to state how long it took them to complete
their undergraduate studies in Accounting, normally a three-year degree. Of the
respondents, 60% completed their degrees in the allocated three years, 30% took four
years, 8% took five years and 2% took longer than five years. Undergraduate
performance was evaluated in terms of success in the QE I, and is presented in Figure 6.
Figure 5: Examination language
Chi-square test: p=0.00
Nu
mb
er
of
respo
nd
en
ts
written in home language: no
57%
43%
Passed Failed
Exam success
0
100
200
300
400
500
600
700
800
900
written in home language: y es
66%
34%
Passed Failed
Exam success
57%
43%
66%
34%
Exam not written in f irst language Exam written in f irst language
E van Wyk 161
Figure 6: Undergraduate performance and success in QE I
Candidates who attained their three-year degrees within the allotted time, were more
likely to pass the QE I. Age, together with its possible related factors such as marriage,
children and family commitments, as well as the poor scholastic background of some
black candidates could also be contributing factors in this regard.
The same comparison was done for the 2010 QE I candidates’ post-graduate
performance and success in the 2010 QE I.
Post-graduate performance
Of the respondents who attempted the 2010 QE I, 60% passed their CTA within one
year, 26% within two years, 11% within three years and three percent took longer than
three years to complete their CTA. Figure 7 shows the success rate in the 2010 QE I
taking into account respondents’ post-graduate performance.
It is evident that candidates who take longer than one year to complete there post-
graduate studies, and especially whose who take longer than two years, find the QE I
more challenging. As in the case of undergraduate performance, possible underlying
factors could be candidates’ age, together with its possible related factors such as
marriage, children and family commitments, as well as the poor scholastic background
of some black candidates. As suggested by Parker (2010), black candidates might have
potential negative factors that they might bring to tertiary study. These factors possibly
include an unsupportive home environment, low socio-economic background and
inferior high-school education (Parker, 2010).
The results might also allude to the findings of Phurutse (2005) who found that
predominantly black schools are characterised by poor educational contexts.
Figure 6: Undergraduate perf ormance and succes in QE I
Chi-square test: p=0.00N
um
ber
of
respo
nd
en
ts
No of years for degree: 3 years
63%
37%
Passed Failed
Exam success
0
50
100
150
200
250
300
350
No of years for degree: 4 years
48% 52%
Passed Failed
Exam success
No of years for degree: 5 years
37%63%
Passed Failed
Exam success
0
50
100
150
200
250
300
350
No of years for degree: longer than 5 years
53% 47%
Passed Failed
Exam success
63%
37%
48% 52%
37%63%
53% 47%
Three y ears to complete degree Four y ears to complete degree
Fiv e y ears to complete degree Longer than f iv e y ears to complete degree
162 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Figure 7: Post-graduate performance and success in QE I
Other advanced degrees
Respondents were required to state whether they held other higher qualifications.
Eighty-five percent of the respondents did not hold any other qualification, with 15%
holding other qualifications. The results indicated that holding another higher
qualification did not have an impact on the success rate in the QE I (p=0.98).
Examination attempts
Of the respondents, 61% had attempted the relevant QE I for the first time, while for
18% it was their second attempt. Twenty-one percent had attempted the QE I more than
twice before.
As indicated in Figure 8, first-time candidates have a greater chance of passing the
QE I. In addition, candidates’ likelihood of success decreases as the number of attempts
increase. This concurs with the findings of Gammie et al. (2003), and Roos (2009).
Figure 8: Past QE I attempts
Figure 7: Post-graduate perf ormance and success in QE I
Chi-square test: p=0.00N
um
ber
of
respondents
years to attain CTA: 1 year
68%
32%
Passed Failed
Exam success
050
100150200250300350400
years to attain CTA: 2 years
42%58%
Passed Failed
Exam success
years to attain CTA: 3 years
29%71%
Passed Failed
Exam success
050
100150200250300350400
years to attain CTA: longer than 3 years
43% 57%
Passed Failed
Exam success
68%
32%
42%58%
29%71%
43% 57%
One y ear to attain CTA Two y ears to attain CTA
Three y ears to attain CTA Longer than three y ears to attain CTA
Figure 8: Past QE I attempts
Chi-square test: p=0.00
Nu
mb
er
of
can
did
ate
s
Past QE 1 attempts: 0
75%
25%
Passed Failed
Exam success
0
100
200
300
400
500
600
700
800
Past QE 1 attempts: 1
52% 48%
Passed Failed
Exam success
Past QE 1 attempts: 2
41% 59%
Passed Failed
Exam success
Past QE 1 attempts: 3
38% 62%
Passed Failed
Exam success
0
100
200
300
400
500
600
700
800
Past QE 1 attempts: 4
35%65%
Passed Failed
Exam success
75%
25%
52% 48%
41% 59%
38% 62%35%
65%
First QE I attempt Second QE I attempt Third QE I attempt
Fourth QE I attempt Fif th QE I attempt
E van Wyk 163
The success of candidates of the different first language groups was analysed in relation
to their number of attempts and is presented in Figure 9.
Figure 9: Past QE I attempts for language groups: Afrikaans, English and other
First-time candidates were relatively evenly spread among the three different first
language groups, with the ‘Other’ first language group showing a slightly lower
percentage of first-timers. Again; age, together with its possible related factors such as
marriage, children and family commitments could be contributing factors in this regard.
Board course attendance
Of the respondents in the study, 95% attended an appropriate Board Course prior to
writing the QE I. Of the 95% who attended Board Courses 57% attended the Gauteng
Board Course and 38% attended the Cape Town Board Course. The remainder of the
Board Course attendees attended ‘Other’ Board Courses.
As is evident in Figure 10, the attendance of a Board Course had a significant impact on
the success rate of candidates. Board Course attendees had a success rate of 65%, versus
the success rate of 35% of those candidates who did not attend a Board Course
(p=0.00). This is in line with the findings of Roos (2009) as well as with studies on the
US CPA examinations (Journal of Accountancy, 1983).
Figure 9: Past QE I attempts f or language groups: Af rikaans, English and Other
Chi-square test: p=0.00
Nu
mb
er
of
respo
nd
en
ts
first language 3 groups: Afrikaans
60%
17%9% 8% 5%
0 1 2 3 4
Past QE 1 attempts
0
100
200
300
400
500
600
first language 3 groups: English
64%
19%
7% 4% 5%
0 1 2 3 4
Past QE 1 attempts
first language 3 groups: other
54%
15% 12% 7% 11%
0 1 2 3 4
Past QE 1 attempts
0
100
200
300
400
500
600
60%
17%9% 8% 5%
64%
19%
7% 4% 5%
54%
15% 12% 7% 11%
Af rikaans English
Other
164 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Figure 10: Board course attendance
Furthermore, as displayed in Figure 11, it was found that the success rates of the Board
Courses differed (p<0.01). Of the respondents who attended Board Courses, the Cape
Town Board Course was found to deliver the highest success rate, with 83% of the
attendees having passed the exam. The Gauteng Board Course delivered a success rate
of 64%, while ‘Other’ Board Courses delivered a success rate of 37%. It is generally
accepted that candidates who contracted to complete their training periods with certain
of the larger RTO/ATO training centres, are often encouraged and financially assisted
to attend the UCT and UJ Board Courses.
Figure 11: Success rates of different board course candidates
Figure 10: Board Course attendance
Chi-square test: p=0.00N
um
ber
of
respo
nd
en
ts
Board Course: Yes
65%
35%
Passed Failed
Exam success
0
200
400
600
800
1000
1200
Board Course: No
35%65%
Passed Failed
Exam success
65%
35%
35%65%
Attended a Board Course Did not attend a Board Course
Figure 11: Success rates of dif errent Board Course candidates
Chi-square test: p=0.00
Nu
mb
er
of
can
did
ate
s
Which Board Course?: UCT (Cape Town)
77%
23%
Passed Failed
Exam success
0
100
200
300
400
500
600
Which Board Course?: UJ (Gauteng)
58%
42%
Passed Failed
Exam success
Which Board Course?: Other
42% 58%
Passed Failed
Exam success
0
100
200
300
400
500
600
77%
23%
58%
42%
42% 58%
Cape Town (UCT) Board Course Gauteng (UJ) Board Course
Other Board Course
E van Wyk 165
Board Course attendance for the different first language groups is displayed in Figure
12.
Figure 12: Board course attendance for language groups: Afrikaans, English and
other
Other factors
Candidates who failed the QE I were required to state the three most important factors
which may have been responsible for their failing the exam. Time management,
insufficient exam preparation, poor performance under exam conditions and academic
workload were listed as the most important factors. Other factors, such as personal or
family crises, stress, struggling to balance study and social commitments, and financial
problems were also listed as important factors, although to a lesser extent. Figure 13
displays the scores attained by the different factors over the total population surveyed.
Figure 13: Factors causing failure for total population
Similarly, the four most important factors indicated by unsuccessful candidates were
also cited by candidates with first languages other than Afrikaans or English. Figure 14
shows the factors cited by this group.
Figure 12: Board Course attendance f or language groups Af rikaans, English and Other
Chi-square test: p<0.01
Nu
mbe
r of
ca
nd
idate
s
first language 3 groups: Afrikaans
33%
61%
6%
UCT (Cape Town) UJ (Gauteng) Other
Which Board Course?
020406080
100120140160180200220240
first language 3 groups: English
44%
53%
3%
UCT (Cape Town) UJ (Gauteng) Other
Which Board Course?
first language 3 groups: other
31%
62%
7%
UCT (Cape Town) UJ (Gauteng) Other
Which Board Course?
020406080
100120140160180200220240
33%
61%
6%
44%
53%
3%
31%
62%
7%
Af rikaans English
Other
Figure 13: Factors causing f ailure f or total population
Count of y es
54% 53%48% 46%
17% 14%11% 10%
Tim
e m
anagem
ent
Exam
pre
para
tion
Perf
orm
under
exam
conditio
ns
Work
load
Pers
onal or
fam
ily c
rises
Str
ess
Bala
nce s
tudy a
nd s
ocia
l
Fin
ancia
l pro
ble
ms
Signif icant f actors of f ailure
0
20
40
60
80
100
120
140
160
Count
of
yes
54% 53%48% 46%
17% 14%11% 10%
Unsuccessf ul candidates' signif icant f actors of f ailure
166 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Figure 14: Significant factors of failure for ‘other’ language group
Finally, Figure 15 displays the different factors mentioned by male and female
candidates which had the greatest effect on their failing the examination. The four
factors mentioned earlier were equally important to males and females, although slight
differences were found between the gender groups.
Figure 15: Significant factors of failure: Male and female
Confirming the findings of Macan et al. (1990), females tended to be better at time
management, although there was no significant difference between female and male
candidates. In relation to stress levels and perceived control over time, the findings of
Macan et al. (1990) were also confirmed, with males faring better than females.
CONCLUSIONS AND RECOMMENDATIONS
The success of candidates in the examinations set by SAICA has become a topical issue
due to a shortage of chartered accountants (CA(SA)s) experienced in South Africa
(SAICA, 2010a).
The aim of the study was:
Figure 14: Signif icant f actors of f ailure f or 'Other' language group
Count of y es
55%52%
42%38%
17%15%
12%8%
Ex
am
pre
para
tio
n
Tim
e m
an
ag
em
en
t
Pe
rfo
rm u
nd
er
exa
m c
on
ditio
ns
Wo
rklo
ad
Pe
rson
al
or
fam
ily c
rise
s
Ba
lan
ce
stu
dy
and
so
cia
l
Fin
anc
ial
pro
ble
ms
Str
ess
Signif icant f actors of f ailure
0
10
20
30
40
50
60
70
80
9055%
52%
42%38%
17%15%
12%8%
Unsuccessf ul candidates' signif icant f actors of f ailure
Figure 15: Signif icant f actors of f ailure: Male and Female
Male
Female
58%
36% 34%
58%
22% 21%
6%
16%
52%
40%
48% 46%
8%
14%
10% 8%
Exam
pre
para
tio
n
Wo
rklo
ad
Pe
rfo
rm u
nd
er
exa
m c
on
ditio
ns
Tim
e m
an
ag
em
en
t
Ba
lanc
e s
tud
y a
nd
so
cia
l
Pe
rson
al or
fam
ily c
rise
s
Str
ess
Fin
anc
ial p
roble
ms
Signif icant f actors of f ailure
0
5
10
15
20
25
30
35
40
45
50
58%
36% 34%
58%
22% 21%
6%
16%
52%
40%
48% 46%
8%
14%
10% 8%
Unsuccessf ul candidates' signif icant f actors of f ailure
E van Wyk 167
to identify underlying factors, by investigating the relevant literature, that may
affect the success of candidates in accounting examinations in general;
to collect and statistically analyse data, regarding these underlying factors from
candidates in 2009 QE I and 2010 QE I;
to ascertain which factors characterise candidates who were successful and
unsuccessful in the 2009 QE I and 2010 QE I examinations;
to construct a set of characteristics, and the possible underlying factors which
cause them; and
to make recommendations to candidates, educators, training centres and to SAICA
to serve as basis for future studies as to how the success rate of candidates in the
QE I might be improved.
