Abnormal Monitoring Costs Charged for Auditing Fair Value Model: Evidence from Jordanian Finance Industry

Abstract

This article inspects the likely outcome of using the fair value (FV) concept on audit firms’ anomalous audit fees. The research performs fixed effects regression to evaluate the given hypotheses using data gathered by hand from 105 Jordanian publicly traded enterprises between 2005 and 2018. The study reveals that FV proxies have a favorable and substantial effect on the atypical audit fees paid by Jordanian enterprises. The findings are more evident for businesses with a higher percentage of subjective FVs (level 3 assets). This research gives current empirical information on the effects of adopting IFRS/IAS for policymakers and standard setters. The results contribute by offering recommendations on the factors that influence audit fees for auditors and clients. The present research updates the Fair Value Disclosure (FVD) auditing model and adds new empirical data to close a gap in the auditing literature. It adds to the limited and inconclusive audit price studies already available by examining the post-implementation of FVD. This research gives current empirical facts on the consequences of adopting an FV model in Jordan for policymakers and standard setters. Additionally, this investigation adds by offering information on the factors that influence audit fees for both auditors and clients. The findings provide regulatory authorities with information on observing and regulating the audit vocation as well as on auditing FVD activities.

1. Introduction

The purpose of this research is to look at the impact of the fair value (FV) model on abnormal audit fees. Given the rising interest in the FV model as a component of the International Accounting Standards Board’s (IASB) agenda [1], the auditing profession now confronts significant new hurdles in producing and validating FVs [2,3]. Additionally, there are issues in discriminating between three distinct input levels for financial instruments [3,4]. level 1 entries represent quoted prices for the equivalent asset or liabilities in effective markets at the measurement date. Level 2 inputs to assets and liabilities are immediately or indirectly visible. Utilizing valuation methodologies, level 3 inputs are dependent on unobservable inputs [4,5].

The main point of emphasis IASB on the FV model is to reflect present economic situations and make current projections regarding future occurrences [4,6]. The use of the FV model has been growing as it promotes such benefits as the disclosure of relevant financial information for decision making, improved transparency of financial reporting, and overcoming the drawback of the historical cost (HC) principle [2,7]. At the same time, Fair Value Accounting (FVA) poses significant audit problems in getting and validating FV inputs [8,9]. High-quality audits as a proxy for the dependability of financial information are under threat from the growing use of sophisticated estimations of FVA [10]. Auditor demand to provide high-quality financial reporting for stakeholders is consequently increased [8]. The inclusion of a higher proportion of fair-valued assets increases the risk of misstatements, increases the auditing efforts and time, especially for level 2 and level 3 inputs of FV [11,12,13,14], and ultimately pushes audit fees higher [15,16]. Abnormal audit fees are an alternative measure of audit fees that shows the variance between the standardized audit fees and the actual paid amount of audit fees. Therefore, this analysis is motivated by the recent call of the IASB for more examinations being done on (post-IFRS) “Fair Value Measurement” consequences on practicing auditing and abnormal audit fees in this study [17].

The empirical efforts on the FV post-implementation are growing but are yet to be an established area of research [15], with limited studies available for developed economies and the unexplored niche of developing countries, particularly for the Middle Eastern (ME) region. Two major studies have examined the impact of the application of FV for financial assets on audit pricing, conducted by [18] in the United States and [19] in the European Union, with varied results. While the experts agreed that sophisticated FV inputs had a beneficial influence on the amount of audit fees, the conflicting findings of past empirical research inspire this study to seek better knowledge on the influence of FVA application on the audit profession in general and aberrant audit fees in particular. Although irregular audit fees have been discovered, they are mostly focused on specialized subjects such as earnings management, cost of capital, and financial restatements and are often ignored in FV research [20]. This study is considered a pioneer one as it investigates the relationship between the FV model and anomalous auditing fees over a long-time duration (2005–2018) for the first time in developing countries [21]. The current research focuses on the Middle East area, namely, Jordan, which has a history of the significant use of FVs owing to industrial concentration [22].

