Skills development is a fundamental necessity in developing countries, as it is central to capability enhancement and industrialisation, both of which ultimately increase national prosperity (Edigheji 2010). South Africa continues to experience a pervasive skills shortage, both as a legacy of apartheid and as a result of ongoing dysfunctionality in its education system (Akoojee 2010). This skills shortage continues to complicate the South African labour market, where most public sector employees lack appointment-appropriate experience (Koelble & Siddle 2014), necessitating the use of private-sector consultants (Wenzel 2007).

The skills shortage and transformation challenges are closely interrelated and reverberate throughout the South African accounting profession, where the processes of closure on the basis of race and class are still prevalent (Barac 2015; Hammond, Clayton & Arnold 2009; 2012). There is a shortage of professional accountants and auditors in the South African public sector, which emphasises the skills training needs of the public sector (World Bank 2013). Despite these challenges, South Africa has maintained its number one ranking for the strength of its Auditing and Reporting Standards for the sixth successive year according to the World Economic Forum’s (WEF) Global Competitiveness Index for 2015–2016 (WEF 2015b).

Drawing on social closure theory and the above-mentioned paradox creates an interesting background for this study, which presents a model that is able to guide public sector auditors in the exercise of planning judgement for a class of transactions, account balance and/or disclosure. Such a model [referred to as an integrated evidence planning model (IEPM)] allows for an audit structure that is ‘a mechanistic approach to decision making that constrains the range of actions available to individual auditors in specific circumstances’ (Knechel 2007:386). By following a structured audit approach, auditors try to reduce the risk of serious errors in judgement within the audit process (Knechel 2000) and thereby reduce the uncertainties and ambiguities surrounding the performance of the responsibilities of inexperienced or unskilled professionals (Knechel 2007). Although the literature review identifies that today’s auditors are resisting formalism and structure in favour of a business risk audit approach (Abdullatif & Al-Khadash 2010; Khalifa et al. 2007; Knechel 2007), such a structured approach could aid public sector auditors in South Africa to sidestep both their own skills limitations and those of their auditees (World Bank 2013) when exercising judgement during their audit planning decision-making. The same applies in other developing countries where public sector auditors (also referred to as Supreme Audit Institutions [SAIs]) are faced with endemic limitations in skills which inhibit their performance of the tasks expected of a modern auditor (UN 2007). Sound auditor judgement is fundamental to the strengthening of public sector transparency and accountability (Iyoha & Oyerinde 2010; Kayrak 2008) in developing countries as these countries are often plagued by corruption (ACFE 2014; Desta 2006; Pillay 2004; Transparency International 2014). It has become axiomatic that the sustainable economic development of these countries is impacted by weak accounting and auditing capabilities, and the consequential high incidence of financial mismanagement in public sectors (Iyoha & Oyerinde 2010).

An auditor should obtain sufficient appropriate audit evidence to reduce audit risk to an acceptably low level; this will then enable him to draw reasonable conclusions on which the audit opinion can be based (Causholli & Knechel 2012; Hayes 2007; Perry 2011; Schmutte & Duncan 2009). Thus, the assessment of the risks of material misstatement informs the audit evidence planning decisions (Blay, Kizirian & Sneather 2008; Fogarty, Graham & Schubert 2007; IAASB 2015c; 2015e). Despite the importance of audit evidence planning considerations, previous studies have not shown a clear relationship between evidence planning and the extent of testing on one hand, and the level of or changes in risks on the other hand (De Martinis, Fukukawa & Mock 2011; Elder & Allen 2003; Fukukawa, Mock & Wright 2006; Hogan & Wilkins 2008; Johnstone & Bedard 2001; Luo 2011; Seidel 2014); nor have they addressed the question of how to combine or aggregate evidence from different sources (Knechel 2007; Srinidhi & Vasarhelyi 1986; Turley & Cooper 1991) in order to reach a final audit opinion. Although the International Standards on Auditing (ISAs) provide some guidance to assist the auditor in deciding the sufficiency and appropriateness of audit evidence to be obtained, they do not dictate exactly how much evidence needs to be gathered, nor the quality thereof (Perry 2011). The basis for determining ‘sufficient appropriate evidence’ is therefore determined by the professional judgement of the auditor, which may mean that there is little consistency regarding the quantity and quality of the audit evidence obtained (Perry 2011), especially where auditors are inexperienced or lack the necessary skills. Recent research indicates that audit quality is not improving in areas where the auditor needs to apply judgement (Kleinman, Lin & Palomon 2014). Consequently, there is a need to develop and communicate overarching principles to guide auditors making these decisions. This is in line with the aforementioned objective of this study.