Conclusions
Factors identified from literature
A number of factors that may affect accounting examination success were identified:
Age;
Gender;
Available study time: Marital status;
Available study time: Children;
Employment;
Language: First language;
Language: Examination language;
Prior academic achievement: Undergraduate;
Prior academic achievement: Post-graduate;
Prior academic achievement: Other advanced degrees;
Examination attempts; and
Board Course attendance.
Characteristics of successful and unsuccessful candidates
Acknowledging the limitations of the study, it was found that successful and
unsuccessful candidates in the 2009 and 2010 QE I papers had one or more of the
following characteristics:
168 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Table 2: Characteristics of successful and unsuccessful candidates
Successful candidates Unsuccessful candidates Statistically
significant with
p<0.01 (Yes/No)
Generally younger than 23.5
years
Generally older than 23.5 years Yes
Female Male No
Unmarried Married Yes
Childless Have children Yes
Employed Unemployed No
Afrikaans or English is their
first language
Afrikaans or English is not their first
language
Yes
Had the opportunity and
attempted the QE I in their
first language
Did not have the opportunity to attempt
the QE I in their first language
Yes
Good prior academic
achievement
Poor prior academic achievement Yes
Passed the QE I at their first
attempt
Attempted the QE I more than once Yes
Attended a Board Course Did not attend a Board Course Yes
Unsuccessful candidates also indicated the following factors to have contributed to their
failure:
poor time management both prior to and during the QE I;
inadequate exam preparation;
poor performance under examination conditions; and
the high academic workload associated with the QE I.
The same factors were prioritised by male, female and black candidates.
As is evident from the results section of the paper, certain factors may affect/cause more
than one characteristic.
Age, marriage, children and the number of examination attempts
Factors linked to the age of unsuccessful candidates might be that they are not ‘fresh out
of school’ as stated by Koh and Koh (1999) when attempting the QE I. This might be
due to ineffective study skills or financial constraints causing them to take longer to
complete their studies, or causing them to complete their studies on a part-time basis.
Older candidates might also more likely be married and might have children at the time
of attempting the QE I. In addition, these candidates might also have attempted the QE I
more than once as a result of one/more of the factors discussed in this section.
First language, examination language and prior academic achievement
Most black candidates were found to have attempted the QE I in a language other than
their first language. Also, these candidates were most likely to have received their
scholastic and tertiary education in a language other than their first language.
E van Wyk 169
Black candidates’ academic performance at tertiary level might also be affected by the
quality of the institution(s) they attended.
In addition, black candidates are more likely to come from an inferior high-school
background in which the resources of schools are limited and in which class size is a
problem.
Board course attendance
Candidates who attended the UCT and Gauteng Board Courses were most likely
successful in the QE I. When one considers that most candidates who attend these
Board Courses are trainees contracted with larger RTO/ATO training centres,
accessibility and financial constraints might prevent candidates who are not contracted
with the larger RTO/ATO training centres to attend these Board Courses.
Recommendations and opportunities for further research
Candidates
From the results and conclusions of the study it is clear that candidates should be made
aware of the characteristics of successful candidates:
they tend to attempt the QE I at a young age;
they attempt the QE I prior to getting married and starting a family; and
they have good academic achievement prior to attempting the QE I.
Candidates should be made aware of the following in respect of the characteristics
found for unsuccessful candidates:
the chances of passing the QE I decrease as the number of exam attempts increase;
in the continued absence of examination papers in languages other than English or
Afrikaans, candidates with a first language other than these should be encouraged
to work hard at mastering Afrikaans or English at the level of writing the QE I,
perhaps enrolling for a subject such as Language Development 114/144 (available
at Stellenbosch University as a first-year subject for B Commerce and
B Accounting students) early in their tertiary studies (Stellenbosch University,
2011); and
attending a Board Course prior to writing the examination might improve their
success rate.
Considering the significant factors listed by candidates who failed the exam, candidates
should be made aware of the benefits of mastering the following:
to adequately utilise their time in preparing for and writing the exam;
to adequately prepare for the exam; and
to deal with stress under exam conditions.
170 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Training centres and SAICA
SAICA, in collaboration with relevant RTO/ATO training centres, may consider
providing or obtaining financial assistance to/for candidates who may not normally be
able to attend Board Courses.
The effect on black candidates of having to study and attempt the QE I in a language
other than their first language should be investigated and, where possible, addressed.
One way of addressing the issue may be that SAICA may consider making the QE I
papers available in the two main black languages isiXhosa and isiZulu.
Educators
In the continued absence of the QE I examination papers in languages other than
English or Afrikaans, candidates with a first language other than these should be
encouraged to enroll for a subject such as Language Development 114/144 (available at
Stellenbosch University as a first-year subject for B Commerce and B Accounting
students) early in their tertiary studies (Stellenbosch University, 2011).
In addition, finding that successful candidates generally have good prior academic
achievements, educators should encourage candidates to work hard at their studies from
early on.
Finally, factors affecting the academic performance of candidates in general deserve
more research. These include their scholastic background including the resource base of
schools, candidates’ home environment, socio-economic background, possible inferior
high-school education and a ‘host of different factors affecting the education of
candidates’ (Phurutse, 2005).
ACKNOWLEDGEMENTS
My sincere appreciation to Professor Martin Kidd who assisted with the statistical
analysis of the survey results.
REFERENCES
Alfan, E. and Othman, N. (2005). Undergraduate students’ performance: the case of
University of Malaya, Quality Assurance in Education, 13(4):329-343.
Anderson, E.D. and Cole, B.S. (2001). Stress factors related to reported academic
performance and burnout, Education, 108(4):497-503.
Auyeung, P. and Sands, J. (1997). Factors influencing accounting students’ career
choice: a cross-cultural validation study, Accounting Education: An International
Journal, 6(1):13-23.
Black, H.T. and Duhon, D.L. (2003). Evaluating and improving student achievement in
business programs: the effective use of standardized assessment tests, Journal of
Education for Business, 79(2):90-98.
E van Wyk 171
Boone, J., Legoria, J., Seifert, D.L. and Stammerjohan, W.W. (2006). The associations
among accounting program attributes, 150-hour status, and CPA-exam pass rates,
Journal of Accounting Education, 24(4):202-215.
Brahmasrene, T. and Whitten, D. (2001). Assessing success on the uniform CPA exam:
a logit approach, Journal of Education for Business, 77(1):45-50.
Boud, D. (1988). Developing student autonomy in learning. 2nd
edition. London: Kogan
Page.
Byrne, M. and Flood, B. (2008). Examining the relationships among background
variables and academic performance of first year accounting students at an Irish
University, Journal of Accounting Education, 26:202-212.
Chan, S.Y. and Ho, S.M. (2000). Desired attributes of public accounting firms in the job
selection process: an empirical examination of accounting graduates’ perceptions,
Accounting Education: An International Journal, 9(4):315-327.
Council on Higher Education (CHE). 2001. Language policy framework for South
African Higher Education. http//:www.info.gov.za/otherdocs/2001/langframe.pdf.
Accessed 1 April 2011.
CHE, vide Council on Higher Education.
Craig, L. and Bittman, M. 2005. The effect of children on adults’ time use: analysis of
the incremental time costs of children in Australia. SPRC Discussion Paper No.43.
www.sprc.unsw.edu.au/media/file/DP143.pdf. Accessed 7 April 2011.
De Lange, P., Waldmann, E. and Wyatt, K. (1997). Personal characteristics and
academic achievement of undergraduate accounting students studying through open
learning, Accounting Education, 6(4):295-306.
Du Plessis, A., Müller, H. and Prinsloo, P. (2005). Determining the profile of the
successful first-year accounting student, South African Journal of Higher Education,
19(4):684-698.
Ernst and Young South Africa. 2011. Board Courses. www.ey.com/ZA/en/Careers/
Board-courses. Accessed 7 April 2011.
Felton, S., Burh, N. & Northey, M. (1994). Factors influencing the business student’s
choice of a career in chartered accountancy, Issues in Accounting Education, 9(1): 131–
141.
Gamaroff, R. 1995. Solutions to academic failure: the cognitive and cultural realities of
English as the medium of instruction among black learners.
www.grammargraph.wordpress.com. Accessed 7 April 2011.
Gammie, E., Paver, B., Gammie, B. & Duncan, F. (2003). Gender differences in
accounting education: an undergraduate exploration, Accounting Education, 12(2):177-
196.
172 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Gierlaskinski, N. (2003). Analysis of factors leading to success in intermediate
accounting, Journal of Academy of Business and Economics. Available online:
www.allbusiness.com/human-resources/employee-development/752311-1.html.
Accessed 11 May 2010.
Gracia, L. and Jenkins, E. (2002). An exploration of student failure on an undergraduate
accounting programme of study, Accounting Education, 11(1):93-107.
Inman, B.C., Wenzler, A. & Wickert, P.D. (1989). Square pegs in round holes: are
accounting students well suited to today’s accounting profession? Issues in Accounting
Education, 4(1):29-47.
Innocenti, N.D. (2009). Accountants like gold dust in South Africa. Available online:
http://news. efinancialcareers.co.za/News_ITEM/newsItemId-19835. Accessed 11
October 2010.
Jackling, B. and Anderson, A. (1998). Study model, general ability and performance in
accounting: a research note, Accounting Education, 7(1):65-73.
Journal of Accountancy. (1983). CPA exam takers show rise in level of education.
January:10-12.
Koh, M.Y. and Koh, H.C. (1999). The determinants of performance in an accountancy
degree programme, Accounting Education, 8(1):13-29.
Macan, T.H., Shahani, C., Diphoye, R.L. and Phillips, A.P. (1990). College students’
time management: correlations with academic performance and stress, Journal of
Educational Pshychology, 87(4):760-768.
Mail & Guardian Online. (2008). SA faces shortage of qualified accountants. Available
online:http://www.mg.co.za/article/2008-11-05-sa-faces-shortage-of-qualified-accoun-
tants. Accessed 11 October 2010.
Mauldin, S., Crain, J.L. and Mounce, P.H. (2000). The accounting principles
instructors’ influence on students’ decision to major in accounting, Journal of
Education for Business, 142 -148.
Moses. O.D. (1987). Factors explaining performance in graduate-level accounting,
Issues in Accounting Education, 2(2):281-291.
Mutchler. J.F., Turner, J.H. and Williams, D.D. (1987). The performance of female
versus male accounting students, Issues in Accounting Education, 2(1):103-111.
Myburgh, J.E. (2005). An empirical analysis of career choice factors influencing first-
year Accounting students at UP, Meditari Accountancy Research, 13(2):35-48.
Parker, K.M. 2010. Does attendance at a historically white university benefit non-white
students of introductory economics in South Africa? South African Journal of
Economics, 78(2):209-218.
E van Wyk 173
Parkin’s Lot. (2005). Best practices in questionnaire design. Available online:
http://parkinslot. blogspot.com/2005/09/best-practices-in-questionnaire-design.html.
Accessed 12 May 2010.
Phurutse, M.C. (2005). Factors affecting teaching and learning in South African public
schools. Report presented to the Education Labour Relations Council.
www.hsrcpress.ac.za. Accessed 7 April 2011.
PWC. 2011. Campus candidates: study support. www.pwc.com/za/en/campus-
candidates/study-support.jhtml. Accesses 7 April 2011.
Roos, S. (2009). Factors affecting Southern African students’ success in CIMA
examinations, Meditari Accountancy Research, 17(1):49-67.
Sadler, E. & Erasmus, B.J. (2005). The academic success and failure of black chartered
accounting graduates in South Africa: a distance education perspective, Meditari
Accountancy Research, 13(1):29-50.
Samenfink, A. & Milliken, R.L. (1961). Marital status and academic success, Marriage
and family living, 23(3):226-227.
South African Institute of Chartered Accountants (SAICA). (2010a). About SAICA.
Available online: https://www.saica.co.za/About/tabid/56/language/en-ZA/Default.
aspx. Accessed 18 February 2010.
South African Institute of Chartered Accountants (SAICA). (2010b). Becoming a CA.
Available online: https://www.saica.co.za/Trainees/BecomingaCA/tabid/157/language/
en-ZA/Default.aspx. Accessed 8 April 2010.
South African Institute of Chartered Accountants (SAICA). (2010c). Examinations.
Available online:https://www.saica.co.za/LearnersStudents/Examinations/tabid/469/
language/en-US/Default.aspx. Accessed 18 February 2010.
South African Institute of Chartered Accountants (SAICA). (2010d). QE I 2011 – New
examination format. Available online: https://www.saica.co.za/LearnersStudents
/QEI2011Newexaminationformat/tabid/1776/language/en-ZA/Default.aspx. Accessed
18 February 2010.
SAICA, vide The South African Institute of Chartered Accountants.
Stellenbosch University. (2011). Jaarboek 2011 Deel 10: Fakulteit Ekonomiese en
Bestuurswetenskappe.
Survey System. (2010). Survey Design – how to begin your survey design project.
Available online: www.surveysystem.com/sdesign.htm. Accessed 12 May 2010.