The study performs fixed effects regression to evaluate the given hypotheses using data gathered by hand from 105 Jordanian publicly traded enterprises between 2005 and 2018. The study reveals that FV proxies have a favorable and substantial effect on the atypical audit fees paid by Jordanian enterprises. The findings are more obvious for businesses that have a higher percentage of subjective FVs (level 3 assets). This investigation contributes to a better understanding of atypical audit fees as a result of the FV model’s application. As a result, this is the first research of its sort to evaluate this association. It adds additional empirical data to the auditing literature by addressing a gap. The paper contributes to the present dearth of conclusive research on audit pricing by examining the post-implementation of FV using data from a single developing nation, Jordan. This research is unique in auditing since it integrates stakeholder conceptions of agency and signaling with an FV model to create and investigate the nature of the link between FV and anomalous audit fees. The analysis’s overall results have consequences for policymakers and standard setters since they provide current empirical data on the FV model’s applicability. The findings presented here will pique the attention of both auditors and clients by providing an update to the current audit pricing models used to determine auditing expenses. This study will help the Jordanian government in formulating more precise rules and regulations that elucidate and provide best practices for practices of funding valuation adjustments (FVA).

The rest of this study is organized as follows: Section 2 reviews the institutional context, Section 3 details the evolution of the theories, and Section 4 discusses the study methodology and sample size determination. Section 5 discusses the design and methods of the investigation as well as summarizes the findings and discussion that follow. Moreover, Section 6 contains the sensitivity analysis, and Section 7 brings the work to a close.

2. Institutional Background

2.1. Fair Value Development Overview

FVA was applied for the first time after the Jordanian government issued IAS 39—“Financial Instruments: Recognition and Measurement”, and it had a high impact on the finance industry globally [18]. Financial assets under IAS 39 are required to be classified in one of the following categories ([4], para 34 and 45):

“Financial assets at fair value through profit or loss, Available-for-sale financial assets, Loans and receivables, and Held-to-maturity investments.”

To begin, financial assets at FV through profit or loss are classified into two sub-categories: those designated on initial recognition are to be measured at FV, including changes in FV due to profit or loss; those designated as Held-For-Trading (HFT), which refers to financial assets acquired for the purpose of selling quickly in order to retain shorter-term profits. All derivatives, with the exception of hedging instruments, are included in this category. Second, available-financial-assets (AFS) are non-derivative financial assets that are immediately accessible for sale upon first recognition. According to IAS 39, AFS assets are evaluated at FV and reported on the balance sheet, with their FVs being changed in the equity statement indirectly. Thirdly, loans and receivables are non-derivative financial assets that have fixed payments but are not actively traded; these assets are valued at amortized cost. Finally, held-to-maturity investments (HTMs) are payments made in exchange for financial assets that a business plans to hold until maturity. Investments in HTMs are valued at their amortized cost. The three FV categories’ primary goal is to educate users of financial information about managers’ judgments regarding the FVs that will be realized on financial instruments. Consequently, market participants can easily understand and use this information in their assessments. Later, several IFRSs, including IFRS 7 and 13, made FVD for financial assets mandatory. These IFRS criteria constitute a framework based on principles that businesses may use to evaluate and/or make public their assets’ fair market value, liabilities, or equity instruments [5].

Further developments resulted in the introduction of IFRS 13 in 2013 to demand more disclosure of the FV hierarchy from companies. IFRS 7 is a more comprehensive implementation of FVM since it includes non-financial assets and liabilities in the definition of FVM. FV has been defined by IFRS 13 [5] as:

“The price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date.”

2.2. Fair Value Adoption and Implications in Jordan

Jordan is a Middle Eastern country with close ties to its neighbors and the rest of the world. Cultural and political considerations have had a significant impact when it comes to how firms operate and how they prepare their financial statistics [23]. In early 1988, Jordan joined the International Accounting Rules Committee, resulting in substantial revisions to accounting standards (IASC). Later that year, in 1989, the Jordanian Association of Certified Public Accountants was founded (JACPA). Jordan has around 300 audit companies, and the Big Four international accounting firms (i.e., the Big 4) are also present [24]. The JACPA is in charge of establishing minimum audit fees for external auditors to avoid conflicts of maintaining a high degree of audit quality between auditors and their clients.

The JACPA was then advised by the IASC to adopt the IASs for all Jordanian firms in 1990. Later on, in 1997, the “Firms Legislation No. 22” was enacted, requiring all companies governed by this law to keep accounting records and disclose audited financial information following “internationally accepted accounting and auditing practices.” Jordan’s governance policy framework was established in 1997 with the passage of the “Companies Law.” Soon after, in 1998, the “Securities Act No. 23” was issued, and the Jordan Securities Commission (JSC) declared that all listed companies are required to follow the financial reporting rules of IFRS, and auditing has to be done under the guidelines of the International Standards on Auditing (ISA). Following Law No. 22/1997, it is the job of an external auditor to protect the interests of stakeholders [3]. More laws requiring Jordanian auditors to adhere to International Accounting and Auditing Standards (IAAS) followed, including ones requiring them to examine and submit the final judgment in accordance with IAAS. Auditors should follow Companies Law No. 22/1997 and Accounting Career Law No. 73/2003 [25]. A potential Jordanian auditor must currently have a professional practice license.

