Auditors maximising their utility: Economic analysis of the supreme audit institution

. Despite several studies on the supreme audit institutions (SAIs), the research question of what maximises SAI auditors' utility remains unanswered. Since SAIs are key public suppliers of audit services, it is essential to determine how they perform their duties and obligations to the stakeholders. This paper follows the grounded theory paradigm using qualitative and quantitative methods. This study is pioneering in many ways because the motivation of SAI's auditors is rarely the subject of research. We formulate the model of SAI's utility and determine the SAI auditor's motivators. The study results prove that SAIs seek to impose their system of preferences through audits and derive usefulness from this. From a long-term perspective, this may lead to a negative autonomy of SAI. Additionally, the study reveals that auditors strive to maximise monetary income and perceive employment stability as a motivational factor. However, job stability could lead to lower auditors' engagement during the audits. The study proposes a pay-for-performance system, which does not reduce the auditor's motivation linked with job stability and simultaneously encourages auditors This study contributes to a better understanding of the problems associated with interactions of public officials inside the state.


INTRODUCTION
Although there are studies on supreme audit institutions (SAIs) and their activities (c.f., Mosher, 1979;Pollitt et al., 1999;Blume & Voigt, 2011;Jeppesen et al., 2017;Dobrowolski, 2017;Cordery & Hay, 2019), little is known about the problems with the motivation of SAIs' auditors. Auditors' motivation is generally thought to be positively associated with performance on various tasks. However, little experimental evidence supports this conclusion, and little is known about the specific mechanisms behind any effect (Kadous & Zhou, 2019). Meanwhile, regardless of how domestic auditing systems are designed, SAIs play a vital role as the primary evaluator of public performance in each country. Therefore, one may assume that understanding the SAI auditors' motivation is crucial to assessing whether the SAI properly fulfils its public task. We formulate the following research questions: 1) How does SAI maximise its utility in the state? 2) What motivates auditors to work in SAI?
The paper proceeds as follows. First, we review previous research on SAIs and auditors' motivations. After that, we present the methodology and show the empirical account of the motivation system in SAI.
We develop a theory about auditors' professional motivation and a mechanism that increases SAI's utility for stakeholders. We build a model and experimentally test whether SAI increases its societal utility through increasing audit demand. Finally, we present conclusions and opportunities for further research.

SAI's unique role and position in the macrostructure
SAIs basically are not part of the judiciary and the executive branch of the state. Finally, SAIs are not part of the legislative branch of the state, although they are subject to parliaments (Kożuch & Dobrowolski, 2014;European Court of Auditors, 2019). Therefore, the SAIs are outside Montesquieu's classical tripartite division, which affects their audit powers. SAI is subordinate to the Parliament, but regarding the method of conducting audits and presenting audit findings, it operates based on the law and within the law, and no entity may influence the audit findings. The only such impact occurs in the case when the auditee submits justified objections to the findings, conclusions, and audit recommendations (Dobrowolski, 2017).
How the SAI carries out its audit derives from the SAI Law enacted by the Parliament and its particular position in the macrostructure -state. SAI has the broadest audit rights. This organisation establishes the facts based on audit evidence, determines the cause and effects of irregularities or best practices, and makes proposals (recommendations) to improve auditees' activities. In most cases (except SAIs from Portugal and Greece -they are part of the judiciary power of the state), SAIs are outside the tripartite division of power, and they do not have the right to order the realisation of audit recommendations by the audited organisations. Therefore, the implementation of audit recommendations by audited organisations is optional. In other words, the audited organisations decide on the scope and method of implementing audit recommendations. On the other hand, the provisions of the law on SAIs oblige audited organisations to inform SAIs about the method of implementing audit recommendations or the reasons for refusing to implement them. It can therefore be generalised that SAIs are not supervisory authorities. SAIs carry out financial and compliance audits and, for the most part, performance audits. Their audits aim to assess whether the activities of the audited organisations were legal, reliable, and in line with the goals set by decision-makers. In the case of performance audits, SAIs declare that they assess whether the activities of the audited organisations were economical, effective, and efficient (Kożuch & Dobrowolski, 2014;European Court of Auditors, 2019;Tamimi & Orbán, 2022). However, researchers (Pollitt et al., 2002) state that SAIs' declarations about performance audits differ from reality and that SAIs have problems assessing the effectiveness of public task implementation. It should also be noted that the SAIs cannot question the political goals (Dobrowolski, 2017).