The remainder of the article is organised as follows. In the next section, social closure theory, as the theoretical underpinning of this study, is explained. This is followed by an explanation of the research method and the context of the study. The building blocks of the model are explained next, with particular reference to the literature on the exercising of judgement by the auditor when planning the audit. Then the suggested model, intended to guide public sector auditors when exercising planning judgement, is presented. In the final section, a conclusion is reached and areas for future research are identified.

Social closure theory, based on the work of Weber, is used to analyse social stratification processes (Murphy 1984). This theory posits that one group monopolises advantages by closing off opportunities to another group, deemed to be ‘outsiders’ (Malsch, Gendron & Grazzini 2011:201). The apartheid laws in South Africa serve as an example of the effective application (and institutionalisation) of the principle of exclusion (Murphy 1984). Even though statutory apartheid was abolished in 1994, there is still evidence of ‘widespread de facto segregation’ in the post-apartheid era (Dixon et al. 2010:405). This manifests as workplace inequality and discrimination, is based on race and results in professional closure (Dixon et al. 2010). The situation reflects the imbalances within the South African education system (Mdepa & Tshiwula 2012). The legacy and echoes of apartheid are thus still reflected in South Africa’s ongoing skill shortages (Akinyemi 2010). Exclusion based on skill shortages is a product of the entire education sector that is still undergoing transformation. South Africa’s past continues to hinder certain students from accessing and participating in the higher education sector, and many students arrive at higher education institutions underprepared for their studies (Mdepa & Tshiwula 2012; Pym & Kapp 2013; Sartorius & Sartorius 2013). An assessment of the quality of mathematics and science education in South African schools, published in the Global Information Technology Report 2015, ranks South Africa in last position (mathematics) and close-to-last (science) (respectively 139th and 143rd of the 143 countries surveyed): when looking at the overall quality of South Africa’s education system, this serves as a clear indication of the magnitude of this problem (WEF 2015a).

Professional closure is a particular genre of social closure (Chua & Poullaos 1993; Hammond et al. 2012; Kim 2004; Lee 2010; Macdonald 1984; Verhoef 2013) which was initially based on class, gender, race and social differentiations (Annisette 2003; Cooper & Robson 2006). More recently, professional closure has come to embody exclusion based on education and professional credentials (Annisette 1999; Bonnin & Ruggunan 2013). Through accredited education programmes, stringent examinations, prescribed practical training and the registration or licensing of those then deemed worthy of being called professional practitioners, closure or exclusion continues to be held in place in today’s South African accounting profession (Heathcote 2012). Indications are that, even in post-apartheid South Africa, the professional closure of the South African accounting profession continues because of difficulties experienced by previously disadvantaged persons to acquire the academic education (South Africa’s education system remains largely dysfunctional, as previously mentioned) and professional credentials necessary to enter the profession (Barac 2015). The shortage of professional accountants and auditors is especially apparent in the South African public sector (World Bank 2013). This is reflected in the membership statistics of the South African Institute of Chartered Accountants (SAICA): of its 32 079 members, only 1415 (4%) members work in the public sector, and of the 10 189 members practicing as auditors, only 584 (6%) are in the public sector (SAICA 2016).

This study employs a theory-building approach (Strauss & Corbin 2008) to develop a model capable of guiding public sector auditors in the exercise of their planning judgement decisions for a class of transactions, account balance and/or disclosure. This study critically analyses and interprets the various relevant principles and concepts emerging from the literature (Leedy & Ormrod 2014) and then systematically links and integrates them in successive stages through the application of inductive reasoning, deliberate thinking and logical interpreting to construct the IEPM that quantifies (explains) the extent of audit evidence required to support a desired outcome.

