The intersection between ‘ The Audit Society ’ and public sector corruption and fraud: – a literature review and future research agenda

Purpose – This paper aims to explore how the phenomena of corruption and fraud in the public sector have been portrayedintheliterature using theAudit SocietyFramework(Power, 1997). Design/methodology/approach – The authors conducted a structured literature review (Massaro et al. , 2016)to unveil relevantliterature inthe area of corruptionandfraudin thepublicsector. Findings – Results highlight that the literature using “ The Audit Society ” theory is still scant. Notwithstanding the call for a more decisive role of auditors in ﬁ ghting corruption and fraud, much is still to be discovered aboutconsequencesof auditing andwhat “ good quality ” is. Research limitations/implications – The main limitation is that only literature in English has been included. Practical implications – This paper helps practitioners and policymakers to take and implement informed decisions with regards to the ﬁ ght against fraud andcorruption. Social implications – In calling for more research in the domain of audit, fraud and corruption in the publicsector, this paper promotes a higherfocusof societyonpublicinterest andthecommongood. Originality/value – This paper investigates one part of The Audit Society related to corruption and fraud, topics that are still very underdeveloped and unexplored by researchers. From the ﬁ ndings the authors suggest possible new avenuesfor further research.


Introduction
The public sector is first devoted to the common good of society, but fraud and corruption, among others, can be conceived as a true "cancer" and a plague (Feil, 2021). Undoubtedly, these latter phenomena prevent society from achieving this remarkable public goal. Although there are different meanings of fraud and corruption in the literature, in line with Cooper et al. (2013), in our study we do not distinguish between them and further acknowledge that "fraud or corruption is likely to vary across contexts and over time" (p. 441). Both have a huge impact on society: certainly, the estimates about the economic impact of frauds and corruption are not always accurate (Wathne and Stephenson, 2021), but the misuse of public funds is a growing concern around the globe. Let us think, for one moment, about the huge amount of money funded by national governments and international bodies for COVID-19, the concurrent flooding in Germany (2021, >e10bn estimated damages), the earthquake in Croatia (2021, e5.5 bn estimated damages) or currently the war in Ukraine (2022, e 68bn estimated physical damages in April) and the temptation of fraud and corruption.
Auditing could have a role in fighting fraud and corruption in public sector and auditors should be at the forefront of the public interest (Jeppesen, 2019;Paterson et al., 2019;Nerantzidis et al., 2020;van Brenk et al., 2022). The occasion of the 25th anniversary of "The Audit Society" by Power (1997) gave us the opportunity to reflect upon this and similar claims, with a particular focus on the public sector. One might expect, due the constant spreading of the auditing practices throughout society and time, not the disappearance of the phenomena, but at least a marked decline in the number of acts of wrongdoing. On the contrary, scholars claim fraud, corruption and bribery as an increasing concern in the public sector (Peter, 2017).
Building upon the ongoing research in public sector auditing, we aim to explore how the community of scholars interested in the role of auditing against fraud and corruption has used Power's theory. We were interested in knowing the research locations, levels of analyses, conceptual frameworks, theories and methodologies used by past literature, to offer preliminary answers to this contradiction (more auditing but not less wrongdoings). Accordingly, we first conducted a structured literature review (Massaro et al., 2016), highlighting how the call by Power (2000) for more empirical work on characteristics of "The Audit Society," its causes and consequences, has been answered. Then, we offered a portrait on the use of "The Audit Society," while concurrently suggesting potential avenues of future research.
The paper is organized as follows. In Section 2, the methodology of the literature review is described. In Section 3, the primary results on how Power's ideas influenced the research on public sector fraud and corruption are presented. In Section 4, the scant literature on the public sector that has mobilized "The Audit Society" arguments is reviewed. Finally, conclusions and suggestions for further research are drawn in Section 5.