As a result of Jordan’s lack of natural resources, the government has been working over the last several decades to strengthen the country’s governance and transparency frameworks in an effort to increase public trust and confidence in the economy. A high degree of openness and comparability of financial information is essential for promoting international commerce between Arab nations and others, and this can only be achieved by developing market economies such as Jordan adopting IAS/IFRS standards. In order to aid users in making choices, such standards aim to raise the number and quality of financial disclosures from companies [24]. Moving to FV as a foundation for accounting measurement, on the other hand, presents significant challenges to the auditing process, particularly in developing countries like Jordan [16].

Jordanian financial enterprises were obliged to adopt FVA by 2005 under IAS 39, and the common assets assessed using FVA are HFT and HFS (HFS). The introduction of FVA in Jordan has caused substantial economic issues. Unrealized gains/losses on fair-valued assets in Jordan boosted share prices during the economic slump. The stock market’s volatility led to bad investing selections owing to a lack of understanding [3]. The growing dependence of the Jordanian economy on exports has increased the use of financial assets by companies, leading to negative news coverage of losses of financial instruments [26]. The challenge of adopting FVA was exacerbated by the increased necessity for FVDs. In response, the Jordan Stock Exchange (JSC) was formed.

The “New Fair Value” regulations were released in February 2008 through the JSC to overcome volatility in the economy and were revised later in 2011 [24]. Recently, during the “boom years” and before the recession, the JSC’s new regulations in 2014 emphasized supervising the role of external auditors in enhancing the quality of disclosed fair-valued information by Jordanian firms. In 2015, the government promulgated its “Jordan 2025” plan to develop commerce with other nations in the area, particularly with the Gulf Cooperation Council (GCC) members, to increase exports [27]. In order to take advantage of free trade agreements, the strategy envisions Jordan as a gateway to the regional markets [28]. To achieve the government’s goals of attracting international investors and ensuring that high-quality financial information is provided, such restrictions may be appropriate. Therefore, Jordan, as the study site for this examination, has been selected for many reasons.

First, the results obtained from this examination can be generalized to the wider ME. The growing literature on accounting in the ME nations is due to their similar heritage, political systems, traditions, and culture [16]. Second, the implementation of IAS/IFRS for almost 30 years in Jordan provides worthwhile insights into how FVMs are prepared and audited under different circumstances [3]. Third, the increased use of financial instruments by firms and companies, as well as the media coverage of financial instruments’ losses, prompts this analysis to focus on the FVA of the financial assets [29]. Finally, Jordan is the only Arab nation with a legal obligation for listed companies to publish audit costs paid in their annual reports since 2001.

2.3. Country Profile

Jordan is an Arab country with strong social and international relationships Factors such as culture and politics have led to several improvements in its corporations and how they do business, especially their preparation of accounting information [23]. Significant improvements in accounting standards began in early 1988 when Jordan became a member of the International Accounting Standards Committee (IASC). This was followed by establishing the Jordanian Association of Certified Public Accountants (JACPA) as a local accounting body in 1989. IASC then advised the JACPA to adopt the IASs for all Jordanian firms in 1990. Later on, in 1997, when “Companies Law No. 22” was issued, this required all Jordanian companies regulated by the Companies Law to prepare accounting records and present audited financial information based on “internationally recognized accounting and auditing principles” [22]. In 1997, the “Companies Law” introduced the framework of government policy in Jordan. Shortly after, in 1998, the “Securities Act No. 23” was issued and declared by the Jordan Securities Commission (JSC); it became necessary for all companies that are mentioned to follow the financial reporting rules of IFRS. Auditing has to be done under the guidelines of the International Standard on Auditing (ISA) [10,22,30].

Given Jordan’s scarcity of natural resources, the government has attempted to strengthen governance and transparency frameworks during the past several decades to increase trust and confidence in the economy [22]. As a result, the adoption of IAS/IFRS standards by developing market economies such as Jordan becomes critical to achieve noteworthy financial information transparency as well as comparability, which consequently will boost global trade between Arab countries and on a global scale [2,29,31,32]. The primary goal of such measures is to enhance the amount of information disclosed in yearly reports. Thus, boosting the quality of financial reporting enables consumers to make more informed judgments [24]. However, the shift to IFSR/IAS has significant consequences for the accounting and auditing professions worldwide, given the deteriorating situation in underdeveloped nations, particularly Jordan [16,33].