Tan, L.M. and Laswad, F. (2006). Student’s beliefs, attitudes and intentions to major in
accounting, Accounting Education: An International Journal, 15(2):167-187.
Titard, P.L. and Russell, K.A. (1989). Factors affecting CPA examination success,
Accounting Horizons, 3(3):53-59.
174 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Toh, L.M. (1994). Some evidence on the determinants of student performance in the
University of Malaya introductory accounting course, Accounting Education, 3(4):331-
340.
Tyson, T. (1989). Grade performance in introductory accounting courses: why female
students outperform males, Issues in Accounting Education, 4(1):153-160.
University of Johannesburg. (2005). Questionnaire design. Analysing survey data using
SPSS13: a workbook.
Wiese, A. (2006). Transforming SA’s chartered accountancy profession: the progress
and obstacles black trainees still encounter, Meditari Accountancy Research, 14(2):151-
167.
Wikipedia. 2011. Gender identity. http://en.wikipedia.org/wiki/Gender_identity.
Accessed 31 March 2011.
G Steenkamp 175
A Note: The perceptions of a CTA class
regarding the changes to the international
accounting standards dealing with financial
instruments
G Steenkamp
University of Stellenbosch
Received: January 2010 SAJAR
Revised: January 2011, May 2011 Vol 25 No. 1
Accepted: May 2011 2011
pp.175 to 189
The IASB has recently begun redeveloping the accounting standards that deal with financial
instruments. During 2009 the IASB issued an exposure draft (ED 7/2009), which proposed
certain changes to IAS 39. This study seeks to investigate the perceptions of a CTA class
regarding these proposed changes. The study also compares the students’ responses to the South
African Institute of Chartered Accountants (SAICA) comments as well as the final IFRS 9. In
general, it was found that the students were positive about the proposed changes, and formulated
valid opinions regarding aspects thereof - although their practical examples were limited. It was
found that the perceptions of the students mostly corresponded to the SAICA comments, and
even more so to the final IFRS 9 drafted. It is conjectured that including students in the
evaluation of an exposure draft could be beneficial to both the process and the students
themselves, as it could develop the students’ pervasive professional skills (such as critical
thinking, life-long learning and change management) as required by the new Competency
Framework: Detailed Guidance for Academic Programmes.
KEY WORDS
Financial instruments; IAS 39; ED 7/2009; IFRS 9; Graduate skills; Competency framework;
Critical thinking
Contact
INTRODUCTION
Recently the International Accounting Standards Board (IASB) has begun to reshape
the accounting standards that deal with financial instruments, in response to
international pressure to improve and simplify these standards. In the current stage of
the redevelopment, IAS 39 Financial Instruments: Recognition and Measurement is
being rewritten in phases. The first of these phases addressed the classification and
measurement of financial assets and financial liabilities. An exposure draft in this
regard, Exposure Draft 7 of 2009 (ED 7/2009), was issued on 14 July 2009 and
comments were due by 14 September 2009. The South African Institute of Chartered
176 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Accountants (SAICA) wrote a comment letter and submitted this to the IASB by the
due date.
The IASB has subsequently drafted the first portion of IFRS 9 Financial Instruments,
refining the original exposure draft based on the comments that were received. IFRS 9
was issued in November 2009 and could voluntarily be applied by companies with a
December 2009 year-end. Originally, the mandatory effective date was set as
1 January 2013, but a later date (1 January 2015) is currently being considered. The
next phases of this project included exposure drafts on impairment, hedge accounting
and derecognition. Each phase was addressed separately during 2010 and 2011, and
certain deliberations are still continuing.
Students completing a Certificate of Theory in Accounting (CTA) study the current
IAS 39 as part of the syllabus for Part 1 of the SAICA Qualifying Examination (QE1).
Increasingly, South African universities offering a CTA are seeking to prepare
graduates more adequately for the practical business environment, rather than simply
teaching them theoretical and technical knowledge. One of the major reasons for this
move away from pure theory towards practice, has been an international debate
regarding the necessary skills that an accounting graduate (and especially a chartered
accountant) should possess. In South Africa, SAICA has also acknowledged the need
for a broader skill set, and have developed a new Competency Framework for the
education and training of prospective CAs (SA). A specific document (called the
“Competency Framework: Detailed Guidance for Academic Programmes”) that
contains guidance on the implementation of the Competency Framework at university
level was also developed (SAICA, 2009a).
A group of students, studying full time towards a CTA at a residential South African
University, was surveyed regarding their opinion on the proposed amendments in
ED 7/2009. The goal was to test whether they had the necessary skills to critically
assess the appropriateness of proposed new accounting standards. They were
specifically asked to answer the questions posed by the IASB in the exposure draft
itself. The views of the students were compared to those contained in the comment
letter submitted by SAICA to the IASB, as well as the final IFRS 9. This was done to
determine whether the students’ views corresponded to these documents and whether
the students could apply their theoretical knowledge to formulate valid opinions
regarding the proposed changes, in line with the requirements of the new Competency
Framework: Detailed Guidance for Academic Programmes mentioned above.
RESEARCH OBJECTIVE AND CONTRIBUTION
Research objective
This research aims to determine the perceptions of a CTA class regarding the changes in
accounting for financial instruments as proposed by ED 7/2009. This is done by posing
the questions asked by the IASB in the exposure draft to the students. The objective of
this is to evaluate how successfully these CTA students can apply their theoretical
knowledge to a practical problem and whether their education has taught them to
critically evaluate new accounting standards, especially in the light of the requirements
of the Competency Framework: Detailed Guidance for Academic Programmes.
G Steenkamp 177
Another aim is to ascertain whether the perceptions and comments of students
correspond with those of other chartered accountants. This will be done by firstly
comparing the students’ comments to those of SAICA (as per comment letter submitted
to the IASB) and the IASB (as per final IFRS 9).
Research contribution
One of the contributions of the research would be to assess the adequacy of the current
educational programme for chartered accountants in South Africa, regarding the
teaching of specific professional skills (such as critical thinking, problem solving and
change management). This is especially relevant in the light of the new Competency
Framework: Detailed Guidance for Academic Programmes, as certain of these skills
must now be developed in a student as part of the academic training programme, as
prescribed by SAICA. This could help universities form opinions on whether they
currently meet those requirements or whether some changes are necessary.
Another possible contribution is to raise the question whether the inputs of students
(and especially post-graduate students) should be solicited during the process of
drafting a comment letter to the IASB or other such institutions. Post-graduate students
have a vast technical knowledge and their input could, as such, be valuable. This
research also adds another layer to the Financial Instruments discussion, as the
viewpoints of students regarding the recent changes have not been researched.
LITERATURE REVIEW
Although every company employs financial instruments, the use and variety of these
financial instruments have expanded dramatically over the past decade. However, the
current accounting standards relating to financial instruments are believed to be
complex and difficult to understand and apply (Lopes & Rodrigues, 2008). Some have
felt that these standards are problematic because they are rules-based, with many scope
exclusions and exceptions, which allows for financial structuring to achieve a pre-
determined financial goal (Schipper, 2003). It has even been argued that the recent
economic crisis could have been avoided, at least in part, if the risks associated with
highly specialised financial instruments had been disclosed thoroughly in the financial
statements of companies (Arnold, 2009).
The development of a simplified standard detailing the accounting rules for financial
instruments has been on the long-term agenda of the IASB since 2005, but the global
financial crisis in the latter half of 2008 prompted a faster response. During 2009 the
IASB commenced with a phased improvement plan to rewrite IAS 39. The first of these
phases addressed the classification and measurement of financial assets and liabilities
through the issue of ED 7/2009. A summary of the proposed amendments to IAS 39 by
ED 7/2009 is presented in the table below:
178 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Table 1: Changes proposed by ED 7/2009
CURRENT IAS 39
PROPOSED AMENDMENTS BY
ED 7/2009
1 CLASSIFICATION APPROACH
Four categories of financial assets (namely 'at fair
value through profit or loss', 'held to maturity',
'loans and receivables' and 'available for sale') and two categories of financial liabilities ('at fair value
through profit or loss' and 'other').
Two primary categories of financial assets
and liabilities, namely those measured at
amortised cost and those measured at fair value. A financial asset or financial liability
would be measured at amortised cost if it has
basic loan features and is managed on a contractual yield basis. Any other financial
instrument would be measured at fair value.
2 EMBEDDED DERIVATIVES
An embedded derivative should be split from its
host contract (however, there are a few exceptions to this rule).
A hybrid contract with a host that is within
the scope of the proposed IFRS (i.e. the host contract is a financial instrument) is classified
in its entirety in accordance with the proposed
classification approach. It will not be split.
3 FAIR VALUE OPTION
An entity may designate any financial asset or liability as 'at fair value through profit or loss' even
though it does not meet the requirements (if it
eliminates or significantly reduces a measurement or recognition inconsistency).
An entity may still elect to carry any financial asset or financial liability at fair value,
although it has basic loan features and is
managed on a contractual yield basis (if it eliminates or significantly reduces a
measurement or recognition inconsistency).
4 RECLASSIFICATION
Reclassifications are permitted between certain
categories of financial assets and liabilities when
certain criteria are met, e.g. from 'Held to maturity' to 'Available for sale' when the entity no longer
intends to hold it until maturity.
No reclassifications are allowed.
5 INVESTMENTS IN EQUITY INSTRUMENTS THAT DO NOT HAVE A QUOTED
MARKET PRICE AND WHOSE FAIR VALUE CANNOT BE RELIABLY MEASURED
Such investments should be carried at fair value, but if the fair value cannot be determined, they may
be carried at cost as an exception. If carried at cost,
they should be tested for impairment.
Such investments should still be carried at fair value, even though this would lead to
additional costs. This approach would provide
better information, simplify requirements and improve comparability.
6 INVESTMENTS IN EQUITY INSTRUMENTS THAT ARE MEASURED AT FAIR VALUE
THROUGH OTHER COMPREHENSIVE INCOME (OCI)
Financial assets classified as 'available for sale'
should be carried at fair value, with changes in fair
value through OCI. An asset is available for sale if it does not meet the requirements of any of the
other categories of financial assets or is designated
as such. Dividends on such investments are shown
in profit or loss.
If investments in equity instruments are not
held for trading, the changes in fair value can
be presented in OCI. This choice must, however, be an irrevocable election at initial
recognition. Dividends on such investments
should also be shown in OCI.
(IAS 39 and ED 7/2009)
G Steenkamp 179
During 2009, SAICA reviewed its academic and training programme for prospective
chartered accountants by developing the Competency Framework: Detailed Guidance
for Academic Programmes (SAICA, 2009a). Although these two matters are unrelated,
the development of the Competency Framework and the changes to the financial
instrument standard coincided. Some of the skills that have received increased attention
in the Competency Framework: Detailed Guidance for Academic Programmes are the
personal attributes and professional skills which a CA(SA) should possess upon
qualification (SAICA, 2009a). These include awareness of changes in the business
environment, analytical/critical thinking, problem-solving skills, innovation, change
management as well as the ability to critically assess an unknown situation using
current knowledge and form an appropriate opinion on the matter (SAICA, 2009a).
From 2011 (on third year level) and 2012 (on CTA level), the academic programme will
have to develop these skills in students, to the extent possible in an academic setting.
This will have certain implications for the way in which accounting is taught, as
students would have to be able to critically analyse current financial standards and point
out flaws (specific competency referred to in Competency Framework is “IC-2
Examines and interprets information and ideas critically”), as well as be aware of
upcoming changes and its effect (specific competency referred to in Competency
Framework is “IB-5 Manages change” and “IB-8 Is a life-long learner”) (SAICA,
2009a). Research conducted by Nettleton, Litchfield & Taylor (2008) and Gammie,
Gammie & Cargill (2002) pointed out that a graduate should possess the following
skills:
The ability to critically assess theory learnt, to determine if it is applicable in a
certain scenario, and adapt and apply this theory in a dissimilar scenario,
The skill to develop the appropriate solution for a specific problem,
A questioning mind, which challenges the current way of doing things,
The ability to analyse situations and critically assess the problem and the possible
solutions, and
The desire to keep learning and improving (lifelong learning).
Lifelong learning is the ability to appropriately and effectively adapt to changes in a
specific work environment, as well as the ability to apply theoretical knowledge in new
and practical situations (Tempone & Martin, 2000). The importance of lifelong learning
was affirmed by Wessels & Steenkamp (2009), who mention that many professional
accountancy bodies require their members to be able to manage change and solve
problems by critical and strategic thinking.
RESEARCH METHODOLOGY
Overall research design and method
Since the objective of this research was to investigate the perceptions of a CTA class
regarding the proposed changes to the standard on financial instruments, an empirical
180 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
research methodology was decided upon, with a literature review to guide the empirical
research, explain the research question adequately and interpret the responses gathered.
Survey instrument (including development)
A survey instrument was developed by, firstly, comparing the current accounting
treatment in IAS 39 to the proposed treatment in ED 7/2009, and tabulating the major
differences (as shown in Table 1). Table 1 only includes the differences which relate to
those aspects of IAS 39 that form part of the QE1 syllabus, as these are the aspects
which the students are familiar with and could be expected to have an opinion on.