Methods adopted for the articles' selection
The keywords search is one of the main phases in a systematic literature review (Cronin, 2008). Specifically, the authors decided to search the string ("Corruption" OR "Fraud") using the Scopus database (Grossi et al., 2020) in title, abstract and keywords, as these are the parts of the articles that typically contain keywords (Dal Mas et al., 2019;Natalicchio et al., 2017;Paoloni et al., 2020;Mattei et al., 2021). In addition, to better identify the articles' sample, as the documents most consistent with the research aim, another boundary was defined. In particular, the authors decided to search the name of the investigated book on "The Audit society" (Power, 1997) in the full text and the name of its author "Power" in the references. Therefore, the multiple initial strings were developed as follows: "Corruption" OR "Fraud" in Title, abstract, keywords AND "The Audit Society" in Full text AND "Power" in References The search was limited to articles written in English to avoid any problem related to translation (Mauro et al., 2017). The research was carried out on December 22, 2021, and no one limitation considering the year of publications has been considered. The initial sample consists of 82 papers.
Considering that the manuscript aims to investigate the public sector specifically, the authors refined the analysis introducing another limitation. Then, searching within the results, they added the word "Public" in the initial string and defined the final sample, composed of 74 articles (8 deleted documents).
Subsequently, all duplicate documents were removed (3 manuscripts) because updated versions of some publications had been published (Figure 1), bringing the sample from 74 papers to 71.
Considering that the intention is to analyze only manuscripts with assured quality, published after a peer review process (Easterby-Smith et al., 2012;Massaro et al., 2016), all documents classified as books or book chapters or conference papers were removed, to limit the sample to English peer-reviewed journal articles. Therefore, after this refinement, the sample drops to 50 papers. The next steps were performed thanks to an in-depth analysis of the papers' content against a set of exclusion criteria (Ardito et al., 2015), which are: Type of publication: Conference proceedings, reviews, letters, notes, editorials, books, book chapters, conference reviews, debates, commentaries, dissertations and updated editions of already published documents.
Research focus: Articles with a focus not so close to the search aim.
First, the authors read the title, the keywords and the abstracts, and when some information was not available within them, the full paper was read. The authors initially analyzed documents individually, following the items before identified, whit reference to the following Framework of Analysis: Research methodthe methods used to conduct the investigation ( and then to address the results in a homogeneous way and to reduce both the number of errors and personal judgments, compared the work done with others. In case of doubts or disagreements, the final allocation was collegially decided. The total number of eliminated articles, considering the exclusion criteria and because of their being off topic, was 20, meaning that the number of eligible articles extracted from the databases was 28.
For the eligible papers, authors analyzed all the elements summarized in the aforementioned Framework of Analysis.

Preliminary analysis of results
The articles are published between 1999 and 2021. As can be seen in the Figure 2, although the time span covered by the publications is wide (22 years) after the first publication in 1999 (Chan, 1999), one must wait nine years (in 2008) to find a document dealing with these topics. In the last 12 years, however, scientific production has increased considerably, especially between 2014 Haines, 2014;Morales et al., 2014;Pilcher, 2014) and 2015 (Anders, 2015;Goddard and Malagila, 2015;Neu et al., 2015;Johansson, 2016;Røiseland et al., 2015) and then also between 2018 (Baird et al., 2018;Gustavson and Sundström, 2018;Shore, 2018;Smart, 2018) and 2019 (Jeppesen, 2019;Johnsen, 2019;Wilcock, 2019). Apparently, the two apexes portrayed by Figure 2 are strongly correlated to the publication of two special issues/sections on corruption and frauds (i.e. on critical perspectives on accounting in 2015; on The British Accounting Review in 2019).
Agency theory was another framework used in the retrieved literature (Pilcher, 2014;Røiseland et al., 2015). Koreff et al. (2021) combined a theoretical framework of power (episodic and systemic power by Clegg and social change by Lawrence) with the topic of algorithmic decision-making. Lastly, papers were structured adopting a post-colonial theory (de Maria, 2008) and the Habermasian concept of colonization (Goddard and Malagila, 2015), this latter in combination with a critical approach and a grounded theory method. Other scholars used different paradigm as decision-making (Johansson, 2016) and neo institutional theory (Gong, 2009a). A literature review and future research agenda 4. Fraud, corruption and the "audit explosion" In the previous section, we highlighted in which ways Power's arguments have been explored by the past literature on public sector management and accounting. Surprisingly, notwithstanding the diffusion of the framework in a variety of disciplines (Spence and Toh, 2021;Skaerbaek, 2022) and despite the richness of the theoretical background used in the publications we analyzed, a detailed evaluation of the latter reveals that Power's ideas and approach on "The Audit Society" (Power, 1997) were only referred to and cited without fully being used as an interpretative lens. This evidence shows there is still a marginal use of Power's innovative ideas, at least in the public sector management literature.