By 2005, Jordanian finance companies needed to use FVA under IAS 39, and joint assets that are measured on the basis of FV through FV are held for trading and held for sale. The adoption of FVA in Jordan is the major issue that brought severe problems to the country’s economy. Moreover, the recognition of unrealized gains/losses of the fair-valued assets in Jordan drove stock prices to their highest levels during the downturn. The volatility in share prices led to poor investment decisions due to low market efficiency. Jordan’s economy’s rising dependence on foreign exports resulted in greater use of financial assets by Jordanian businesses, which finally resulted in negative news coverage of financial instrument losses [3,26]. The complexity of applying FVA is compounded by the increasing need for the FV reporting of financial assets. These developments compelled the government to take action via the JSC to address the issues raised by the adoption of FV on Jordan’s stock market., which were revised later in 2011. Recently, during the years of boom and before the recession, the JSC’s new regulations in 2014 emphasized supervising external auditors’ role in improving the quality of disclosed fair-valued information by Jordanian firms. In 2015, the government promulgated the “Jordan 2025” plan, which emphasized an export-oriented economic policy through increasing commerce with other regional nations, particularly the Gulf Cooperation Council (GCC) members. Jordan’s strategy is to become a gateway to regional markets and to capitalize on free trade agreements [28]. Such restrictions might help Jordan’s government achieve its aims of attracting international investors and publishing high-quality financial information by sending good signals about the country’s enterprises’ financial viability.

The Middle East (ME) region is economically diverse, with large discrepancies in natural resources between countries. These countries are most likely to rely on foreign investments and international funds, connecting to the global environment, which becomes paramount to sustain their economic cooperation and political integration [34]. Clarity and transparency of accounting practices become extremely important in such circumstances as connecting to investors, trading partners, and foreign buyers. Considered an attractive setting, Middle Eastern nations operate under a unique political, cultural, legal, and economic environment [35]. Because of their shared history, culture, language, religion, beliefs, customs, and location, researchers are increasingly interested in studying the financial systems of Middle Eastern nations [36]. Accounting reforms such as the implementation of International Financial Reporting Standards (IFRS) are made possible by a trifecta of political, financial, and technology advancements [29]. Although Jordan enjoys political stability, the political and economic conditions in its vicinity have severely influenced the Jordanian economy [35]. Despite the increased involvement of the state in these political conditions in neighboring economies, the foundations of the economy were already built [37,38,39]. These cultural and political elements have resulted in many enhancements to Jordanian firms’ behavior and communication of financial information [29]. Although the Middle Eastern accounting environment data (for Jordan, in particular) are insufficient, the research indicates a growing interest in the region as a gateway for international investments [24], accompanied by dramatic changes happening in the Middle Eastern business environment [3]. This study could help the government authorities to investigate current and future policies in order to provide favorable reporting circumstances for financial statements. Hence, integrating and promoting Jordan and the Middle East in worldwide commerce is a one-way road.

Therefore, to begin with, the outcomes of this research may be generalized to the whole ME, which is why Jordan was chosen as a case study. They share many characteristics, including a shared history and government system, shared religious beliefs, and common cultural traditions and language [23,24]. Second, the rising usage of financial instruments by Jordanian enterprises and the media coverage of financial instrument losses further urge this investigation to focus on the FVA of Jordanian financial assets [3]. Thirdly, since 2001, Jordan has been the first Arab nation to compel publicly traded companies to publish the audit fees they have paid in their annual reports. Finally, Jordan’s almost 30-year experience with IAS/IFRS implementation and auditing allows for useful assessments of how FVMs are created and audited under various conditions [29].

3. Theoretical Perspective, Literature Review, and Hypotheses Development

3.1. Theoretical Perspective

The main objective of this investigation is to explore the correlation between the FV model and abnormal audit fees. FV research and knowledge of audit fees have shown that this phenomenon is studied within agency theory, signaling theory, and stakeholder theory [8,12,15,19,20,21,26]. Scholars have agreed that auditors are viewed as a control mechanism to mitigate agency conflict between managers and agents. Therefore, corporations seek to give an assurance to corporate stakeholders that their financial reports, prepared by corporate managers, are reliable and free from material misstatements [15]. An FV audit not only requires auditors to devote additional effort and time to understanding and validating the complex models and methods used by companies but also exposes auditors to potential future litigation and audit risks. Naturally, it is expected that auditors are compensated for their extra effort and risks through expensive audit fees. Information related to FVA may vary due to the limited market discipline over such complex estimates, and no amount of reliable assurance can be obtained perfectly. Correspondingly, incumbent auditors as an independent third party are expected to provide reliable verification of firms’ FV estimates [15,40,41]. Following previous literature and the theoretical framework of corporate disclosure, these three theories are merged to fulfill the purpose of the current study (see Figure 1). The discussion of each theory is explained below, as follows.