SAIs evaluate the activities of state organisations as well as entities that benefit from state funds. In some countries, for example, Poland, SAI also assesses the activities of local governments and entities using their funds. However, the statutory criteria for assessing such activity are limited. SAI cannot assess the purposefulness of actions undertaken by local government. It results from the adopted system paradigmlocal government participates in exercising public authority) (Kożuch & Dobrowolski, 2014;European Court of Auditors, 2019).
SAIs prepare audit reports in which they present the findings and the resulting assessments of the audited activity. These findings are based on evidence in the working papers. Audited organisations may, prior to the publication of reports, raise objections to SAI's findings, conclusions and audit recommendations. SAIs consider objections the audited organisations raise, and their position is presented in the SAI's audit report. These reports are submitted to audited organisations and parliaments (relevant parliamentary committees) and may be available to the public if the SAI law provides and does not contain classified information (Kożuch & Dobrowolski, 2014;European Court of Auditors, 2019).

Motivation factors
Organisations use a range of tools to plan, monitor, evaluate, and reward the job performance of their employees (c.f. Nicholson, 1977;Murphy & Cleveland, 1995;& Byrne, 2001;Pulakos & O'Leary, 2011;Aguinis, 2013;Pulakos et al., 2015;Murphy, 2020). One may analyse these motivation tools from different perspectives, including employee performance appraisal (Kuvaas, 2006;Murphy, 2020) and the connection between human resource management practices and behavioural outcomes (Gould-Williams, 2007;Alfes et al., 2012), the role of salary in the motivation of employees (Rynes et al., 2004). One may also analyse the motivation problem from individual employees' and superiors' perspectives. It can also be analysed from the perspective of the entire organisation, assuming that it is an aggregate set of individual preferences, in some cases divergent (e.g., Benson, 1995), but in many cases convergent preferences which transform into the directions of the organisation's activities and its outcomes.
Auditors' motivation is analysed from their impact on financial reporting or the determinants of individual performance in professional accounting firms (Ferris & Larcker, 1983;Codori, 1988;Christensen et al., 2012;Iskandar et al., 2012). Auditors' motivation is generally thought to be positively associated with performance on various tasks. However, there is only a little experimental evidence supporting this conclusion, and one knows little about the specific mechanisms behind any effect (Kadous & Zhou, 2019).
Prior empirical studies demonstrate that performance incentives cause individual auditors to increase the amount of effort devoted to the audit task (Mohd-Sanusi & Iskandar, 2007). Iskandar et al. (2012) pointed out that a non-financial performance incentive increases auditors' efforts. They tried to explain two incentive mechanisms: accountability pressure as an external mechanism and self-efficacy as an internal mechanism, to improve audit judgment performance. They showed that each incentive, accountability pressure and self-efficacy, are significantly and directly related to audit judgment. The results are in line with Libby and Lipe's (1992) studies and Bonner and Sprinkle's (2002) studies. Srimindarti et al. (2020) noticed that an auditor's job satisfaction is addressed under two major headings, which are internal and external factors. The internal factors include achievement, advancement, recognition and growth of knowledge and skills. The external factors include but are not limited to organizational policy, relationship with colleagues, workplace security, relationship with superiors, and salary conditions (Tan & Waheed, 2011). Malka and Chatman (2003) found that income and job satisfaction correlated positively. Achievement is known to be a significant satisfaction. Achievement refers to employees' willingness to solve problem efficiently (Wernimont, 1966;Ruthankoon & Ogunlana, 2003). Work success enhances an employee's sense of accomplishment and self-confidence (Savery, 1996;Saat et al., 2021). Furthermore, the possibility of promotion or an improvement in work status is one of the variables that can determine job satisfaction which is referred to as advancement (Woodruffe, 2006). However, little is known about the motivators of public auditors -SAI staff. The noticed research gap resulted in formulating the research questions presented below.