Secondly, the questions asked by the IASB in the ‘Request for Comments’ section of
ED 7/2009 pertaining to the abovementioned changes were tabulated and modified
slightly to enable the students to better understand each question. In many cases the
questions were shortened slightly.
From the information above, a survey was developed that listed and explained the
changes (as indicated in Table 1) and then asked the students to indicate whether they
thought the change was Positive (“P”) or Negative (“N”). The results of this part of the
survey are presented in Table 3. The survey further posed one or two questions (adapted
from the questions in the IASB’s ‘Request for Comments’) for each change in Table 1,
to which the students were required to answer with a Yes (“Y) or a No (“N”). Provision
was also made for a narrative comment or response alongside each question. The
adapted IASB questions posed to the students (to which they could answer Yes or No)
can be found in Table 4, together with the results of the survey pertaining to these
questions.
Data collection
This survey was handed out in class to the 205 students enrolled for the CTA
programme at a residential university in 2009. A total of 51 students completed the
survey on paper.
Data analysis
The data from the completed surveys was transferred to an electronic spreadsheet,
tabling the responses from each student to each change (positive/negative) and each
question (yes/no). The percentage of the students who held each view (positive/negative
or yes/no) concerning each change or question was calculated. The data was also
inputted in a statistical computer programme (STATISTICA) for statistical analysis.
The narrative responses to each question were summarised and will be discussed under
the specific heading or question to which they relate.
EMPIRICAL RESEARCH FINDINGS
Biographical details of the sample in comparison to the population (academic strength)
The biographical details of the sample (the 51 students who responded) were compared
to the population (entire CTA class, i.e. 205 students) in terms of their academic
strength to ascertain whether the responses were properly balanced between stronger
G Steenkamp 181
and weaker students. If this was not the case, it might have led to a bias in the sample.
Biographical data was only available regarding 49 of the 51 students who responded.
The academic strength of the respondents was based on final marks (average for all
subjects) at the end of the academic year.
Table 2: Academic strength of respondents (based on final marks)
Sample Population
Mean 52.1 51.3
Standard deviation 9.6 8.0
The mean final mark and the standard deviation therein for both the population and the
sample are detailed in Table 2 above. Although the sample’s mean and standard
deviation are slightly higher than the population’s, the differences are small enough to
suggest that the sample is not biased towards stronger or weaker students. Any
significant differences (in the responses to the questions asked) between students of
differing academic strength will be discussed in the appropriate sections below (refer
tables 5 and 7).
Summary of the findings
The results of the empirical research are listed below. Note that the responses do not
always add to 100% as not all the respondents answered all the questions.
Table 3: Percentage who were positive or negative about each change
Proposed changes in ED 7/2009 (refer to Table 1 for
details of changes) Positive Negative
1 Classification approach 73% 18%
2 Embedded derivatives 76% 14%
3 Fair value option 67% 24%
4 Reclassifications 35% 51%
5
Investments in equity instruments that do not have a quoted
market price and whose fair value cannot be reliably
measured 65% 22%
6 Investments in equity instruments that are measured at fair
value through other comprehensive income (OCI) 65% 16%
182 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Table 4: Percentage of students who answered ‘Yes’ or ‘No’ to each question
Number of
question by
IASB
IASB's question (slightly modified) Percentage of
students who
answered 'Yes'
Percentage of
students who
answered 'No'
CLASSIFICATION APPROACH
Q1 (a) Does amortised cost provide decision-useful information for a financial asset or financial
liability that has basic loan features and is
managed on a contractual yield basis?
94% 2%
Q3 (b) Do you believe that other conditions would be
more appropriate to identify which financial
assets or financial liabilities should be measured at amortised cost?
12% 75%
EMBEDDED DERIVATIVES
Q4 (a) Do you agree that the embedded derivative requirements for a hybrid instrument with a
financial host contract should be eliminated?
92% 8%
FAIR VALUE OPTION
Q5 (a) Do you agree that entities should still be
allowed to elect to carry any financial asset or liability at fair value through profit or loss if it
eliminates or significantly reduces a measurement
or recognition inconsistency?
86% 14%
Q6 (b) Should the fair value option be allowed under any other circumstance?
16% 51%
RECLASSIFICATION
Q7 Do you agree that reclassifications should be prohibited?
41% 53%
INVESTMENTS IN EQUITY INSTRUMENTS THAT DO NOT HAVE A QUOTED
MARKET PRICE AND WHOSE FAIR VALUE CANNOT BE RELIABLY
MEASURED
Q8 (a) Do you believe that more decision-useful
information about investments in equity
instruments results if all such investments are measured at fair value?
78% 18%
Q9 (b) Are there circumstances where the benefit of
improved decision-usefulness does not outweigh the costs of providing this information?
41% 37%
INVESTMENTS IN EQUITY INSTRUMENTS THAT ARE MEASURED AT FAIR
VALUE THROUGH OTHER COMPREHENSIVE INCOME (OCI)
Q10 (a) Do you believe that presenting fair value
changes and dividends for particular investments
in equity instruments in OCI would improve financial reporting?
78% 12%
Q11 (b) Do you agree that an entity should be
permitted to present in OCI changes in fair value and dividends of any investment in equity
instruments (other than those held for trading),
only if it elects to do so at initial recognition?
71% 20%
G Steenkamp 183
Discussion of the findings
Classification approach
ED 7/2009 proposed that the four categories of financial assets currently contained in
IAS 39, namely ‘at fair value through profit or loss’, ‘held to maturity’, ‘loans and
receivables’ and ‘available for sale’, should be reduced to only two categories –
amortised cost and fair value. It proposed that a financial asset should be measured at
amortised cost if it has basic loan features and is managed on a contractual yield basis.
Any other financial asset should be measured at fair value. A clear majority of 73% of
the students felt that this was a positive change, while only 18% believed it to be
negative.
Regarding question 1 posed by the IASB, 94% of the students felt that amortised cost
provided decision-useful information if a financial asset or liability has basic loan
features and is managed on a contractual yield basis. This corresponds with the response
of SAICA to the IASB (SAICA, 2009b).
In relation to question 3, 75% of the students did not believe that other conditions (i.e.
other than basic loan features and management on a contractual yield basis) should be
used to identify those instruments that are measured at amortised cost. Some of the
students mentioned that the duration, as well as the identity of the issuer of the loan,
should also be considered. Similarly, SAICA (2009b) did not support the use of other
conditions, but asked that these conditions be better defined, and that guidance be
provided for their application in certain scenarios. SAICA also mentioned many
practical problems and specific instruments that had to be addressed in the guidance
(SAICA, 2009b). In the final international financial reporting standard (IFRS) that
flowed from the exposure draft (known as IFRS 9), the wording was modified slightly
to better describe the conditions for measurement at amortised cost and more guidance
was provided.
Embedded derivatives
ED 7/2009 proposed that embedded derivatives with a financial host should not be split
(bifurcated) into its two parts (namely the embedded derivative and the host) as
currently required by IAS 39. Once again, 76% of the students felt that this was a
positive change, while only 14% believed it to be negative. In addition, 92% of the
students believed that the current requirement to split a hybrid instrument (with a
financial host) should be eliminated (responding to question 4 (a) which the IASB
posed). One student mentioned the fact that the embedded part of the contract should
not be material (greater than 50%), while others felt that splitting the hybrid instrument
in two (as required by the current IAS 39) provides a substance-over-form view of the
economic substance of the transaction. However, another student also mentioned that
this concession would reduce complexity and would thus be in line with the objective of
the exposure draft.
SAICA reported that most of their commentators agreed that eliminating the
requirement to split hybrid instruments with financial hosts would reduce complexities,
but that a minority felt that the current approach (splitting the embedded derivative from
its host) provided more useful information (SAICA, 2009b), thereby having views
184 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
consistent with those of the students. The final standard (IFRS 9) does not require the
bifurcation of embedded derivatives with a financial asset host, in line with ED7/2009.
Fair value option
ED 7/2009 proposed that an entity should still have the option of carrying at fair value a
financial asset that has basic loan features and is managed on a contractual yield basis
(and which would normally be measured at amortised cost) if this reduces or eliminates
a so-called accounting mismatch, a stance similar to that of the IAS 39 in issue at
1 July 2009. In response, 67% of the students felt that the retention of the fair value
option was positive, while 24% believed it to be negative. Regarding question 5, 86% of
the students felt that the fair value option should be permitted, concurring with the
comments submitted by SAICA to the IASB (SAICA, 2009b).
In relation to question 6, 51% of the students did not believe that the fair value option
should be allowed under any other circumstances than those mentioned above, as did
SAICA (SAICA, 2009b). A minority of 16% of the students proposed that it should be
allowed under other circumstances, with a number of students mentioning that the fair
value option should be allowed under all or most circumstances as they felt that fair
value always provides better and more decision-useful information. IFRS 9 allows the
fair value option in the circumstance listed in ED7/2009.
Reclassifications
ED 7/2009 proposed that no reclassifications should be permitted between the two
categories, while a number of reclassifications are allowed under the IAS 39 in issue at
1 July 2009. SAICA (2009b) concurred with this view as it reduces complexities, but
mentioned that some problems could arise, such as the sale and repurchase of assets to
affect a change in category. SAICA also mentioned that reclassifications allow
companies to show more relevant and useful information (SAICA, 2009b).
Interestingly, 51% of the students felt that this was a negative change, while only 35%
believed it to be positive.
Regarding question 7, 53% of the students felt that reclassifications should be
permitted, while 41% felt that they should not be permitted. Many students felt that
reclassifications should be permitted in the following circumstances: if material changes
are made to the contract between the holder and the issuer of the instrument (nature of
instrument changes) or if there has been a change of intention regarding the asset held
(similar to the principle in another accounting standard, namely IFRS 5 Non-current
assets held for sale). One student used the following example: If an entity held a
financial asset with the intention of keeping it and collecting cash flows (carried at
amortised cost) and subsequently decided to sell it, then the fair value might be a much
more useful carrying value for decision-making purposes. In IFRS 9, reclassifications
are permitted if the entity changes its business model, which would be a rare
occurrence.
Investments in equity instruments where the fair value cannot be determined reliably
Under IAS 39, investments in unquoted equity instruments can be carried at cost and
tested for impairment if their fair value cannot be determined reliably. ED 7/2009
G Steenkamp 185
proposed that all investments in equity instruments should be carried at fair value, no
matter the effort or cost involved in determining that fair value. In response, 65% of the
students believed this to be a positive change, while 22% did not.
Regarding question 8 posed by the IASB, 78% of the students felt that carrying all
investments in equity instruments at fair value results in more decision-useful
information. This corresponds to the response of SAICA to the IASB (SAICA, 2009b).
SAICA (2009b) also commented on the fact that testing for impairment is similar to a
fair value valuation. However, some of the students commented on the fact that
measuring an asset at fair value, when the fair value cannot be determined reliably,
might lead to inaccurate information that misrepresents the entity’s financial situation.
Some felt that cost is a better indication if the fair value cannot be determined reliably.
Regarding question 9, responses were almost evenly split between those students who
believed that the costs of determining the fair value of an investment might outweigh
the benefits thereof in certain circumstances, and those who believed that the costs
never outweighed the benefits. This disparity was evident in the narrative comments of
the students. Many believed that the costs of providing the information could outweigh
the benefits thereof, especially if the investment in equity instruments is small in value.
Students also mentioned that determining a fair value for instruments that are not traded
could prove to be very problematic. They proposed that additional disclosures be made
where the fair value cannot be determined reliably or it is impracticable. These
disclosures should provide detailed information about the financial situation of the
issuer as well as liquidation value of the investment.
SAICA (2009b) stated that they were not aware of any circumstances where the benefits
did not outweigh the costs, and proposed that no exemption be made and all
investments in equity instruments be carried at fair value. This view was also adopted in
IFRS 9.
Investments in equity instruments that are measured at fair value through other comprehensive income
ED 7/2009 proposed that an entity be allowed an irrevocable choice at initial
recognition to carry investments in equity instruments (which are not held for trading)
at fair value with changes in fair value recognised in other comprehensive income
(OCI). In response, 65% of the students believed the availability of this choice to be
positive, and 78% believed that allowing this would improve financial reporting
(question 10). Some students, however, believed that this would allow an entity to
exclude the capital loss made on a non-beneficial investment from its bottom line, and
could lead to less useful information about an entity’s activities. SAICA (2009b) stated
that the majority of their respondents did not believe that allowing entities to report fair
value changes in OCI would enhance financial reporting, and felt that this would be
inconsistent with the principles in a number of other accounting standards.
SAICA (2009b) did, however, agree that an entity should only be allowed to report fair
value changes in OCI if the entity elects this at initial recognition, and if the investment
is not held for trading (question 11), concurring with 71% of the students on this matter.
Some of the students mentioned that, if reclassifications are allowed, then an entity
should be able to designate an investment into this category at a later stage as well. The
186 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
students mentioned that additional disclosures should be made in respect of these
investments. IFRS 9 has retained this OCI-option.
