Portfolio rebalancing in South Africa

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.

mailto:[email protected]

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

[email protected]

[email protected]

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.


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


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.


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


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


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).


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.


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


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.


(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?”


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.


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


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


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.


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.

http://www.ifac.org/

http://www.irba.co.za/index.php?option=com_docman&task=doc...35


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.


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.

http://www.uct.ac.za/downloads/uct.ac.za/about/policies/academicstaff.pdf

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


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

[email protected]

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,


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:


“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.


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.


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.


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


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


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.


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.


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.


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.


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

[email protected]

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,

mailto:[email protected]


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


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


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


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,


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


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


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


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


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.


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.


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-


adjusted return. The output is presented in Table 4 and labelling follows the


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


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


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


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).


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


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


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


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.

http://dbpaustin.com/pwm/TrumanClarkEfficientRebalancing.pdf


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.


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


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


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


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

[email protected]

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


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.


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.


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.


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


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.


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).


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


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.


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


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.


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


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


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


Trafford, V. and Leshem, S. (2008). Stepping Stones to Achieving your Doctorate.

Berkshire: McGraw-Hill.


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

[email protected]

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 &


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).


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


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


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


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.


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


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”.

http://www.sustainabilitysa.org/LinkClick.aspx?fileticket=QIng8AK7SYU%3d&tabid=79

http://www.sustainabilitysa.org/LinkClick.aspx?fileticket=3HtLz9AEtOo%3d&tabid=79


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).


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


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


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


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.


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


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



public sector entities disclose sufficient information regarding their stakeholder

engagement efforts.


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.


Disclosure of GRI content and use of GRI guidelines



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.


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



public sector entities included performance data of their organisations with reference to

integrated, economic, environmental and social performance indicators.


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%)


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.


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

http://www.accountability.org/about-us/publications/aa1000.html

http://www.accountingforsustainability.org/output/page146.asp


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


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.

https://www.cdproject.net/en-US/WhatWeDo/Pages/overview.aspx

http://www.cisco.com/web/strategy/docs/education/schl-green-wp.pdf


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.


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


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


Holcroft, N.M. (1999). Corporate environmental reporting: a review of national and

international trends in structures and systems necessary for the production of effective

http://www.hm-teasury.gov.ulk/d/FRAB(64)03-SUSTAINABILITY-REPORTING.pdf

http://www.paulgosling.net/2010/09/sustaianibilty-in-the-public-sector-accounting-business

http://www.paulgosling.net/2010/09/sustaianibilty-in-the-public-sector-accounting-business

http://globalblogs.deloitte.com/deloitteperspectives/2010/07/what-is-corporate-responssibili

http://globalblogs.deloitte.com/deloitteperspectives/2010/07/what-is-corporate-responssibili


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


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).

http://www.iodsa.co.za/membership.asp?CatID=145

http://www.sustainabilitysa.org/

http://www.iso.org/iso/iso_catalogue/management_and_leadership_standards.htm

http://www.asci.org.in/publications/journal1998.asp

http://wwww.edie.net/news/news-story.asp?id=9401


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.

http://sd.defra.gov.uk/2010/07/public-sector-leadership-on-sustainability


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


Prizma. (2011). GRI Sustainability Reporting in the Public Sector. Available from:

http://prizmablog.com/2010/02/13/gri-sustainability-reporting-in-the-public-sector


http://0-onlinelibrary.wiley.com.ujlink.uj.ac.za/doi/10.1002/bse.657/pdf

http://ujdigispace.uj.ac.za:8080/dspace/handle/10210/2030

http://www.oecd.org/dataoecd/56/36/1922428.pdf

http://www.pwc.com/gx/en/sustainability/industries/goverment-and-public-sector.jhtml

http://prizmablog.com/2010/02/13/gri-sustainability-reporting-in-the-public-sector


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).

http://www.bptawards.com/post_awards_brochure_sustainability_ps_09.pdf

http://www.socialinvest.org/resources/research/documents/SIF-SIRAN-KLDReportforEMTransparency2008.pdf

http://www.socialinvest.org/resources/research/documents/SIF-SIRAN-KLDReportforEMTransparency2008.pdf

http://www.sustainabilitysa.org/sustainabilityreporting/Wowwellarecompaniesdoingrepo

http://www.sustainabilitysa.org/sustainabilityreporting/Wowwellarecompaniesdoingrepo

http://www.sustainabilitysa.org/sustainabilityreporting/StandardsandGuidelines.aspx


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


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.

http://www.unep.fr/scp/gri/pdf/SectorMeet.pdf

http://www.unglobalcompact.org/docs/news_events/8.1/GC_Summit_Report_07.pdf

http://www.sustainability.com/programs/engaging/history-reporting.asp


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

[email protected]

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.