The model reported on in this article was developed for incorporation into the audit methodology used by the Auditor-General of South Africa (AGSA) (Mentz 2014). In terms of the Constitution of the Republic of South Africa (Act No. 108 of 1996), the AGSA acts as the external auditor of all national and provincial state departments, public entities, municipalities and any other institutions required by legislation to be audited by the AGSA (RSA 1996). The AGSA has to submit audit reports to any legislature that has a direct interest in the audit outcome, and to any other authority prescribed by national legislation; in addition, all reports must be made public (RSA 1996). The functions of the AGSA are described in section 188 of the Constitution; it is furthermore subject to the requirements of the Public Audit Act (Act No. 25 of 2004) (PAA), which mandates the AGSA to perform constitutionally required audits and other functions (RSA 2004).

The AGSA’s staff complement, including trainee auditors and short-term contractors, was 3435 on 31 March 2015. Their focus is mainly on the performance of financial audits, performance audits, audits of performance information and audits of compliance with legislation (AGSA 2015a). Even though the AGSA experiences a scarcity of skills in some areas, by recruiting 1126 employees in 2014–2015 (33% of the full staff complement is thus relatively inexperienced) it manages to ensure that the institution’s staffing levels are adequate for effective performance of its statutory and other duties (AGSA 2015a). The organisation has, over the past years, managed to significantly increase the number of qualified audit professionals that make up its staff complement (in 2015, 547 staff members were chartered accountants [the SAICA website on 31 March 2016 reported this as 584 chartered accountants (SAICA 2016)]). Thus, 16% of the staff could practice as private sector auditors (AGSA 2015a).

In terms of the PAA, the AGSA determines the standards and methodology applied in audits of public sector entities (RSA 2004). The AGSA is one of a few SAIs in the world that has fully implemented the ISAs (AGSA 2015a). Although having no legal oversight role over the AGSA, the Independent Regulatory Board for Auditors (IRBA) in South Africa reviews and provides credibility to the internal quality control monitoring process the AGSA uses (this is also performed at the individual private sector firm level) (World Bank 2013). That the IEPM adheres to the ISA’s requirements is confirmed by the fact that the model has been incorporated into the audit methodology of the AGSA, and this methodology has been successfully assessed for compliance with the ISAs through a monitoring inspection performed by the IRBA (Mentz 2014).

The World Bank (2013) reported significant skills shortages in the South African public sector (auditees of the AGSA), which also manifests as a shortage of skills available to support financial management roles in public sector entities. Internal capacity constraints and high vacancy rates sometimes lead public sector entities to appoint consultants to support the continuity of service delivery, a practice which the World Bank (2013) considers a main cause of consistently poor audit outcomes in the three tiers of government. For example, 17% of the 325 auditees at local government level (268 municipalities and 57 municipal entities) received unqualified audit reports with no findings in the 2013 and 2014 financial year. For the 2012 and 2013 financial year, this percentage amounted to 9% and 5% for the 2011 and 2012 financial year (AGSA 2015b). Even though since 2011 and 2012 the number of municipalities and municipal entities achieving what is commonly called a ‘clean audit’ has more than tripled, and a quarter of all public departments and entities have attained a clean audit status, these audit outcomes remain areas of concern (World Bank 2013).

The model quantitatively integrates the determination of the extent of audit evidence to be obtained with the assessed risks of material misstatement and the materiality of each assertion, class of transactions, account balance or disclosure. The risk level of each assertion, class of transactions, account balance or disclosure is identified as ‘significant’, ‘standard’ or ‘slight’ (first and second building blocks) and a corresponding quantitative level of assurance is allocated (fourth building block). Based on the identified level of assurance, the model quantitatively links the extent of audit evidence required to the sample sizes needed for performance of tests of controls and tests of details, and the precision levels needed for the performance of substantive analytical procedures (sixth and seventh building blocks). By establishing this inter-relationship between the risks of material misstatement and the assurance that is required to reduce audit risk to an acceptable low level, the IEPM increases the audit work in high-risk areas. This effectively addresses the situation identified in earlier studies that found no clear correlation between evidence planning and extent of testing and the level of or changes in risks (De Martinis et al. 2011; Elder & Allen 2003; Fukukawa et al. 2006; Hogan & Wilkins 2008; Johnstone & Bedard 2001; Luo 2011; Seidel 2014).