Both the students and SAICA (2009b) disagreed with the recognition of dividends
received in OCI, as proposed by ED 7/2009. IFRS 9 also relented on this point, and
requires that the dividends received on such investments be recognised in profit or loss.
Comparing the responses of the students to their exam mark
Table 5: Comparison of ‘Yes/No’ responses in relation to exam mark
Number of question by IASB
Mean exam mark
of student who said
‘Yes’
Mean exam mark
of student who said
‘No’
F-test P
Q1 * * * *
Q3 54.5 51.7 0.40 0.53
Q4 51.2 62.5 5.51 0.02
Q5 51.8 54.7 0.47 0.50
Q6 56.9 50.7 3.05 0.09
Q7 51.2 52.7 0.27 0.60
Q8 52.2 53.0 0.05 0.83
Q9 53.7 52.1 0.33 0.57
Q10 50.6 61.5 7.32 <0.01
Q11 51.2 54.8 1.14 0.29
*As only one student answered “No” to Question 1, no F-test could be done
As can be seen from the above table, except for Questions 4 and 10 no statistically
significant correlations were found between a student’s academic strength and his/her
response to the questions. In relation to the highly technical area of embedded
derivatives (Q4) fewer of the academically stronger students agreed that the embedded
derivative requirements for a hybrid instrument with a financial host contract should be
eliminated; they were happy to remain with the current legislation which requires the
host and embedded derivative to be bifurcated. This difference was statistically
significant at a 2% level. A definite trend is visible in Question 10, where the
academically stronger students were much less likely to allow a choice to report
changes in fair value in OCI, if so elected at initial recognition. This difference was
significant at a 1% level.
Comparing the students’ responses to SAICA and IFRS 9
To enable an understanding of the statistical significance of the similarity between the
students’ comments and both the SAICA submission and the final IFRS 9, the
following process was followed:
G Steenkamp 187
1. A perfect “score” was developed for the SAICA submission – i.e. what SAICA’s
comments were on each of the 10 IASB questions listed above in table 4.
2. A perfect “score” was also developed for the final IFRS 9 standard – i.e. what the
final standard adopted in relation to each of the questions.
3. The response of each student was then rated according to each of the scales above
(in both 1 and 2), allocating a mark out of 10 to each student, denoting the number
of responses by the specific student that correlated with either the response in the
SAICA comment letter or IFRS 9. This mark was then translated to a percentage.
The results of the above can be seen in the table below:
Table 6: Statistical evaluation of significance of similarity
Students compared to SAICA
Students compared to
IFRS 9
Mean 63.53% 71.37%
Standard deviation 17.53% 19.39%
From the above table, it can be seen that the answers of the students corresponded well
to both SAICA’s comment letter and the final IFRS 9. Interestingly, the students’
answers showed a higher correlation with IFRS 9 than the SAICA comments.
Furthermore, a statistical analysis was done to determine if there was any correlation
between the similarity of a students’ response to either SAICA or IFRS 9, and their
final exam mark, i.e. their academic strength. The results of this can be seen in the table
below:
Table 7: Statistical evaluation of significance of similarity to academic strength
Students compared to SAICA
Students compared to
IFRS 9
Spearman r -0.09 -0.11
Spearman p 0.55 0.44
No statistical correlation was found between the student’s academic strength and to
which extent his/her responses corresponded to SAICA’s comment letter and the final
IFRS 9. This might indicate that academically weaker students are not necessarily less
adept at formulating a correct or valid opinion, but that their academic inferiority might
stem from other issues, such as examination technique.
CONCLUSION AND RECOMMENDATIONS
Conclusion
Following the recent international financial crisis, the IASB initiated the redevelopment
of IAS 39, the primary accounting standard dealing with financial instruments. The first
stage of this process addressed the classification and measurement of financial assets
(such as investments in shares, bonds and loans) through the issue of ED 7/2009, which
188 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
proposed certain fundamental changes to IAS 39. This study sought to investigate the
perceptions of a CTA class regarding the changes proposed by ED 7/2009 and to
compare the students’ perceptions to the comments presented by SAICA to the IASB,
as well as the final IFRS 9.
In general, it was found that the students were positive about the proposed changes. The
students applied their theoretical knowledge well and could formulate valid opinions,
although their practical experience and examples were limited. This indicates that their
university education has equipped them to some extent to think critically, to evaluate
new accounting standards and their appropriateness, to be innovative and to solve
problems – as required under the new Competency Framework: Detailed Guidance for
Academic Programmes’ pervasive skills. It was also found that the perceptions of the
students mostly corresponded to the comments submitted to the IASB by SAICA, and
to the position taken in the final IFRS 9.
The views of the students and the comments of SAICA were almost identical regarding
the changes in the classification approach, accounting for embedded derivatives and the
fair value option. The students believed that some reclassifications should be allowed,
which differed from SAICA’s view. Interestingly, in IFRS 9 (the final standard
produced from ED 7/2009 after international comments had been received),
reclassifications are allowed in limited circumstances. Many students differed from
SAICA regarding whether the costs of measuring the fair value of an unquoted equity
instrument might outweigh the benefits thereof in a certain situation. The students and
SAICA also differed regarding the choice to account for the fair value changes on
equity investments not held for trading in other comprehensive income (OCI). SAICA
did not regard this as appropriate, while the students did. The final IFRS 9 adopted the
OCI-option.
From the statistical evaluation, no significant correlation was found between academic
strength and similarity of a student’s answers with both SAICA and IFRS 9. This could
point out that academically weaker students can still form valid opinions, but struggle to
apply their knowledge in an examination situation.
Recommendations
Academics could consider the usefulness of exposing CTA students to current exposure
drafts. This could be a way to foster life-long learning, to prepare students for an ever-
changing business and accounting environment, to help them be aware of upcoming
changes and develop the ability to assess the appropriateness of new accounting
standards. The exposure drafts (and the questions contained therein) could be used as a
tool in the education process.
Students could be included in the consultation process, when an exposure draft is
issued, as they have detailed academic knowledge of the accounting standards. It is
agreed that their practical knowledge is limited, but their viewpoint would be a
supplementary one to that of practice.
Lastly, it must be noted that the Competency Framework: Detailed Guidance for
Academic Programmes will have to be implemented at the academic programme level
from 2011. This would have certain implications for accounting academics, as the so-
called pervasive skills (such as awareness of changes in the business environment,
G Steenkamp 189
analytical/critical thinking, problem-solving skills, innovation, and change
management) would have to be developed in students to some extent. From the findings
of this research it can be seen that the current programmes do develop these skills to
some extent, but still more emphasis should be placed on the practical implications of
theoretical knowledge. As such, students should be made more aware of the practical
issues that could arise on the application of a specific financial standard in real life
situations and the business context in which such transactions might take place.
Students should also be allowed the opportunity to express their views and critically
assess both current and forthcoming accounting standards.
REFERENCES
Arnold, P.J. (2009). Global financial crisis: The challenge to accounting research.
Accounting, Organizations and Society, 34: 803-809.
Exposure draft 7/2009 (ED7/2009), Financial instruments - Classification and
Measurement. (2009). International Accounting Standards Board (IASB). London.
Gammie, B., Gammie, E. and Cargill, E. (2002). Personal skills development in the
accounting curriculum. Accounting Education, 11 (1): 63–78.
IAS 39, Financial Instruments: Recognition and Measurement. (1999). International
Accounting Standards Board (IASB). London.
IFRS 9, Financial Instruments. (2009). International Accounting Standards Board
(IASB). London.
Lopes, P.T. and Rodrigues, L.L. (2008). Accounting for financial instruments: a
comparison of European companies’ practices with IAS 32 and IAS 39. Research in
Accounting Regulation, 20: 273-275.
Nettleton, S., Litchfield, A. and Taylor, T. (2008). Engaging professional societies in
developing work-ready graduates. Proceedings of the 31st Annual International
HERDSA Conference, 1-4 July, Rotorua, New Zealand. http://www.herdsa.org.au/wp-
content/uploads/conference/2008/media/Nettleton.pdf. Accessed 17 May 2011.
South African Institute of Chartered Accountants (SAICA). (2009a). Competency
Framework: Detailed Guidance for Academic Programmes.
https://www.saica.co.za/Portals/0/documents/FINAL_-_Detailed_Guidance_for_the_
academic_programme_(Nov_2009).pdf. Accessed 17 May 2011.
SAICA. (2009b). SAICA submission on Exposure Draft Financial Instruments:
Classification and Measurement. http://www.saica.co.za/Portals/0/documents/SAICA%
20Submission% 20on%20Financial%20Instruments%20%20Classification%20and%20
Measurement.pdf. Accessed 17 May 2011.
Schipper, K. (2003). Principles-Based Accounting Standards. Accounting Horizons,
17(1): 61-72.
Wessels, P.L. and Steenkamp, L.P. (2009). An investigation into students’ perceptions
of accountants. Meditari Accountancy Research, 17(1): 117-132
TF Prinsloo & M Oberholzer 191
A Note: Estimating the relationship between
environmental performance and economic
performance of South African mining
companies
TF Prinsloo
Faculty of Economic and Management Sciences, North-West University
M Oberholzer
Faculty of Economic and Management Sciences, North-West University
Received: January 2011 SAJAR
Revised: May 2011, September 2011 Vol 25 No. 1
Accepted: September 2011 2011
pp.191 to 207
The purpose of the study is to determine the relationship between environmental performance
and economic performance in the South African mining industry, where the economic
performance consists of market-based and accounting-based estimates. In this regard,
Spearman’s ranking-order correlation was applied. The study found that there is some
contradiction in the direction of the relationship between accounting-based estimates and
market-based estimates and their relationship with environmental performance. Furthermore,
there is some contradiction in the direction of the relationship between economic performance
and the environmental performance of the different mining sectors. The main limitation is that
only a limited number of mining companies reported on environmental-related issues. Only ten
mining companies provided sufficient annual data over a five-year period. The contribution of
this study is that it pointed out that causation can be argued both ways between environmental
performance and economic performance. When a change in the environmental performance
influences the economic performance, it can be determined whether it does/does not pay to be
green. This one-sided view, used in previous studies, presents probably only half of the bigger
picture.
KEYWORDS
Accounting-based estimates; environmental performance; economic performance; market-based
estimates; sustainability development
Contact
192 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
INTRODUCTION
In recent years, the growing importance of environmental and social issues has put
pressure on companies to implement environmental and social systems
(Laurinkevičiūtė, Kinderytė, & Stasiškienė, 2008:69). With the pressure on companies
to improve environmental performance, accounting can provide a valuable tool that
enables companies to respond to environmental challenges. The Kyoto Protocol for
Greenhouse Gas Emissions, the World Summit on Sustainable Development held in
Johannesburg, South Africa, in 2002 and the United Nations Climate Change
Conference held in Copenhagen, in December 2009, show the concern that stakeholders
have about the negative impact that industries have on the environment and humans
(Ambe, 2008:54). Companies have started to show a lot of interest in the areas of
corporate social responsibility as well as social and environmental accounting. As a
result of this interest, a growing number of companies are publishing triple bottom line
and sustainability reports (Brown & Fraser, 2006:103). Many companies are voluntarily
reporting according to the Sustainable Reporting Guidelines (GRI, 2002), provided by
the Global Reporting Initiative (GRI) to enhance the quality of environmental reporting
(Ambe, 2008:54).
In the past, many managers and companies viewed environmental protection as
additional costs and an expenditure that will lower profits (Telle, 2006:195). Bragdon
and Marlin (1972), Jaggi and Freedman (1992), Pava and Krausz (1995), Hart and
Ahuja (1996), King and Lenox (2001), Rivera (2001), Orlitzky, Schmidt and Rynes
(2003) and Ambec and Lanoie (2008) challenged this idea by asking the question does
it pay to be green? Therefore, they have tested this relationship between environmental
performance and economic performance. These studies found mainly positive
relationships (at different degrees) between environmental performance and economic
performance. The reasons are that better environmental performance will lead to higher
profits, because by reducing pollution, future cost savings are made by improved
efficiency, reducing compliance costs and minimising future liabilities (Porter & Van
der Linde, 1995:98). This links with the view of Bhat (1999:450) that lower pollution
means lower costs, e.g. waste discharge consists of labour, materials and equipment
hours that the company has already paid for. No value is added through waste
management; instead, costs are added through transportation, handling and disposal. If
pollution is high, it implies that the manufacturing processes are inefficient. If pollution
is low, it will lower energy usage, make the workplace safer, reduce production costs
and enhance the quality of the product. A positive relationship between environmental
performance and economic performance will make mining companies more aware of
environmental issues and the economic gain they have if the environment is taken care
of (Gallarotti, 1995; Hart, 1997; Orlitzky et al., 2003). Orlitzky et al. (2003) also argue
that certain environmental regulations can be relaxed if there is a positive relationship
between environmental performance and economic performance. Consequently, a
negative relationship implies that economic gains are made at the expense of the
environment.