METHODOLOGY
We took through a literature review using Google Scholar, Scopus and Web of Science (Snyder, 2019). It enabled us to find the research gap and formulate two research questions: RQ 1: How does SAI maximise its utility in the State? RQ 2: What do motivate auditors to work in SAI?
The paradigm we decided to use is interpretivism or symbolic-interactionism (Sułkowski, 2012;Burrell & Morgan, 2017). There are two reasons for choosing a qualitative approach. First is the research phenomenon's open nature, with a few studies. It creates the need for a reflective, open, dialogical approach, with respondents being key informants and experts simultaneously. Second is the confidentiality of the researched matter. Auditors usually examine matters covered by the auditee's secret, so they have to trust that the researcher does not present data available only to auditors.
The study's rigour is based on open questions and a comparison of the answer of respondents. We used data and methods triangulation to get a broader context of the studied issues (Campbell & Fiske, 1959;Greene et al., 1989). The data was collected from two primary sources: unstructured interviews and documents (SAIs' Acts and annual reports). The sample is purposive but gives insight into crucial informants' perspectives.
Based on the auditor's motivators identified in the literature (mentioned above), we assumed that there are the following motivators: salary, post position in SAI (possibility of promotions), stability of employment, workplace security, the possibility of upgrading qualifications, SAI reputation as an employer, proper task realisation, superior's recognition, the nature of tasks carried out, power over auditees.
We decided to use a qualitative approach based on individual interviews conducted by one researcher in person. This researcher has long-term experience in SAI and could conduct interviews. Of course, it also means some subjectivity and personal engagement, which is also possible in qualitative research (Clark, 2010). The list of open questions was formulated after a literature review and has been tested by five interviews conducted with key informants (Faifua, 2014). To assess the auditors' motivators, we used unstructured, open interviews with 11 questions. The transcripts of the interviews with respondents were analysed using the Nvivo14 software. In order to ensure the accuracy of our statements, we have kept the anonymity of the respondents. The respondents were informed that interviews are anonymous and that we will not use direct citations from interviews that could identify respondents or their organisations. Therefore, we chose non-standardised interviews with 28 auditors from 2014 to 2017. These auditors represent 20 SAIs. Although the sample of auditors is small, we consider it acceptable for concluding. Twenty-eight randomly selected auditors represented 20 SAIs from almost all continents (except Australia). These 20 SAIs represent 10,2 per cent of Full Member (196) in the International Organization of Supreme Audit Institutions -INTOSAI (SAIs of all countries which are members of the United Nations Organization or any of its Specialised Agencies and to the SAIs of those supranational organisations which are a subject under international law and are endowed with a legal status and an appropriate degree of economic, technical, organisational or financial integration) (INTOSAI, 2022). The organisation of the research was possible because participants attended conferences in which one of the authors of this study also participated and had the opportunity to conduct interviews. From the perspective of this research goal, it was not important what the organisation's size was from which the auditors came and which countries they represented. We assumed that auditors' answers might be subjective but allow us to understand the processes in which they participate. Such subjectivism is permissible from the perspective of a qualitative approach (Van Manen, 2016).
We decided to prove the model of the SAI's utility based on Polish SAI activity presented in the SAI annual reports. The selection of this SAI for research was deliberate. Firstly, it resulted from analysing the SAI Act and literature studies. We have found that this SAI has one of the most considerable auditing powers among other SAIs in the world and that SAI may undertake audits on its initiative and not only on behalf of Parliament. Secondly, the method of conducting audits by the Polish SAI does not differ from other SAIs (Dobrowolski, 2017).

The model of SAI's utility
The analysis of Polish SAI work shows that this SAI blends parliamentary requests with SAI's audit initiatives (NIK, 1998(NIK, , 2008(NIK, , 2016(NIK, , 2017(NIK, , 2018(NIK, , 2019. For example, in SAI Annual Report in 2017, SAI has pointed out that SAI has submitted 456 audit topics to the 2017 SAI Audit Work Plan. Employees and managers of SAI reported 344 topics in total, while 112 proposed topics -as their suggestions -were submitted by the organs of the Parliament, the Ombudsman and the Prime Minister. SAI has included most of the proposals of the organs of the Polish Parliament in the audit work plan of SAI. A significant part of the proposals of external parties was, to some extent, consistent with those of the SAI (NIK, 2018).