3.1.1. Agency Theory

According to agency theory, managers make decisions on behalf of business owners and should act according to the interests and needs of such persons. However, due to the difference between ownership and management, the latter may not work in favor of the owners. This can lead to mismanagement, fraud, and, consequently, significant errors in the reporting of information [41]. Thus, managers may provide incomplete, misleading, and disjointed information about a company’s financial performance to the detriment of owners [16]. Auditors are a form of monitoring tool that helps shareholders and stakeholders in general to reduce the problem of asymmetric information. Agency conflict between managers and shareholders can be reduced by providing transparent and reliable financial reports audited by a third party, the external auditor [20,21]. Accordingly, the amount of audit fees paid by companies contributes to bridging the information gap between managers and investors [41]. Actually, auditing FVDs requires extreme care as managers choose the quantity and quality of information to be disclosed. FV measurement relies heavily on management judgment, especially in the case of active markets, where managers use valuation methods to prepare such complex estimates [42]. The main effect of these estimates may appear on recorded profits, such as unrealized profits or losses, and such behaviors reduce the quality of published financial reports and thus manipulate shareholders [2].

3.1.2. Signaling Theory

Signaling theory explains the disclosure of company information by managers; it deals with the inconsistency of information due to the agency problem caused by the separation between managers and owners [11]. According to the theory of signaling, external auditors are an indication of the quality of the company’s financial information disclosure [15]. Firms may hire quality auditors to send positive signals to stock market stakeholders, resulting in higher audit fees [18]. Given the increasing use of complex FV accounting estimations, the opportunity for management bias puts more pressure on obtaining high-quality audit services as an alternative to the reliability of financial reporting [29,30,35]; this ultimately results in higher audit fees due to the effort and time involved in auditing these uncertain estimates [15,18].

In general, agency and signaling theories are related to the problem of information asymmetry, and the combination of both theories will help form more ideas about the choice of accounting methods and procedures [43,44]. The rationale for choosing the FV model can be clearly explained using these two theories [45]. The high audit fees paid by companies help control managerial behavior and address information inconsistencies caused by agency conflicts [15]. Additionally, higher audit fees are positive signals sent to various stakeholders as they increase their confidence in financial reporting, attract investment, reduce the cost of capital, reduce capital volatility, and thus develop relationships with stakeholders [21].

3.1.3. Stakeholder Theory

According to the stakeholder theory, the company has a duty to assess the needs of stakeholders and provide them with all necessary information that can be used to make informed decisions. [9,46]. In comparison with agency theory, which expresses the conflict between organization managers and shareholders, stakeholder theory includes a wider range of internal and external users.

The adoption of the FV model has led to increased client complexity and risks to auditors, which sometimes leads to fraud when using such complex estimates [17,21,40]. This situation arises from the thriving use of uncertain FV measurements and complex valuation techniques to determine the FV of assets and liabilities [20]. Hence, this may lead to the manipulation and misleading of external users due to greater management bias in selecting evaluation models [32]. Therefore, attracting external financing (investors, suppliers, or creditors) to continue the normal operations and the necessary obligations requires the presence of a third party (i.e., an external auditor) to verify the credibility and transparency of the managers preparing the FVs of the assets. Hence, these companies apply higher audit fees to obtain this approval due to the complex activities and the high level of agency cost [15].

In general, based on stakeholder theory, disclosure of corporate information is considered a necessary tool to satisfy stakeholder interests in making decisions on resource allocation. In contrast to agency theory, stakeholder theory sees the firm within its broader social fabric, with responsible managers facing a wider range of stakeholders. In this regard, stakeholder theory broadens the focus of agency theory. Thus, managers must support all stakeholder groups with high-quality financial information to make responsible decisions about resource allocation, allowing the merging of both agency theory and stakeholder theory since the latter theory does not use the concept of information asymmetry [38]. Therefore, when discussing the relationship between FVDs and audit fees, integrated theories can be used to explain the relationship between them and justify that audit fees can reduce information asymmetry between companies and all stakeholders.

Figure 1. The conceptual framework.

Figure 1. The conceptual framework.