Only a few papers seem to contribute in more depth to the insights offered in "The Audit Society" and related works (Shiraz Rahaman, 2009;Neu et al., 2013;Gustavson and Sundström, 2018;Koreff et al., 2021). Shiraz Rahaman (2009) explicitly refers to the explosion of (financial) audit in a developing word setting, because of the pressure by external agencies and the big accounting firms on government, unveiling a predominant emphasis of practices detecting fraud and corruption; however, this focus served legitimacy purposes (Power, 2003), rather than effectively countering the phenomenon. Neu et al. (2013) offered a portrait of a unique case of corruption in an "influence-market" setting of government auditing, detailing how political concerns had a role in the backstage of auditing practices, especially in the judgments and reporting on (potentially) corrupted activities by internal auditors. Confirming a previous intuition of Power (1997), they also claimed that "auditing has a remarkable capacity to be invulnerable to not only its own limitations, but also to the inherently social and political nature of audit practice" (Neu et al., 2013, p. 245), because in the aftermath the implemented solution in the country there was a further increase in auditing practices. Gustavson and Sundström (2018) tried to understand the effects of the audit quality on public sector corruption, the independence of public sector auditors, their professionalism and their communication strategies of the results. The authors also mentioned certain factors identified by Power (1997) as binding elements for the audit and which may therefore be able to reassure the public audience about the quality of public sector audit. Koreff et al. (2021), suggested the reinforcement and strengthening of "The Audit Society" caused using data analytics as well as provided evidence on how this society is created at a microlevel.
A paradox appears from this literature review. In the researched settings, where the practices of auditing, verifying and evaluating have continuously grown in scale and relevance both in society and within organizations, even contributing to construct the "auditee" individual, the expectation may be of less corrupt environments. How was it possible, on the contrary, that fraud and corruption persisted in and around public services, notwithstanding the "explosion" of audit, at least in some countries?
A couple of examples, extracted by the retrieved literature, confirm this contradiction. Canada has been portrayed as a country with a high level of administrative oversight, increased especially at the beginnings of early 2000s, in which auditors shifted their focus toward dimensions of compliance, performance and risk (Liston-Heyes and Juillet, 2021); nevertheless, the country has been the site of a major scandal concerning "sponsorship" to political parties . Another example is from a rather different context, Ghana, in which the phenomenon of fraud and corruption were still depicted as endemic (Shiraz Rahaman, 2009), despite an explosion in the use of financial auditing practices boosted by international agencies as a solution to the "poor performance" problem of African governments (Shiraz Rahaman, 2009).
The answer to this paradox is not so simple. Notably, findings from our structured literature review unveil just initial and contradictory results in the public sector field. First, we do not have sufficient evidence from both groups of countries that experienced an "audit explosion" and/or, as labeled by Johnsen (2019), an "audit contraction," although, as just recapitulated, we know from the past literature that an expansion of audit activities not always couple with the absence or lower levels of fraud or corruption. The problem could be of quality rather than quantity, that is, "good auditing" (Gustavson and Sundström, 2018). In this respect, Power (2005) has suggested some dimensions to explore for the design of a good system and the evaluation of existing ones: the institutional logics, the sanctions, the content of the knowledge base, the question of independence of auditors, the operational processes of auditing and the reporting forms. The knowledge base, in our opinion, deserves attention in the near future, because current content seems inadequate to counter such complex phenomena and digitalization is impacting auditing (Power, 2022). As Shiraz Rahaman (2009) concluded in his analysis of the sub-Saharan setting, the public sector needs a multidimensional approach to defeat corruption and, we might add, the right "alchemy" between "good" external and "good" internal auditing.