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)


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.


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.


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


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:


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:


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.


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


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).


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).


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).


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.


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.

http://schwert.ssb.rochester.edu/f533/ff_jf_96.pdf

http://schwert.ssb.rochester.edu/f533/ff_jf_96.pdf

http://www.afajof.org/jofihome.shtml

http://www.topcompanies.co.za/topcos2003/stories/tc1.htm


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


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

[email protected]


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.


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.

http://en.wikipedia.org/wiki/Gender

http://en.wikipedia.org/wiki/Female

http://en.wikipedia.org/wiki/Male

http://en.wikipedia.org/wiki/Biological_sex


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


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.


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


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.


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


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


97%

3%

no yes

Have children

0

100

200

300

400

500

600

700

800


93%

7%

no yes

Have children


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


Nu

mb

er

of

respo

nd

en

ts


66%

34%

Passed Failed

Exam success

0

100

200

300

400

500

600


66%

34%

Passed Failed

Exam success


53% 47%

Passed Failed

Exam success

0

100

200

300

400

500

600

66%

34%

66%

34%

53% 47%

Af rikaans English

Other


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


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


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


48% 52%

Passed Failed

Exam success


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


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


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


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


52% 48%

Passed Failed

Exam success


41% 59%

Passed Failed

Exam success


38% 62%

Passed Failed

Exam success

0

100

200

300

400

500

600

700

800


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.


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


Nu

mb

er

of

respo

nd

en

ts


60%

17%9% 8% 5%

0 1 2 3 4

Past QE 1 attempts

0

100

200

300

400

500

600


64%

19%

7% 4% 5%

0 1 2 3 4

Past QE 1 attempts


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


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


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


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


33%

61%

6%

UCT (Cape Town) UJ (Gauteng) Other

Which Board Course?

020406080

100120140160180200220240


44%

53%

3%


Which Board Course?


31%

62%

7%


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


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


0

10

20

30

40

50

60

70

80

9055%

52%

42%38%

17%15%

12%8%


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


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%


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:


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.


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.


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.

http://www.sprc.unsw.edu.au/media/file/DP143.pdf.%20Accessed%207%20April%202011

http://www.ey.com/ZA/en/Careers/%20Board-courses

http://www.ey.com/ZA/en/Careers/%20Board-courses


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.

http://www.allbusiness.com/human-resources/employee-development/752311-1.html.%20Accessed%2011%20May%202010

http://www.allbusiness.com/human-resources/employee-development/752311-1.html.%20Accessed%2011%20May%202010

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


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.

http://www.hsrcpress.ac.za/

http://www.pwc.com/za/en/campus-candidates/study-support.jhtml

http://www.pwc.com/za/en/campus-candidates/study-support.jhtml

https://www.saica.co.za/About/tabid/56/language/en-ZA/Default

https://www.saica.co.za/Trainees/BecomingaCA/tabid/157/language/

https://www.saica.co.za/LearnersStudents%20/QEI2011

https://www.saica.co.za/LearnersStudents%20/QEI2011

http://www.surveysystem.com/sdesign.htm


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.

http://en.wikipedia.org/wiki/Gender_identity

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

[email protected]

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


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:


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.


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


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%


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


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


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


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

http://www.herdsa.org.au/wp-content/uploads/conference/2008/media/Nettleton.pdf

http://www.herdsa.org.au/wp-content/uploads/conference/2008/media/Nettleton.pdf

https://www.saica.co.za/Portals/0/documents/FINAL_-_Detailed_Guidance_for_the_%20academic_programme_(Nov_2009).pdf

https://www.saica.co.za/Portals/0/documents/FINAL_-_Detailed_Guidance_for_the_%20academic_programme_(Nov_2009).pdf

http://www.saica.co.za/Portals/0/documents/SAICA%20Submission%25

http://www.saica.co.za/Portals/0/documents/SAICA%20Submission%25


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


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

[email protected]


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


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).


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).


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.


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)

http://www.nbi.org.za/


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. =

∑


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)


44 000

40 480

39 000

33 930

40 000

39 200

37 000

41 810

C: Production (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).


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


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.


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


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.


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


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.

http://www.cdproject.net/


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).

http://www.ipcc.ch/

http://mcgregorbfa.co.za/


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.

http://www.eartlife.org.za/


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.

Date post:	22-Feb-2023
Category:	Documents
Upload:	imf
View:	0 times
Download:	0 times

Portfolio rebalancing in South Africa

Documents