In addition to incorporating the different types and combinations of audit procedures needed to obtain audit evidence in response to the assessed risks (third building block) and establishing the compensatory inter-relationship between types of audit procedures, the IEPM also quantitatively integrates the relative assurance being drawn from the different types of audit procedures (fifth building block). Thereby, the IEPM clarifies how to combine or aggregate evidence from different types of procedures in order to reach a final audit opinion. By following the IEPM, public sector auditors obtain guidance, enabling them to determine the extent of audit evidence required [which is linked to sample sizes (for tests of controls and tests of details) and precision levels (for analytical procedures)], in relation to the assessed risks of material misstatement. It aids the public sector auditor who is required to respond to the question of how much to test, and in so doing responds to numerous conclusions in scholarly literature that there is an increasing need for guidance in how to determine the extent of testing (Bierstaker, Houston & Wright 2006; Charles, Glover & Sharp 2010; CICA 1980; Fogarty & Rigsby 2010; Hayes 2007; Kinney 2005; Seidel 2014). In addition, the IEPM answers the question that arose from the literature review: how to combine or aggregate evidence from different types of procedures (tests of controls, substantive analytical procedures and tests of details) (Knechel 2007; Srinidhi & Vasarhelyi 1986; Turley & Cooper 1991) in order to reach a final audit opinion.

Thus, the IEPM provides a framework to guide the auditor’s exercise of his judgement regarding audit evidence gathering. However, it is not a substitute for professional judgement. Rather, the model provides a framework within which the auditor may confidently approach the exercise of judgement (an area where ‘flexibility’ predominates) regarding the types, extent and combinations of audit procedures to be employed to gather sufficient appropriate audit evidence to respond to the assessed risks. Although audit decision-making cannot be reduced to the mechanical application of a quantitative model, and will always require the application of considerable (human auditor) judgement, studies have suggested that decision-making models that aid the auditor when required to exercise judgement in complex, multi-dimensional evidence planning situations may improve audit efficiency and effectiveness (Bedard & Graham 2002; Curtis & Turley 2007; Jeppesen 2007; Knechel 2007; Nelson, Tan & Trotman 2005). The IEPM responds to this need. Furthermore, the model, which has now been incorporated into the audit methodology of the AGSA, is also a practical tool to manage adherence to the ISAs (Mentz 2014). This has been confirmed by the IRBA’s assessment of the model against established assessment criteria and standards, during the IRBA’s review of the AGSA’s audit methodology (Mentz 2014).

As is evident from the above discussion, the IEPM clarifies the audit evidence decision-making sequences, inter-relationships and contingent dependencies of the different audit procedures. Furthermore, it quantifies the compensatory inter-relationships between the types of audit procedures and the desired levels of overall assurance required in response to the assessed risks of material misstatement. Such a direction is of value to inexperienced auditors in the South African public sector, and this structured approach could reduce their uncertainty about what course of action is appropriate, and the ambiguity inherent in their implementation of the audit plan (Knechel 2007). In this way the IEPM supports professional accountants and auditors with limited skills and experience in the public sector (World Bank 2013). In so doing (improving skills and competence levels), the professional closure experienced as a consequence of apartheid (Murphy 1984) is countered.

The concepts underpinning the IEPM are those of a traditional audit risk approach. This is in contrast with the business risk approach to auditing followed by large audit firms (Fogarty & Rigsby 2010). In addition to guiding inexperienced auditors (as explained above), a traditional audit risk approach (as incorporated in the IEPM) fits well within the public sector audit environment. Firstly, the AGSA does not blend consulting and auditing (nor do other SAIs). This blending of consulting and auditing is a prominent driver of the current business risk audit approach preferred by large audit firms (Fogarty & Rigsby 2010). Secondly, in terms of the IEPM, the extent of audit evidence gathered and the combinations of audit tests are still determined by the auditor and guided by professional standards; these are not ‘managed’ so as to produce higher levels of satisfaction for the client, as is the case in business risk auditing (Humphrey & Moizer 1990). This is an important consideration, because reaction to the social closure legacy of apartheid has resulted in a public sector environment characterised by a tendency to rate racial representation and political loyalty higher than integrity, skills and qualifications, and experience (Koelble & Siddle 2014; Wenzel 2007). The resulting skills shortage has led to poor audit results and a perilous financial state (Ngoepe & Ngulube 2014). In such an environment, a traditional audit risk approach, with its focus on audit risk, the financial statements and internal controls, rather than on the auditee’s strategic direction (Knechel 2007), ensures an audit approach that searches for those factors that might lead to a misstatement of an account balance, class of transactions or disclosure (Fogarty & Rigsby 2010). Finally, the IEPM makes use of materiality levels, evaluation of internal controls and audit procedure combinations, while planning for business risk auditing might not lead to a similar responsiveness to proportionate changes in levels of substantive testing and follow-up procedures (Fogarty & Rigsby 2010).