3.2. Literature Review and Hypotheses Development

Unlike the HC, the IASB introduced FVA to increase financial reporting accuracy. It correctly portrays the true economic condition [43]. Using FVA helps enhance financial reporting and accounting harmonization [14,35]. The increasing information load complicates the auditing procedure [30]. Inherent ambiguity is dangerous [10,11]. Auditors react by giving more time, effort, and the use of valuation experts, resulting in higher audit fees [41]. In other words, FVD allows for a lot of latitude in management assessments. Thus, auditors must spend more time analyzing reputation and lawsuit risk, increasing time spent validating FVEs [15]. Implementing FVA is tough in poor nations [9,31]. Because Jordan’s capital market lacks efficient markets, evaluating the FV of financial assets is challenging [22]. Gains/losses on financial assets boosted share prices to historic highs throughout the recession. As a consequence, share prices dropped in the years that followed [24,25]. Because of the agency issue, managers participated in FV fraud and abuse [26]. The need for independent FVE verification has risen to prevent earnings management tactics, increasing audit expenses [38,39]. Because of this, rising audit costs for Jordanian companies are communicated in stakeholders’ high-quality financial information [45,47,48]. Audit fees rise as more assets are used that have a questionable FV [49]. The risk Is increased for level 2 and level 3 FV inputs, resulting in an increased workload for auditors and, consequently, an increase in audit fees [41]. Due to the agency conflict, enterprises in Jordan actively employ FV to satisfy managers’ objectives. As a result, the Jordanian capital market’s share prices became more volatile due to this exploitation [24]. The absence of an active Jordanian market, inadequate corporate governance systems, and a lack of uniform criteria on how FV should be defined and audited are the root causes of fraud and abuse [22]. Auditors’ expectations in Jordan are higher because of the information asymmetry they must overcome to discover management fraud and misrepresentation. Enhancing a company’s financial reporting’s credibility is, hence, seen as a good indication by stakeholders [48].

Audit fees paid by a company to guarantee that the correct application and disclosure of FVA are classified into two types: average audit fees and atypical auditor costs. The first pertains to the auditor’s work and time spent conducting audits, which is indicative of labor costs and projected damages in lawsuits arising from those audits [49]. However, it includes unusual audit profits since this indicates the economic links between auditors and their clients [50]. An audit charge that is unusually high indicates either excessive effort by the auditors or a danger to the auditors’ integrity. When it comes to the incentives for abnormality, there are two competing ideas in the auditing hypotheses:

Hypothesis 1 (H1).

There is a significant relationship between the share of fair-valued assets and the anomalous audit fees.

Hypothesis 2 (H2).

Among Jordanian listed companies, the correlation between assets at fair value and anomalous audit fees is greater for entities with higher subjective asset fair value ratios (level 2 and level 3).

4. Data and Methodology of the Study

4.1. The Selection of Data

The study data were manually collected from the annual reports of the Jordanian companies listed and published on the Amman Stock Exchange website for the period 2005–2018 [38]. This study commenced in the year 2005 mainly because that was when the FV for financial assets in Jordan became law, as required by ISA 39, followed by the amendment of IFRS 7 in 2009. Accordingly, the study period selected corresponds to the first and most recent schedules of FVD requirements, as required by various IAS/IFRSs, such as IAS 39 (2005), IFRS 7 (2009), and IFRS 13 (2013). The data concerning 2019 and beyond is either not available or is disturbed because of the COVID-19 impacts.

As presented in

Table 1. Sample selection procedure.

	Total Firms	Pooled
Initial sample	235	3290
(-) Firms with missing data	(13)	(182)
(-) Firms belonging to the non-finance industry	(24)	(336)
(-) Firms belonging to industries with less than 10 firms	(72)	(1008)
(-) Firms using other accounting methods (firms with non-FVD)	(21)	(294)
Final sample	105	1470

, the total sample comprised 235 firms, excluding: first, 13 firms with missing financial ratio data; second, excluding 24 firms belonging to the non-finance industry; third, excluding 72 firms operating in industries with less than ten businesses, following [24,34,36]. The ISIN was utilized to classify each industry; finally, the study excluded 21 firms using other accounting methods (i.e., historical cost accounting) or/and firms with missing FVDs for financial assets. Consequently, the overall sample consisted of 105 firms.

Table 2. Final distribution of the sample by industry.

	Total Accepted Firms	Percentage
Financial industry sample
Real estate	28	27
Diversified Financial Services	38	36
Banks	16	15
Insurance	23	22
Total sample from the financial industry	105	100

categorizes the final accepted firms into their primary sub/industries.