The above-mentioned studies mainly investigated developed countries, and the
economic performances were estimated by using, except Orlitzky et al. (2003) and Pava
and Krausz (1995), either accounting-based or market-based economic performance
estimates, but no combination of the two. Furthermore, it is assumed that the causation
is that environmental performance influences economic performance, while the opposite
is neglected. However, in spite of widespread research on the relationship between
TF Prinsloo & M Oberholzer 193
environmental performance and economic performance, the research questions are:
What is this relation in an emerging economy such as South Africa and is there a
difference in the results when accounting-based and market-based economic
performance estimates are used? The importance of this topic for accountants is that the
economic performance includes both accounting-based and market-based performance
estimates and, although there is a growing awareness of environmental accounting in
South Africa, studies pointed out that there is still a need to improve this reporting as
well as the use of such environmental accounting information (KPMG, 2001; Mitchell
& Quinn, 2005; Ambe, 2007).
This study focuses on the South African mining industry, since they disclose
environmental information as a result of their immense environmental impact (De
Villiers & Barnard, 2000), which is a much greater impact on the environment in
comparison to other companies (Antonites & De Villiers, 2003). This study included
only South African mining companies with demonstrable measures of environmental
performance, i.e. companies that are listed on the GRI database. Data from only ten
companies in three mining sectors were found appropriate for this study. Using this
limited data was a choice of trade-off between quantity and quality of data. The premise
is that it is more important to investigate these few companies rather than many
companies whose environmental performance is not objectively verified from their
reporting. Therefore, the results of this study must be seen against the backdrop of a
lack of controls and its inability to generalise the results. With this in mind, the purpose
of the study is to determine the relationship between environmental performance and
economic performance of the sample companies. Evidence of inter alia trends and
contradictions from this exploratory research, where accounting-based and market-
based economic measures are set opposed to the environmental performance in the three
sectors, can be helpful to detect areas that need to be further investigated, especially to
support the requirement of integrated reporting as stated in the Code of Governance
Principles for South Africa 2009 (King III) (Robberts, 2011:13).
The study used Spearman’s ranking-order correlation and found some contradiction
between the direction of accounting-based estimates and market-based estimates and
their relationship with environmental performance. Furthermore, it found some
contradiction in the direction of relationship between economic performance and
environmental performance of the different mining sectors. The contribution of this
study is that it pointed out that causation can be argued both ways. When a change in
the environmental performance influences the economic performance, it can be
determined whether it does/does not pay to be green. This one-sided view presents
probably only half of the bigger picture.
The organisation of the paper is a background of sustainable development and triple
bottom line accounting, environmental performance and economic performance in the
next section, followed by the hypothesis, methodology of the study and the empirical
results. The study is concluded in the final section.
BACKGROUND
Sustainable development and triple bottom line accounting
Sustainable development seeks economic growth while ensuring future generations’
ability to do the same while protecting the environment at the same time (Shi, 2002:88).
194 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Sustainable development consists of three dimensions, namely environmental
sustainability, social sustainability and economic sustainability. To meet the needs of
future generations, these three dimensions need to be integrated to address the balance
between dimensions of sustainability (IUCN, 2006:2). The interaction between these
three dimensions is also known as triple bottom line accounting, which measures their
performances and interaction (Figure 1).
Figure 1: Three dimensions of sustainable development (IUCN, 2006:2)
In South Africa, the King III report states that a firm’s strategy, risk, performance and
sustainability are inseparable, and therefore it requires an integrated report from South
African firms to give the full circle picture of a firm (Robberts, 2011:13). As a result of
the King III report, the JSE added a new listing requirement, stating that firms must do
integrated reporting for financial years (Le Roux, 2011:22). The social dimension
includes elements such as number of employees and payments for taxes, donations and
to suppliers for goods and services (Anglo Platinum Limited, 2008:62-63). The social
dimension is excluded from this study, which focuses only on the relationship between
two of the dimensions of triple bottom line accounting, namely economic performance
and the environmental performance. The reason is, as highlighted earlier, to contribute
to the increasingly important question as to whether there is a direct link between these
two dimensions.
Environmental performance
In this study, the environmental performance is estimated by considering three
elements, i.e. greenhouse gas (GHG) emissions, water usage and energy usage. These
three elements are indicated in the GRI reports. A mining company with a high
environmental performance is one with relatively low GHG emissions, combined with
low water and energy usage. This environmental performance is important since it has a
direct impact on the most important threat to mankind, namely climate change (Moser,
2010; Morrissey & Reser, 2007). The effect of climate change can be seen in the
increase of global air and ocean temperatures, widespread melting of sea ice and the
rising of global sea levels. The average temperature went up by 4% or 0.55˚C from
1970 to 2005 (IPPC, 2007:30).
TF Prinsloo & M Oberholzer 195
The first element in estimating environmental performance is GHG emissions. Global
warming is caused by the overproduction of GHG. The main sources of GHG due to
human activities are (IPPC, 2007:37):
The burning of fossil fuels and deforestation resulting in a higher concentration of
carbon dioxide (CO2) emissions;
Fully-vented septic systems that enhance and target the fermentation process –
livestock enteric fermentation, manure management and paddy rice farming are
sources of atmospheric methane (CH4);
The use of fertilisers in agricultural activities is a source of nitrous oxide (N2O)
emissions; and
Halocarbons (a group of gases containing fluoride, chlorine and bromine) are
emitted from the use of refrigeration systems, fire suppression systems and
manufacturing processes.
The second element in estimating environmental performance is water usage. With
South Africa being a climatically sensitive and water-stressed country, the effects of
climate change will have a very big impact on crops and other agricultural activities
(SECCP, 2009:17). The availability and quality of water are major concerns that will
have a very big impact on the South African economy and it is anticipated that the
problem will only worsen (Kiker, 2000). Due to the mining industry’s strong
dependence on water, such companies are very vulnerable to the effects of climate
change.
Companies are expecting that GHG emission regulations and carbon taxes will soon be
implemented by the South African government. These regulatory requirements will
have direct and indirect implications on the third element of environmental
performance, namely the price increases of energy usage (CDP, 2008:61). Despite the
risks facing the mining industry, climate change has also created a number of
opportunities. The potential for emission trading projects, the opportunities around
clean development mechanism projects and carbon-trading opportunities are being
considered by companies as investment opportunities (CDP, 2008:62). According to a
study done by the Carbon Disclose Project (CDP) (2008), mining companies have
identified, inter alia, the following opportunities based on the effect of climate change,
namely the development of new technologies that will result in large-scale energy
savings and a reduction in GHG emissions, and investing in projects in renewable
energy that will provide long-term carbon trading opportunities.
Economic performance
As mentioned earlier, the economic performance consists of both an accounting-based
approach and a market-based approach. This study used a combination of different
approaches previously used. Studies that used the accounting-based approach are
Ambec and Lanoie (2008), Rivera (2001), Hart and Ahuja (1996), and Jaggi and
Friedman (1992). The measures included are revenue, net income, return on sales,
return on equity, return on assets, cashflow to equity and cashflow to assets. Studies that
used the market-based approach are King and Lenox (2001) and Bhat (1999). The
measures included are market value of shares and market-to-book value of shares.
196 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Since accounting returns are subject to managers’ discretion of funding different
projects and adopting certain policies, the rationale of using an accounting-based
approach is that it reflects managerial performance within the business environment
(Orlitzky et al., 2003:408). Therefore, a company’s internal performance is measured to
some extent. The rationale of using a market-based approach is that the market price of
equity relies on the perspective of market participants’ perception of past, current and
future share returns and risks (Orlitzky et al., 2003:407-408). This perception may be
influenced by the extent a company is harming the natural environment. Comparing the
market-based performance estimates with environmental impact must be done with
great caution, because the market price of equity discounts a great deal of information
and environmental impact is only one aspect (Roy & Ghosh, 2011:13).
HYPOTHESIS
The conceptual framework is that an improvement in environmental performance will at
the same time increase both the accounting-based and market-based economic
performance estimates and vice versa. The null- and alternative hypotheses will be
helpful to test this statement.
Ho: There is no relationship between environmental performance and economic
performance (accounting-based and market-based) of the companies under review.
H1: There is a relationship between environmental performance and economic
performance (accounting-based and market-based) of the companies under review.
METHODOLOGY
Sample and data
The population for this study was selected from the following companies:
South African mining companies that are listed on the JSE Limited Stock
Exchange in Johannesburg, South Africa;
South African mining companies that subscribe to the South African Business
Council for Sustainable Development hosted by the National Business Initiative
(NBI) (www.nbi.org.za); and
South African mining companies that report on environmental performance based
on the GRI guidelines and that are listed on the GRI database.
A convenience sample of ten South African mining companies was selected for the
study and their financial information for the past five years was used. The reason why
only ten mining companies were selected was due to the limited environmental data
available and due to the limited mining companies that reported on environmental-
related issues prior to 2005. The ten companies operate in the following sectors of the
mining industry: Four in the platinum-mining sector; three in the gold-mining sector
and three in the coal-mining sector. This study attempts to contribute to the literature by
considering the potential impact of environmental performance on economic
performance and should be considered an exploratory research project. Blumberg,
Cooper and Schindler (2008:146) and Terre Blance, Durrheim, and Painter (2008:44)
TF Prinsloo & M Oberholzer 197
stated that an exploratory study is undertaken when not much is known about the
situation at hand or when limited information is available about a particular problem or
scenario. These ten companies provide 50 data points over the five years under review.
The small sample size corresponds to other similar studies, e.g. Pava and Krausz
(1995), who compared both accounting-based and market-based estimates to
environmental performance and used 14 companies and reported seven r-values.
Wokutch and Spencer (1987) compared accounting-based estimates to environmental
performance and investigated 12 companies and reported three r-values. Wolfe (1991)
investigated the quality of nine companies’ corporate social disclosure.
Documentary data from internal company sources, such as annual reports and
sustainability reports, were used to acquire the information needed for this study. The
McGregor BFA database supplied the economic performance information used in this
study.
Estimating economic performance
From an accounting-based point of view, economic performance was estimated by
using the three profit ratios of the Du Pont analysis, because their strength is that they
aggregate the firm’s performance in three broad categories, namely income, investments
and capital structure (Correia, Flynn, Uliana & Wormald, 2007:5-20). The ratios are
return on sales (ROS), return on assets (ROA) and return on equity (ROE).
The first market value ratio is the price/earnings (P/E) ratio, which is an indication of
how much investors would be willing to pay per Rand (South African currency) of
profit (Fairfield & Harris, 1993:591). The price/book value (P/B) ratio is a market value
ratio that provides an indication of expectations of future performance by relating the
market price of a share to the book value of the share (Dunis & Reilly, 2004:231). The
value of any organisation is the present value of the future free cashflows and therefore
the price/cashflow ratio (P/CF) is useful, especially where the price of a share is more
related to cashflows than net income (Park & Lee, 2003:335).
Estimating environmental performance
The environmental effects of a company’s activities were estimated through the
company’s environmental performance. Some of the frequently applied estimates are
those of Hart and Ahuja (1996), and King and Lenox (2001), who measured the
percentage change in the environmental performance elements. However, the approach
of Jaggi and Freedman (1992), who created an index of the emissions of various
pollutants to estimate environmental performance, was used in this study. Only GHG
emissions are used to explain this approach; however, in this study water usage,
measured in m3, and energy usage, measured in GJ, were also used to estimate
environmental performance. Environmental performance will be measured by
normalising the emissions of each pollutant with production. A company, whose
normalised production is low relative to another company’s normalised production, is
greener. This is known as the Jaggi-Freedman Index (JFI). JFI ranges from zero to one,
and the closer JFI is to one, the greener the company.
JFIit. =
∑
198 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
epit = Emissionspit/Productionit
where:
epit = Normalised emissions
p = Pollutant
i = Company
t = Time period
The following example will explain how environmental performance is measured.
Table 1 exhibits the production in tons for companies A, B and C for four years. This
data will be used to calculate the normalised emissions for each company. Table 1 also
exhibits the GHG emissions in tons for companies A, B and C for years one to four.
Table 1: Production (tons) and GHG emissions (tons)
Company Year 1 Year 2 Year 3 Year 4
A: Production (tons)
GHG emissions (tons)
40 000
20 000
30 000
15 900
50 000
28 500
45 000
23 850
B: Production (tons)
GHG emissions (tons)
44 000
40 480
39 000
33 930
40 000
39 200
37 000
41 810
C: Production (tons)
GHG emissions (tons)
20 000
14 600
25 000
18 500
30 000
24 000
35 000
29 750
Three steps are used to calculate environmental performance. Step 1 normalises
emissions with production. This is calculated by dividing emissions with production for
each company for the specific year. Table 2 indicates the normalised emissions for
companies A, B and C for the four years (epit = Emissionspit/Productionit).
Table 2: Normalised emissions
Normalised
emissions (epit)
Year 1 Year 2 Year 3 Year 4
A 0.50* 0.53 0.57 0.53
B 0.92 0.87 0.98 1.13
C 0.73 0.74 0.80 0.85
* 40 000 tons / 20 000 tons = 0.50
Step 2 is to calculate the baseline pollutant. After the normalised emissions have been
calculated, the company with the lowest emissions of pollutant p will be used as the
baseline. The other pollutants will then be valued relative to the baseline company. The
pollutants of company A (0.50, 0.53, 0.57 and 0.53) are used as the baseline, because it
has the lowest normalised emissions for all four years. Step 3 is to calculate
environmental performance by dividing the baseline emissions into the normalised
emissions. The company whose JFI is closest to one, performs the best (Table 3).