Usage of such a planning approach may increase the SAI's usefulness for society in a short time. Audit recommendations from audits carried out from SAI's initiative provided to auditees may help resolve some public problems. Nevertheless, this does not mean that such audits are crucial for the interest of the whole country and society in a long-term perspective.
To prove the validity of such a generalisation, we reviewed how SAI dealt with the severe fiscal problem -the VAT gap. The Polish SAI reviewed a lot of public tasks, but the number of published SAI audit reports concerning only VAT gap issues did not exceed annually 5% of all SAI reports in the year 2010-2018 (NIK, 2011(NIK, -2019; NIK, P/13/042, P/15/012). Meantime, the VAT gap in Poland (defined as the difference between expected VAT revenues and VAT collected) was one of the biggest in the EU. For example, Poland had one of the most significant 2017 VAT gaps as a percentage of VAT total tax liability according to the law (VTTL), 14% versus 11% and 10% for the United Kingdom and Germany, respectively (CASE -Center for Social and Economic Research, Institute for Advanced Studies, 2018; European Commission, 2019; Sarnowski & Selera, 2019).
The above data allowed us to verify the hypothesis and solve the research problem positively. One may generalise that any SAI tries to maximise its usefulness to society, knowing that strengthening the SAI position relative to other State agencies depends on lawmakers' support. The SAI budget includes the funds for audit activity. These funds must be acceptable to both parliamentarians and the public. Taking this fact into account, SAI tends to increase audit supply by increasing its audits' scope. In the case of an active parliament exercising control functions over public administration, the increase in the number of audits is derivative of the Parliament's interest in government activities. There will be an increase in the Parliament's audit requests compared to all audits undertaken in such a situation. Increasing the number of audits undertaken and the involvement of parliamentarians in audit planning legitimises SAI activities' costs. In the case of Parliament being less active in requesting audits to SAI, the SAI themselves generates demand for audits. The above generalisation is illustrated in the figure (Fig. 1) below.

Figure 1. Model of increasing the SAI's utility by generating demand for audits
Source: Dobrowolski (2009).
The "DA" curve means demand for audits. "A" means the initial parliament's needs for SAI audits. Audits' supply should consider the public interest, articulated in demand from citizens and their representatives. Equipping the SAI with the power to undertake audits on its own initiative (including after considering citizens' requests) will shift the supply curve to point "B". The supply and demand balance point shifts from "C0" to "C1" and this means an increase in the cost of audit from "Kk0" to "Kk1". Society's expenditure related to SAI activities changed from "Kk0Xo" to "Kk1X1". It means it can be in the interest of SAI to stimulate the additional supply for audits carried out by SAI (Dobrowolski, 2009).

The SAI auditor's motivators
After analysing the SAIs' audit rights resulting from the SAI Acts (Kożuch & Dobrowolski, 2014), we found that the auditors' motivation system is the most critical factor influencing the SAI results. There is a need to formulate some assumptions about the motivation of auditors before the start of a study. One must assume that the SAI auditor is guided by the auditor's interest, like every rational person. In the case of judges, research conducted by Posner showed that the salary of a judge is the key motivational factor  (Posner, 1973). Is this the case with the SAI auditor? We interviewed 28 auditors from 20 SAIs (SAIs data has been anonymised to ensure the reliability of auditors' answers. They pointed out that they prefer anonymity, so we fulfilled their requests). 22 out of 28 interviewed auditors confirmed that the possibility of obtaining a better salary motivates them the most. For example:  Auditor No 1 said: "I work for money because I have a family. The salary obtained in my SAI is higher than in other public organisations. If I know that I receive a better salary, I am more motivated".  Auditor No 7 said: "I am sure that all auditors are financially motivated. I expect to obtain additional financial remuneration because of my involvement in the audit".  Auditor No 28 said: "My salary depends on work experience in my SAI and my organisational position. A higher position in SAI's structure means a higher salary. Therefore I want to be promoted to a higher post."  The other 19 auditors responded similarly, using different words. Six out of 28 auditors said they could not unequivocally admit that salary for work is their most important motivator. In addition to high salaries, they are also motivated by (Auditor No 3: "recognition of colleagues", Auditor No 11:"recognition of superiors "; Auditor No 17:" feeling that I do socially beneficial work "; Auditor No 22, 23, 25:" personal development". When asked which salary system was the best, 20 out of 28 people said that pay-for-performance. The rest said that the merit system is the best. For example,  Auditor No 19 said: "pay-for-performance gives more possibilities than the merit system because I can get money for the work done, not for how long I have been working at SAI".  