4.2. Research Design and Variable Measurements

This analysis builds on the quantitative tradition. This expands the prior audit price models of FVA by [15,18,20,51,52]. Further, the current study introduces new variables, including the abnormal audit fees (AbFees). AbFees is borrowed from [15,52]. Therefore, this study is the first attempt to explore the relationship between abnormal audit fees (AbFees) and similar FV models such as the asset share of FV and the FV hierarchy level inputs; level 1, level 2, and level 3 (FV, FV1, FV2, FV3). The current study extends the previous auditing and FV models into the following two basic equations in order to test the study’s developed hypotheses:

Baseline Model: AbFees = δ0 + δ1LnASSETit + δ2SUBSit + δ3ROIit + δ4LEVit + δ5GROWTHit + δ6RECINVit + δ7FAM_OWNit + δ8FIN_OWNit + δ9BIG4it + δ10TENUREit + δ11OPINIONit + IndFE + ɛ

(1)

Model (1): AbFees = δ0 + δ1FVti + δ2LnASSETit + δ3SUBSit + δ4ROIit + δ5LEVit + δ6GROWTHit + δ7RECINVit + δ8FAM_OWNit + δ9FIN_OWNit + δ10BIG4it + δ11TENUREit + δ12OPINIONit + IndFE + ɛ

(2)

Model (2): AbFees = δ0 + δ1FV1ti + δ2FV2ti + δ3FV3ti + δ4LnASSETit + δ5SUBSit + δ6ROIit + δ7LEVit + δ8GROWTHit + δ9RECINVit + δ10FAM_OWNit + δ11FIN_OWNit + δ12BIG4it + δ13TENUREit + δ14OPINIONit + IndFE + ɛ

(3)

As abnormal audit fees (AbFees) is a dependent variable, this is used for it. Some authors [42,52] often use unusual audit fees as a way to check the validity of their findings. It was employed by [21] to look into how audit fees change when a company has a lot of other comprehensive income. LnAFEES has been replaced by AbFees, which is based on the baseline model’s expected value but not the FV metrics variables; [21,42,52,53] used the term “standardized abnormal audit fees” to describe audit fees that were higher than normal audit fees.

The regression analysis requires that the study data (2005–2018) conforms with four assumptions (normality, linearity, homoscedasticity, and multicollinearity) (see Appendix A). The models were originally examined using panel data to choose the best estimate for the current investigation. The Hausman test looks to see whether the estimates from the fixed and random effects models are significantly different from each other. If p < 0.05 as the χ² value is above the critical value (0.34 to 0.85), then reject the null hypothesis and conclude that fixed effects are to be preferred. Then, the Breusch-Pagan-Lagrange (LM) multiplier test was used to assess whether the multivariate analysis should use random effects or a simple OLS regression. The non-tabulated p-value of the LM test was highly significant (p = 0.001). Therefore, it is clear that the random-effects model is preferable to the current situation.

Following preliminary analyses [49,51], the current study utilized panel data analysis, where fixed effects are used to estimate the possible changes in audit fees over a period of time at the firm level [41]. Certain control variables are incorporated into the current study, including these that have been utilized in prior auditing models [15,46,51], i.e., SIZE, SUBS, ROI, GROWTH, RECINV, FAM_OWN, FIN_ OWN, BIG4, TENURE, and OPINION. All variables were defined, as shown in

Table 3. Variables’ measurements.

Variable	Measurement
AbFees	Audit fees at an abnormally high level (the residuals)
LnAFEES	The natural log of audit fees.
FV	Firm’s total fair-valued assets are deflated by total assets.
FV1, FV2, FV3	Total fair-valued assets divided by total assets.
SIZE	The natural log of total assets of a firm.
SUBS	The number of the firm’s subsidiaries.
ROI	The net income by total assets.
GROWTH	Relative to previous year sales.
RECINV	Entire receivables minus total inventory.
FAM_OWN	The percentage of a firm’s shares held by family members.
FIN_OWN	Total shares owned by financial institutions proportion to a company’s total share count.
BIG4	PwC, KPMG, Deloitte, and E&Y are the Big 4 audit companies; otherwise, the dummy variable is coded as 0.
TENURE	Three-year audit tenure; 1 if the audit firm remains the same, 0 if it does.
OPINION	If you have an unqualified opinion, it is 1, and if you have an unqualified opinion, it is 0.

.

5. Results and Discussion

5.1. Descriptive Statistics and Correlation Analysis

The descriptive analysis of all empirical study variables is summarized in

Table 4. Descriptive statistics.