TF Prinsloo & M Oberholzer 199
Table 3: Jaggi-Freedman index
JFI Index Year 1 Year 2 Year 3 Year 4
A 1** 1 1 1
B 0.54*** 0.61 0.58 0.47
C 0.68 0.72 0.71 0.62
** 0.50/0.50 = 1 *** 0.5/0.92 = 0.54
Company A has the best environmental performance for all four years because its JFI is
closest to one (note that Company A is the baseline company). Company C has the
second best environmental performance, and Company B has the third best
environmental performance.
Relationship between economic performance and environmental performance
The relationships between economic performance and environmental performance were
determined separately for each mining sector. This will provide information to make a
conclusion with regard to each sector. The reasons why all the mining companies could
not be grouped together are:
Environmental performance estimates are industry and mineral specific. Because
the production output, e.g. gold, platinum and coal, is measured in ounces and tons
respectively, and because the pollutants are normalised with production, the
environmental performances between the minerals cannot be compared. The
environmental performance will not give an accurate reflection of the company’s
environmental performance if it is compared to other companies that mine with
different minerals. Therefore, the environmental performance of platinum can only
be compared with the environmental performance of other platinum mine
companies, and so forth.
The economic performance estimates of each company are company and industry
specific and cannot be compared with the economic performance estimates of other
companies.
Because gold-mining companies are depleting asset companies, the economic
performance estimates of gold-mining companies should not be compared with the
economic performance estimates of other mining companies as a result of reporting
differences (McGregor BFA, 2010).
Although regression and correlation analysis is widely used to define the structural
relationship between variables and the strength of this identified association (Wegner,
2007:407), limited data were available due to the limited number of mining companies
that reported on environmental-related issues prior to 2005. In a time series analysis, the
simple linear regression model needs at least 10 data points (10 time periods) to be
scientific (Kvanli, Pavur & Keeling, 2006:413), thus the five data points (5 years, 2005
to 2009) of this study are not enough to perform scientific results. When the data of
each sector for each year are grouped together, i.e. three times five years, four times
five years and three times five years for the gold-, platinum- and coal-mining sectors,
respectively, the problem arises that a previous year’s data may have a direct impact on
the next year’s data, and the effect of autocorrelation could not be determined as too
200 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
little time-series data were available. Nevertheless, to perform quantitative analysis to
determine the existence and nature of the trends, it was decided to use the rank order
correlation of Spearman (not tested for autocorrelation). Rank order correlation is a
non-parametric technique for qualifying the relationship between two variables. Non-
parametric means that the correlation statistics are not affected by the type of
mathematical relationship between variables, unlike the least square regression analysis
that requires the relationship to be linear (Vose, 1996:33). The Spearman rank order
correlation coefficient is a more general measure of any kind of monotonic relationship
between x and y. This measure is based on ranks and therefore not as sensitive for
outliers (Millard & Neerchal, 2001:534).
EMPIRICAL RESULTS
Table 4 exhibits the results that are based on Spearman’s correlation coefficient.
Wegner (2007:266-267) provides the following rules regarding to the level of
significance: A significance level that exceeds ten per cent implies it can be accepted
that there is not sufficient evidence to infer that H1 is true. A significance level that lies
between ten per cent and five per cent implies weak sample evidence, which is not
statistically significant and cannot be used to reject H0 in favour of H1. If the p-value is
smaller than one per cent, and between one per cent and five per cent, then there is
overwhelming strong sample evidence and strong sample evidence, respectively, to
reject H0 in favour of H1. Therefore, only the relationships which are significant in
Table 4 at one per cent and five per cent levels will be further discussed, since the null-
hypothesis will be rejected at a significance level of one and five per cent, where ρ < α
= 0.01 and 0.05, respectively (two-tailed).
The following summarises where H0 is rejected in favour of H1:
Gold-mining companies
Accounting-based economic performance and environmental performance: There
is a negative statistically significant relationship between ROS and environmental
and water usage, and ROE and water usage.
Market-based economic performance and environmental performance: P/E has a
negative statistically significant relationship with three of the four environmental
performance indicators where it is positive between P/B and three of the four
environmental performance indicators. P/CF has a significant negative relationship
with energy usage.
Platinum-mining companies
Accounting-based economic performance and environmental performance: All
these relationships, except one, are negative and statistically significant to reject
H0.
Market-based economic performance and environmental performance: P/CF has a
positive significant relationship with all four environmental performance indicators.
Coal-mining companies
Accounting-based economic performance and environmental performance: There
is a positive significant relationship between ROE and energy usage and GHG
emissions.
Market-based economic performance and environmental performance: P/B has a
positive significant relationship with all the environmental performance indicators.
TF Prinsloo & M Oberholzer 201
Table 4: Spearman’s correlation coefficient: Trends and significance
Economic performance
Accounting-based Market-based
ROS ROE ROA P/E P/B P/CF
Gold mining
Environmental performance# Neg** Neg Neg Neg*** Pos*** Neg
Water usage Neg** Neg** Neg Neg** Pos*** Neg
Energy usage Neg Pos Pos Neg* Pos Neg**
GHG emissions Neg Neg Neg Neg** Pos** Neg
Platinum mining
Environmental performance# Neg*** Neg** Neg*** Pos Pos Pos***
Water usage Neg*** Neg Neg** Pos Pos Pos**
Energy usage Neg*** Neg** Neg*** Pos Pos Pos***
GHG emissions Neg*** Neg** Neg*** Pos Pos Pos***
Coal mining
Environmental performance# Pos Pos* Pos Neg Pos** Neg
Water usage Neg Neg Pos Pos Pos** Pos
Energy usage Pos* Pos** Pos Neg Pos** Neg
GHG emissions Pos* Pos** Pos Neg Pos** Neg
#Aggregation of water usage, energy usage and GHG emissions.
* Significant at 10% (two-tailed)
** Significant at 5% (two-tailed) *** Significant at 1% (two-tailed)
CONCLUSION
This study investigated the relationship between the economic performance
(accounting-based and market-based) estimates and environmental performance of ten
gold-, platinum- and coal-mining companies in the emerging economy of South Africa
from 2005 to 2009. The null-hypothesis states that there is no relationship between
environmental performance and accounting-based and market-based economic
performance of the companies under review. Spearman’s rank-order correlation was
used to determine the relationship.
There are a few limitations of the study that restricted the scope and in turn affected the
outcomes of the study. Firstly, limited data was available due to limited mining
companies that reported on environmental-related issues prior to 2005. Furthermore,
only the market leaders in the mining industry reported on their environmental
performance and the smaller players did not report on their environmental performance.
If more mining companies reported on their environmental performance, the sample
size would have been bigger and the JFI could discriminate to a greater extent between
companies. Secondly, environmental performance estimates are industry and mineral
specific and economic performance estimates are also industry specific. Therefore,
gold-, platinum- and coal-mining companies were separately investigated. Data of only
five years were available. This is not a sufficient time period to test for the effect of
autocorrelation in the data. A third limitation that should be noted is that possible
double counting may have taken place with the analysis of the market-based economic
202 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
performance estimates when coal-mining companies were tested. The reason is that two
coal-mining companies are not listed on the JSE Limited and therefore no market
performance estimates are available. The market performance estimates of the mother
companies, which are listed, were used.
The research question seeks an answer as to whether there is a relationship between
economic performance and environmental performance, and whether there is a
difference between using accounting-based and market-based economic performance
estimates. Definite trends were recognised by using Spearman’s rank-order correlation.
It is only of value to report on the relationships that are statistically significant to reject
the null-hypothesis in favour of the alternative hypothesis.
For gold-mining companies, it was found that some of the accounting-based economic
performance estimates (ROS and ROE) have a negative statistically significant
relationship with some environmental performance estimates. For platinum-mining
companies, almost all the accounting-based economic performances have negative
statistically significant relationships with all the environmental performance estimates.
These findings correspond with Jaggi and Freedman (1992) who mainly found negative
relationships between environmental performance and accounting-based estimates,
which concluded that it does not pay to be green for these companies. A contradiction is
found with regard to coal-mining companies, where accounting-based economic
performance estimates (ROE) indicate a positive statistically significant relationship
with most of the environmental performance indicators. This corresponds with the
findings of Rivera (2001), who also used accounting-based estimates, and concluded
that it pays to be green for these companies.
Although the above-mentioned conclusions are developed to satisfy the research
question and the hypotheses, they are not necessarily true, since causation can be argued
both ways and the relationship can be argued both ways. Therefore, a negative
relationship can imply the following:
Firstly, that an increase in environmental exploitation (decrease in environmental
performance) increases a company’s profitability, implying it does not pay to be
green, and the opposite, taking more care of the environment (increase in
environmental performance), is costly and decreases a company’s profitability,
implying it does not pay to be green.
Secondly, from an economic perspective, a negative relationship can imply that an
increase in a company’s profitability is a trigger for increased environmental
exploitation (degrease in environmental performance), and the opposite, a decrease
in a company’s profitability, implies that there is less money available to exploit
the environment (increase in environmental performance).
When the relationship is positive, it can imply the following:
Firstly, from an environmental perspective, an increase in exploiting the
environment (decrease in environmental performance) may imply inefficient
processes that decrease profits, implying it pays to be green, and the opposite,
taking care of the environment (increase in environmental performance), implies
efficient processes that increase a company’s profitability, implying it pays to be
green.
TF Prinsloo & M Oberholzer 203
Secondly, from and economic perspective, a positive relationship can imply that
increasing a company’s profitability indicates that it can afford to be green, and
the opposite, a decrease in a company’s profitability, implies that it cannot afford
to be green.
For gold-mining companies it was found that the market-based economic performance
estimate (P/E and P/CF) has a negative statistically significant relationship with some
environmental performance estimates. This corresponds with the findings of Bhat
(1999), but is contradictory to the P/B ratio that has a positive statistically significant
relationship with the environmental performance. Furthermore, it is also contradictory
to P/CF and P/B of platinum and coal-mining, respectively, which also have a positive
relationship with the environmental performance. Again, causation can be argued both
ways and the relationship can be argued both ways. Therefore, similar arguments can be
raised as the above-mentioned between accounting-based estimates and environmental
performance. The point is, as mentioned earlier, using market-based estimates should be
done with great care, since the market price of equity discounts a great deal of
information where environmental performance is only one aspect. Finally, another
contradiction is that the direction of the relationships between economic performance
and environmental performance differ among the three sectors. It was evident that coal-
mining companies tend to have a more positive relationship between environmental
performance and economic performance than the gold- and platinum-mining
companies.
To summarise, the study indicated that there is a contradiction between the direction of
accounting-based estimates and market-based estimates and their relationship with
environmental performance. Furthermore, there is a contradiction in the direction of
relationship between economic performance and environmental performance of the
different mining sectors. The contribution of this study is that it pointed out that
causation can be argued both ways, i.e. where a change in the environmental
performance influences the economic performance and, the opposite, where a change in
the economic performance influences the environmental performance. The formal
causation was used by a number of previous studies (see Introduction) to answer the
question, does it pay to be green? Ignoring the latter implies ignoring the other half of
the bigger picture. The practical implication is that studies that investigate relationships
between the environmental performance and economic performance should be argued
from both sides.
The value of the study is found in the fact that it is the first study to investigate the
relationship between environmental performance and economic performance in the
South African mining industry, which operates in an emerging economy. Further
research should be done to determine which of accounting-based, market-based or a
combination of these economic measures are the most suitable to set opposed
environmental performance. Research should also be done to determine which side of
causation is dominant to understand which of environmental performance or economic
performance is the dependent and which is the independent variable. Following this, the
relationship should also be determined where time-lags are taken into account to
understand whether there is a lag between the time of change in the independent
variable and the change the dependent variable. Another gap that has also not been
investigated is to set a benchmark for mining companies with regard to the efficiency
that environmental exploitation is converted into economic performance. The results of
such further studies will be helpful to develop a framework of best practices between
204 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
the different mining sectors and between individual companies in a sector. This
framework can be used to enhance the quality of integrated reporting.
REFERENCES
Ambe, C. (2007). Perspectives on environmental management accounting (EMA) in
South Africa, Southern African Journal of Accountability and Auditing Research, 7:59-
66.
Ambe, C. (2008). Environmental management accounting in South Africa: Status,
challenges and implementation framework (PhD thesis). Pretoria: Tshwane University
of Technology.
Ambec, S. and Lanoie, P. (2008). Does it pay to be green? A systematic overview,
Academy of Management Perspectives, 22(4):45-62.
Anglo Platinum Limited. (2008). Annual report: Volume 2 – Sustainable development
report, [On-line]. Available from: http:www.angloplatinum.com. (Accessed on 20
March 2009).
Antonites, E. and De Villiers, C.J. (2003). Trends in South African corporate
environmental reporting: A research note, Meditari Accountancy Research, 11:1-10.