Auditor No 4 said: "I think pay-for-performance encourages me to work more than the merit system". The  Auditor No 18 said: "I want to see link between my involvement and my salary".  other answers were similar. Six people who said that the merit system is better than pay-for-performance said that audits could not be compared because one is more accessible. The other audit can be more difficult (Auditor No 11,12,22,26,27,28). Better is the salary system based on the number of years worked in SAI and the position held. (Auditor No 22). Auditor No 15 said: "since I achieved a high official position in SAI, I expect a certain salary, not pay-for-performance payment". Auditor No 16 said: "I am not interested in pay-for-performance. It is good in business, not in the public sector." All of the interviewed auditors admitted that the stability of employment is considered a motivational factor. Employees' job position during their career is inviolable (auditors cannot be moved to lower official positions except for situations specified in disciplinary proceedings). For example, Auditor No 8 said: "I like SAI because I know they can not dismiss me for no reason. Moreover, my job position and salary are stable. No one can demote me unless I commit a disciplinary action. For that, it will not happen". Auditor No 14 also stressed that "the advantage of working at SAI is the stability of my position as an auditor".
Asked whether the possibility of achieving a good reputation in their SAI motivates them, three auditors said "yes". Twenty-three people said that it does not matter to them. Others said "no". When asked whether the possibility of preparing an outstanding audit program or carrying out an audit in an outstanding manner without interest in achieving an additional salary motivates them, four out of 28 people said "yes". The rest of interviewed people said "no".
These results have shown that wages are a crucial motivator. It was also indicated by Rynes, Gerhart and Minette (2004). The interviewed auditors pointed out two main motivational factors:  They think that a higher salary brings better motivation to work.  The stability of employment is considered a motivational factor. Employees' job position during their career is inviolable (auditors cannot be moved to lower post except for situations specified in disciplinary proceedings).
SAIs are budget units. It means that the number of jobs and salaries cannot be freely determined but must be within a specific range adopted in the budgets (Kożuch & Dobrowolski, 2014). In the absence of freedom in determining auditors' salaries, the solution is to combine a merit-based system in which the salary is increased by the number of years worked with a pay-for-performance system based on financial rewards for work results. Out of 28 auditors interviewed, 20 said the pay-for-performance system is an incentive to work. Eight people stated that they prefer a pay system in which the salary depends only on the job position and the number of years worked.
Another study confirms our conclusion presented above. We asked the auditors the following question: Does the stability of employment in SAI decrease the involvement in the audit? We explained that this stability means that an auditor cannot be transferred to a lower official position if he does not perform his job well. 25 out of 28 auditors admitted that employment stability understood in this way may reduce the auditor's involvement in audits. For example, Auditor No 19 stated: "In my case, employment stability does not reduce my commitment to work at SAI. However, I observe other auditors and see signs of less commitment to work." Auditor No 3 said: "I am convinced that if the auditor is sure that the low commitment to work will not negatively affect his salary, the auditor will not get involved". Three people disagreed. Auditor No 28 said: "I do not think that the stability of employment causes a lack of commitment to work. I do not see it in my case. On the contrary, I am involved and participate in this conference, for example." Auditor No 4 said: "I do not confirm the relationship between employment stability and commitment to work at home. How is it with others? I do not know that." Auditor No 1 said: "In my opinion, the stability of employment does not affect commitment to work, but the amount of salary. The higher, the greater the motivation." Our findings and conclusion fit published findings that there is a positive relationship between payfor-performance perceptions and pay-raise satisfaction, pay-level satisfaction, and overall pay satisfaction (Heneman et al., 1998;Cadsby et al., 2007). The situation in which the employer cannot use vertical promotions to the full extent weakens the incentive system of the auditors. One can propose the following solution. Each auditor should win a competition for a specific official position after the end of the term of office. It would mobilise employees.