Furthermore, according to Power's theory, the explosion of auditing may also have detrimental consequences (Power, 1997). For instance, the study by Koreff et al. (2021) shows that in the data analytics era, acceleration brought by audit trails in interpreting deviant A literature review and future research agenda behavior as fraud, with harmful effects on the auditees, led to normalize their behavior "industry-wide," a form of "audit failure" in the language of Power (1997).
To summarize, the research on the phenomena of fraud and corruption in the public sector within the framework of "The Audit Society" is in our opinion still at an early stage of maturity, even though 25 years have passed from the publication of the first edition of Power's masterpiece. In the next section, some directions for further research are suggested.

Conclusions and suggestions for future research
Auditing and auditors can be conceived as natural organizational change agents (Power, 2003). In line with this, recent literature has been calling for a stronger role of accountants and auditors in the construction of the "disciplined and ethical subject" Jeppesen, 2019) and has acknowledged the development of a new occupational group of forensic accountants and auditors interested in avoiding and detecting fraud (Cooper et al., 2013). Yet, auditors have been showing a willingness to act more in preventing corruption, rather than detecting it (Jeppesen, 2019), or even supporting it .
In his masterpiece "The Audit Society," Power (1997) offered a radical and critical new portrait of auditing practices within organizations and society. Notwithstanding the wealth of insights from the book, from our structured literature review we found a still insubstantial use of its argumentations with reference to fraud and corruption in the public sector. What are the consequences of the increase of auditing aimed to contrast corruption? What are the main challenges in designing and implementing a "good" auditing system aimed to lower fraud and corruption? Which content should auditing have to be effective in fighting fraud and corruption? Is the increase of auditing practices the only solution to counter fraud and corruption? The current literature only provides partial answers.
Based on findings discussed in previous sections and echoing some of the calls for research formulated by Power, we suggest the following avenues of future studies, mainly concerning different methods, research settings and theories.
First, other areas of empirical research based on Power (1997) which could be worthwhile exploring are studies that measure the impact in terms of reduction of fraud and corruption in the public sector as well as auditing reforms before and after their implementation both in a context of explosion or contraction of auditing.
Second, comparisons of different public areas in corrupted countries facing different types of auditing reforms could be useful. Certainly, a further valuable contribution of Power (1997) to the comprehension of the corruption outside the traditional public sphere may, for example, derive from the analysis of corruption as it occurs within the context of hybrid organizational arrangements, such as private-public partnerships, outsourcing, public procurement and state-owned enterprises (Grossi and Pianezzi, 2018).
Third, we see great potential in interpretive research for this field (Denzin and Lincoln, 2011), an innovative approach combining quantitative survey data and textual/visual analyses with qualitative data from interviews and case studies. Generally, the field could benefit from comparing auditing initiatives to reduce fraud and corruption in different jurisdictions and across countries (Mattei at al., 2021).
Fourth and related to the previous point, though we found case studies in countries from all over the world, literature on auditing, fraud and corruption would benefit from more elevated numbers of case studies especially, but not exclusively, from Latin America, European countries, Africa and Asia.
Moreover, the literature on the public sector management would benefit in widening the array of theories or theoretical frameworks used in the analysis, and researchers could also apply mixed research methods to the public sector auditing and corruption.
Finally, a potential future strand of research is the auditing on the multiple-level funding regarding emergency and recovery from calamities deriving from hydrogeological risk, climate change and biological hazards. Therein, whilst we still know precious little about the role played by accounting in natural disasters (Sargiacomo, 2014;Sargiacomo and Walker, 2022), such as the pandemics (Ahrens and Ferry, 2020;Ahrens and Ferry, 2021;Sargiacomo et al., 2021), and man-made disasters, such as war and related humanitarian emergencies, the auditing agenda for multiple-level funding of disasters and the connected use and misuse of public funds connected to growing cases of frauds and corruption at the intersection of the public and private sector is an encyclopedia that still needs to be written.
In conclusion, we all have been spectators over time of the "the expanded quantification of performance and the colonization of organizations by audit and evaluation requirements" (Power, 2022) but notwithstanding increasing internal and external practices of auditing in the public sector devoted to fighting against fraud and corruption, the phenom is not disappearing. We are confident that the above topics will deserve much more attention both by the accounting and by the interdisciplinary community.