This article presents seven building blocks from which a model has been constructed to guide public sector auditors in the exercise of professional judgement in planning decisions for classes of transactions, account balances and/or disclosures. The IEPM clarifies the audit evidence decision sequences and the inter-relationships and contingent dependencies of the different audit procedures, and quantifies the compensatory requirements of the inter-relationships between the types of audit procedures and the desired overall levels of assurance in response to the assessed risks of material misstatement. Although the model is aligned with the current ISAs, it relates better to the traditional audit risk approach than to the currently preferred business risk audit approach of large audit firms (Fogarty & Rigsby 2010). The IEPM has been included in the audit methodology of the AGSA, and its compliance with the ISA requirements has been assessed through a monitoring inspection performed by the IRBA (Mentz 2014).

When following the IEPM, South African public sector auditors employ a more traditional audit risk approach, in preference to the business risk audit approach used by large audit firms. It is also in contrast to present-day private sector auditors’ resistance to formalism and structure (Abdullatif & Al-Khadash 2010; Khalifa et al. 2007; Knechel 2007). Such a structured approach fits well in the public sector environment, effectively addressing its shortage of experienced professional accountants and auditors (World Bank 2013) (an apartheid legacy). Social closure [institutionalised by the apartheid laws, and a principal form of exclusion (Murphy 1984)], has resulted in a South African public sector that still has significant skills shortages (Koelble & Siddle 2014; Wenzel 2007) and this continues to result in poor audit results and an ineffectively managed financial state (Ngoepe & Ngulube 2014). This situation is also faced by SAIs in other developing countries where there are limited skills. This inhibits the performance of the tasks expected of a modern (public sector) auditor (UN 2007) and audit function, as is the case of the AGSA. The outcome is a corruption-stricken public sector environment that is shy of transparency and accountability (Iyoha & Oyerinde 2010; Kayrak 2008).

The study has acknowledged limitations. The focus of the IEPM developed in this study is at the class of transactions, account balance, disclosure and assertion levels, and not at the overall financial statement level. The model does not consider the business risk assessment (the backbone of business risk auditing) or the evaluation and the reporting phases of the audit, and it is set at the point in the audit sequence after the assessment of risk at the relevant class of transactions, account balance, disclosure and assertion level has been completed. It is specifically concerned with quantifying the extent or sufficiency of audit evidence required to enable the public sector auditor to express an opinion on the financial statements, and it limits the consideration of the nature and timing of audit procedures only to the degree that they impact the auditor’s decisions regarding the extent of audit evidence.

Future studies should consider developing a business risk audit approach for use by public sector auditors that makes use of significantly detailed risk assessments at all levels of inquiry (including corruption and political influences at each level) and that shows clear linkages between audit procedures and risk assessments. These studies should also consider how inexperienced auditors and auditors with limited skills could be developed to follow the revised business risk audit approach. Another area of consideration is the impact of such an approach on the mandate of SAIs, particularly because past results have shown that regulatory changes (such as the Sarbanes-Oxley Act in the USA) that required improved regulatory scrutiny have caused auditors to act with greater circumspection while following a business risk audit approach (Fogarty & Rigsby 2010). This line of research would require renewed consideration of the relations between SAIs, their states and regulatory agencies and the users of public sector financial information. In this regard, a future study should investigate the logic of professionalism among public sector auditors as it will determine how public sector auditors ought to act, how SAIs should be organised and how professional norms are best enforced (Suddaby, Gendron & Lam 2009).