According to Herzberg's work we categorized motivation into two factors (Herzberg et al., 1959). We proved that intrinsic factors, such as achievement and recognition, produce job satisfaction. Extrinsic factors, such as pay and job security, may produce job dissatisfaction. We also confirmed Vroom's theory that employee effort will lead to performance, and performance would lead to rewards (Vroom, 1964).
There are many measures of the effective operation of SAIs. Considering the public interest and financial effectiveness of SAI's audits, we chose two factors which depend on how their employees perform audit tasks: 1) effectiveness of combating corruption, the most visible indicator of failures in the public sector. ; 2) the extent to which the audit recommendations are considered accurate and are implemented by the audited organisations. About the first factor, we have found that SAIs' outcomes in fighting corruption remain almost at the same level (Transparency International, 2022). Regarding the second factor, first, we asked 28 auditors whether the extent to which audited organisations implement the audit recommendations depends mainly on the quality of the audits, which depends on the auditors' motivation. All interviewed auditors representing 20 SAIs confirmed such a relationship. However, they also pointed out the political and administrative culture as a factor determining the willingness of audited organisations to implement audit recommendations. They also underlined the role of media in verifying whether SAI's recommendations are implemented.
We decided to use the example of Polish SAI to verify the effectiveness of implementing SAI recommendations by the audited organisations. We found that the media actively inform public opinion about SAI's audits. We also found that it is generally accepted that SAI's recommendations should be implemented if they are valid for audited organisations. SAI, however, has no audit rights to order the audited organisations to carry out the recommendation audit. SAI only has the statutory right to obtain information from the audited organisations on the manner of implementation of the recommendations or the reasons for refusing to implement them.
The analysis of the Polish SAI annual reports showed that the effectiveness of SAI measured by the number of recommendations implemented by the auditees was not high. For example, the percentage ratio of SAI audit recommendations implemented by auditees to the total SAI audit recommendations formulated in 2013-2018 was from 67 per cent in 2013 and 51,5 per cent in 2017 to 52,7 per cent in 2018 (NIK, -2019. Implementing SAIs' audit recommendations depends on the audited organisations. It also depends on the quality of these recommendations, which, in turn, on the auditors' motivation to provide audit services at the highest level.

CONCLUSION
This research was, in many cases, pioneering because the motivation of SAI's auditors is rarely the subject of research. No studies, except this research, presented the model of maximising SAIs' utility and SAI auditor's motivators. The research allowed us to determine what do motivate auditors to work in SAI. It proved that auditors seek to impose their system of preferences through audits and derive usefulness from this. From a long-term perspective, the strategy of audit planning, in which the initiative of the audit comes from SAI's auditors, may lead to the negative autonomy of SAI. SAIs can ignore significant social and economic problems or analyse them to a small extent. This study confirms that auditors strive to maximise monetary income similarly to judges, as Kimenyi, Shughart, and Tollinson (1985) empirically proved.
Besides, the auditors pointed out the stability of employment as a motivational factor. They emphasised that their employees' position during their career is inviolable (auditors cannot be moved to lower posts except for situations specified in disciplinary proceedings). However, they admitted that such an understanding of job stability could lead to an organisational problem revealed by lower auditor engagement during the audits. The study proposes a pay-for-performance system, which does not reduce the auditor's motivation linked with the job stability but simultaneously encourages auditors to achieve better job results.
As with any research, this study also has some limitations. One is a small research sample with few auditors and 20 SAIs. However, since the auditors participating in the research were randomly selected, as were the SAIs, it can be generalised that their answers certainly contribute to further research on the motivators of SAIs auditors. Future research should involve more auditors and their employers. The model we developed does not depend on the research sample size. It was developed based on the principles of audit planning applied by SAIs. We have verified this model's truth on the Polish SAI's example.
Nevertheless, subsequent tests should include more SAIs. We also think that future research should analyse in detail the factors limiting the effectiveness of the implementation of SAI's audit recommendations. In other words, to what extent does it depend on the quality of audits and auditors' motivation, and to what extent on the political conditions in which SAI's audit findings and recommendations are not taken seriously by the audited organisations.
Concluding, we are not saying that we have formulated an invariant audit theory. We researched the motivators of SAIs auditors, and our research is preliminary.