Bhat, V.N. (1999). Does it pay to be green? International Journal of Environmental
Studies, 56(4):497-507.
Blumberg, B., Cooper, D.R. and Schindler, P.S. (2008). Business research methods.
2nd
European Edition. London: McGraw-Hill.
Bragdon, H.H. and Marlin, J. (1972). Is pollution profitable? Environmental virtue and
reward: Must stiffer pollution controls hurt profits? Risk Management, 19(4):9-18.
Brown, J. and Fraser, M. (2006). Approaches and perspectives in social and
environmental accounting: An overview of the conceptual landscape, Business Strategy
and the Environment, 15(2):103-117.
Carbon Disclose Project (CDP). (2008). Report 2008: JSE Top 100, [On-line].
Available from: www.cdproject.net. (Accessed on 19 Jan. 2010).
Correia, C., Flynn, D., Uliana, E. and Wormald, M. (2007). Financial Management. 6th
Edition. Cape Town: Juta.
De Villiers, C.J and Barnard, P. (2000). Environmental reporting in South Africa from
1994 to 1999: A research note, Meditari Accounting Research, 8:15-23.
Dunis, C. and Reilly, D. (2004). Alternative valuation techniques for predicting UK
stock returns, Journal for Asset Management, 5(4):230-250.
Fairfield, P.M. and Harris, T.S. (1993). Price-earnings and price-to-book anomalies:
Test of an intrinsic value explanation, Contemporary Accounting Research, 9(2): 590-
611.
TF Prinsloo & M Oberholzer 205
Gallarotti, G.M. (1995). It pays to be green: The managerial incentive structure and
environmentally sound strategies, Columbia Journal of World Business, 30(4):38-57.
Global Reporting Initiatives (GRI). (2002). Sustainable Reporting Guidelines: 2nd
Generation Guidelines, Amsterdam, [On-line]. Available from:
http://www.globalreporting.org. (Accessed on 20 May 2009).
Hart, S. (1997). Beyond greening: Strategies for a sustainable world, Harvard Business
Review, 75(1):66-76.
Hart, S. and Ahuja, G. (1996). Does it pay to be green? An empirical examination of the
relationship between emission reduction and firm performance, Business Strategy and
the Environment, 5(1):30-37.
International Plant Protection Convention (IPPC). (2007). Climate change 2007:
Synthesis report: An assessment of the intergovernmental panel on climate change,
Valencia, Spain, [On-line]. Available from: http://www.ipcc.ch. (Accessed on 5 Jan.
2010).
International Union for Conservation of Nature (IUCN). (2006). The future of
sustainability: Re-thinking environment and development in the twenty-first century,
Report of the IUCN Renowned Thinkers Meeting, 29-31 Jan. 2006.
Jaggi, B. and Freedman, M. (1992). An examination of the impact of pollution
performance on economic and market performance: Pulp and paper firms, Journal of
Business Finance and Accounting, 19(5):697-713.
Kiker, G.A. (2000). Synthesis report for the vulnerability and adaptation assessment
section: South African country study on climate change, Pretoria: Department of
Environmental Affairs and Tourism.
King, A.A. and Lenox, M.J. (2001). Does it really pay to be green? An empirical study
of firm environmental and financial performance, Journal of Industrial Ecology,
5(1):105-116.
KPMG. (2001). Research and survey report on environmental accounting in South
Africa. Unpublished, Johannesburg.
Kvanli, A.H., Pavur, R.J. and Keeling, K.B. (2006). Concise managerial statistics.
Mason, OH: South-Western Thomson Learning.
Laurinkevičiūtė, A., Kinderytė, L. and Stasiškienė, Ž. (2008). Corporate decision-
making in furniture industry: Weight of EMA and a Sustainability Balanced Scorecard,
Environmental Research, Engineering and Management, 1(43):69-79.
Le Roux, C. (2011). The JSE SRI index (Special Report: Sustainability), Accountancy
SA, December 2010/ January 2011.
McGregor B.F.A. (2010). Database, [On-line]. Available from:
http://mcgregorbfa.co.za. (Accessed: 1 Jan. 2010-30 Apr. 2010).
206 SA Journal of Accounting Research Vol. 25 : No. 1 : 2011
Millard, S.P. and Neerchal, H.K. (2001). Environmental statistics with S-plus. Boa
Raton, Florida: CRC Press.
Mitchell, C.G. and Quinn, N.W. (2005). Environmental reporting disclosure in South
Africa: A comparative study of the expectations of selected group of preparers and user,
Meditari Accounting Research, 13(2):17-33.
Morrissey, S.A. and Reser, J.P. (2007). Natural disasters, climate change and mental
health considerations for rural Australia, Australian Journal for Rural Health, 15:120-
125.
Moser, S.C. (2010). Now more than ever: The need for more societally relevant
research on vulnerability and adaptation of climate change, Applied Geography, 30:646-
474.
Orlitzky, M., Schmidt, F. and Rynes, S. (2003). Corporate, social and financial
performance: A meta-analysis, Organization Studies, 24(3):403-441.
Park, Y.S. and Lee, J-J. (2003). An empirical study on the relevance of applying relative
valuation models to investment strategies in Japanese stock market, Japan and the
World Economy, 15(3):331-339.
Pava, M.I. and Krausz, J. (1995). Corporate responsibility and financial performance:
The paradox of social cost. Westport: Quorum.
Porter, M. and Van der Linde, C. (1995). Towards a new conception of environment
competitiveness relationship, Journal of Economic Perspectives, 4(4):97-119.
Rivera, J. (2001). Does it pay to be green in the developing world? Participation in a
Costa Rican voluntary environmental program and its impact on hotels’ competitive
advantage, Business Strategy & Environment, 10(3):176-182.
Roberts, L. (2011). Sustainability Intro (Special Report: Sustainability), Accountancy
SA, December 2010/ January 2011.
Roy, A. and Ghosh, S.K. (2011). The bilateral association between discretionary
environmental disclosure quality and economic performance: An Asian perspective, The
IUP Journal of Accounting Research & Audit Practices, X(21):7-27.
Shi, X.Q. (2002). Dynamic analysis and assessment for sustainable development,
Journal of Environmental Sciences, 14(1):88-94.
Sustainable Energy and Climate Change Partnership (SECCP). (2009). Climate change,
development and energy problems in South Africa: Another world is possible, [On-line].
Available from: www.eartlife.org.za. (Accessed on 13 Jan. 2010).
Telle, K. (2006). “It pays to be green” – A premature conclusion? Environmental and
Resource Economics, 35:195-220.
TF Prinsloo & M Oberholzer 207
Terre Blance, M., Durrheim, K. and Painter, D. (Editors). (2008). Research in Practice:
Applied methods for social sciences. 2nd
Edition. Cape Town: University of Cape Town
Press.
Vose, D. (1996). Quantitative risk analysis: A guide to Monte Carlo simulation
modelling. Chichester: Wiley.
Wegner, T. (2007). Applied business statistics: Methods and Excel-based applications.
2nd
Edition. Cape Town: Juta.
Wolfe, R. (1991). The use of content analysis to access corporate social responsibility,
Research in Corporate Social Responsibility and Policy, 12:281-307.
Wokutch, R.E. and Spencer, B.A. (1987). Corporate sinners and saints: The effects of
philanthropic and illegal activity of organizational performance, California
Management Review, 29:62-77.
208 SA Journal of Accounting Research Vol. 25: No. 1: 2011
PUBLICATION RECORD TO JUNE 2011
Submissions to date
In Process
- Awaiting review
- Awaiting resubmission
- Awaiting publication
1
21
0
391
22
Processed
Rejected
52%
369
191
Published 48% 178
PRIZE WINNERS
The Editorial Board awarded the following prizes for Volume 24:
G Holman, J Shev and F Zheng: Evidence of share option back-dating in the JSE
Top 40 since 2001 – R2000.
CG Mitchell and TR Hill: An exploratory analysis of stakeholders’ expectations
and perceptions of corporate social and environmental reporting in South Africa –
R2000.
SA Journal of Accounting Research Vol. 25: No. 1: 2011 209
Information for Authors Aims and Scope The SA Journal of Accounting Research (SAJAR) publishes peer-reviewed original
research papers, notes and commentaries that address issues relevant to accounting
academics and professional accountants in Southern Africa and elsewhere. This
includes areas of interest in the study and practice in financial accounting, auditing,
taxation, financial management, management accounting, finance, ethics and
information systems. Research papers should be analytical and make a contribution to
knowledge in the field. They may be empirically based (including survey and case
study methods) or review and theoretically based. Notes and commentaries should
meet all the criteria for good research, however their interest and topicality may
compensate for the research problem being less rigorously pursued. Notes and
commentaries would typically be shorter than research papers, and a faster turnaround
time is expected.
To provide a balanced presentation contributions are welcomed from the fields
mentioned above, and from related areas, such as environmental accounting, corporate
law, corporate governance, and accounting education; these fields may be approached
from a wide variety of perspectives such as the behavioural, technological, institutional,
organisational, regulatory, societal, educational, or environmental.
Manuscripts should preferably have been presented to peers, for example at
conferences and seminars, or reviewed by colleagues. Clarity of argument and
readability is an essential attribute. Authors should pay particular attention to
presenting the paper in such a way that complex matters do not unduly hinder the
readability of the paper. The judicious use of appendices is encouraged. Manuscripts
should be severely edited for jargon, style and unnecessarily complex language and
vocabulary.
Articles must be submitted and will be published in English. Should an author wish,
an extended abstract can be included in one of the other official South African
languages.
Review Procedure The editor will screen manuscripts submitted to the journal. Those considered
inappropriate or for which there appears to be a low probability of acceptance will be
returned promptly to the sender. Manuscripts that pass the initial screening will be sent
to two reviewers for evaluation using a double blind review procedure. The reviewer
will be asked to provide a recommendation to accept, revise or reject the manuscript.
Most published manuscripts have been revised one or more times before acceptance.
Awards Standard Bank sponsors three annual prizes for the best articles published in the journal;
one prize of R10 000 and two prizes of R2 000 may be awarded. No awards will take
place if the judges believe that a sufficiently high academic standard was not achieved.
All articles published will be considered in determining the awards. Articles
(co)authored by members of the board are eligible for the prize but the board member
does not receive the monetary award.
210 SA Journal of Accounting Research Vol. 25: No. 1: 2011
Manuscript Submissions Authors should submit their manuscripts by e-mail to [email protected].
Alternatively they may submit 3 hard copies to Prof Enrico Uliana, Editor, SAJAR,
Finance Department, Bremner Building, University of Cape Town, Private Bag X3,
Rondebosch, 7701, South Africa. An abstract and up to ten keywords must accompany
each submission on a separate sheet. Manuscripts must be free of any identification of
the author(s), affiliations, and acknowledgements. A separate sheet should include the
title, affiliation, contact address, phone number, fax and email address and any
appropriate acknowledgements. The letter to the editor should contain a statement that
the manuscript or a similar one has not been published and is not, nor will be, under
consideration for publication elsewhere while being reviewed by SAJAR.
The above instructions must be adhered to for resubmission.
On acceptance of the paper an electronic copy must be submitted.
Manuscript Preparation and Style Manuscripts should be laser quality printed on one side of A4 paper and be 1.5 or
double-spaced. Authors are requested to use a 12 point easily legible font, and margins
should be at least one inch to facilitate editing and duplication.
Language usage. All submissions must ensure grammatical and typographical
correctness before submission to the editor. The editor reserves the right to make minor
editorial adjustments without consulting the author.
References in the text. Cited information must be identified accurately. The
surname(s) of the author(s), year of publication and page number(s) appear in
parentheses after the quotation, for example (Coetzee, 2003:2-5), (Brown & Jones,
2003:2-5). Omit the page number(s) if the entire publication is referred to, for example
(Berger, 2001). In works by three or more authors the surnames of all the authors
should be given in the first reference to such a work, for example ‘A recent study
(Jones, Smith, Boren & White, 2002) shows … .’ In later references to this work only
the first author’s name is given, and the abbreviation et al., a comma and the year of
publication. For example: (Jones et al., 2002).
References at the end of the manuscript. More details about sources referred to in the
text must appear at the end of the manuscript under the caption ‘References’. All
sources must be arranged alphabetically according to the surnames of the first author. If
more than one publication by the same author(s) appear in one year they must be
distinguished by an a, b, etc., for example 2003a, 2003b.
References from books. After the year of publication, follows the title. The Edition,
Place of publication: publisher, total number of pages:
Steers, R.M. and Porter, L.W. (1991). Motivation and work behaviour. 5th Edition.
Singapore: McGraw-Hill.
References from journals. After the year of publication, follows the title of the article,
title of the journal, volume, number, page(s).
SA Journal of Accounting Research Vol. 25: No. 1: 2011 211
Doyle, R.J. (1983). Gainsharing – A total productivity approach, Journal of
Contemporary Business, 11(2):57-70.
Proof-reading Final proofs of the paper will be sent to the author for final proof-reading. It is in the
author’s interest and is his/her responsibility to ensure that the paper is error-free.