Variable	Mean	Median	Standard Deviation
AbFees	0.061	0.085	0.566
LnAFEES	9.398	9.196	1.085
FV	0.148	0.073	0.180
FV1	0.124	0.055	0.160
FV2	0.014	0.000	0.042
FV3	0.004	0.000	0.013
SIZE	17.330	16.913	1.972
SUBS	2.254	1.000	3.458
ROI	1268.633	1226.500	762.507
GROWTH	1.636	1.007	3.357
RECINV	0.229	0.176	0.219
FAM_OWN	0.222	0.166	0.216
FIN_OWN	0.288	0.200	0.274
BIG4	0.405	0.000	0.491
TENURE	0.533	1.000	0.499
OPINION	0.877	1.000	0.329
N	1470

for the period 2005–2018, including mean, median, and standard deviation. The variable for abnormal audit fees is AbFees. AbFees’ mean (median) value is 0.061 (0.085), with a standard deviation of 0.566. Audit fees are varied (LnAFEES). LnAFEES has a mean (median) of 9.398 (9.196) and a low standard deviation of 1.085, indicating that audit fees across Jordanian listed companies have a slight variance. In terms of independent empirical variables, the proportion of total assets with FV has a mean (median) value of 0.148 (0.073) and a standard deviation of 0.180. This study shows that assets at FV in Jordanian firms average around 0.13. The volume of assets measured at FV is less than reported by [47] at 0.17 in the United States and by [19] at 0.31 in the European Union, where the capital market differs significantly from that of Jordan. The percentage of total assets measured at FV through the input data for the three-tier hierarchy—level 1 (FV1), level 2 (FV2), and level 3 (FV3)—have mean (median) values of 0.124 (0.055), 0.014 (0.000), and 0.004 (0.000), respectively. The study shows that level 1 assets represent the vast majority of FV assets held by Jordanian firms, accounting for 0.124 of total FV assets. In the US context, the hierarchical means are similar to those published [15,21].

For dependent and independent variables, the results of the Spearman correlation matrix are shown in

Table 5. Correlation matrix.

		1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1	AbFees	1
2	FV	0.0253	1
3	FV1	0.0347	0.921 ***	1
4	FV2	−0.00143	0.319 ***	−0.0151	1
5	FV3	−0.00908	0.237 ***	0.0930 ***	0.144 ***	1
6	SIZE	0.192 ***	−0.269 ***	−0.316 ***	0.0701 **	0.0339	1
7	SUBS	0.0537*	−0.0787 **	−0.0754 **	−0.0319	0.0971 ***	0.250 ***	1
8	ROI	0.108 ***	−0.00452	0.0138	−0.0313	−0.0314	0.357 ***	−0.161 ***	1
9	GROWTH	0.00499	0.0515 *	0.0554 *	−0.00737	−0.00583	−0.0558 *	−0.00592	0.0536 *	1
10	RECINV	−0.0248	−0.0424	−0.0464	0.0204	−0.0126	0.0768 **	−0.199 ***	0.179 ***	−0.0387	1
11	FAM_OWN	−0.0656 *	0.0419	0.0577 *	−0.0657 *	−0.0674 **	−0.106 ***	0.00883	0.0550 *	0.018	−0.0744 **	1
12	FIN_OWN	0.110 ***	0.00157	−0.0283	0.0553 *	0.141 ***	0.186 ***	−0.0152	0.115 ***	0.0733 **	0.0176	−0.431 ***	1
13	BIG4	0.107 ***	−0.159 ***	−0.209 ***	0.111 ***	0.0788 **	0.514 ***	0.0628 *	0.223 ***	−0.0614 *	0.119 ***	−0.129 ***	0.289 ***	1
14	TENURE	0.0947 ***	0.0205	0.0122	0.0367	−0.0105	0.131 ***	0.0879 ***	0.013	−0.035	−0.0378	−0.0408	0.104 ***	−0.0276	1
15	OPINION	−0.0286	0.138 ***	0.130 ***	0.0535 *	0.032	0.0438	−0.147 ***	0.235 ***	0.006	0.0778 **	0.123 ***	0.125 ***	0.0897 ***	−0.0149	1

*, ** and *** denote significance at the 0.10, 0.05, and 0.01 levels, respectively.

. It is not difficult to visualize that the independent variables are not significantly correlated, as shown by the multicollinearity test (MCT), where the mean VIF for each model was less than 2.

5.2. Univariate Analysis

Table 6. Univariate analysis.

Variable	Mean		t-Value (Sig.)
	Fair Value Model (FVA = 1) N = 172 Firm	Cost Model (FVA = 0) N = 50 Firms
AbFees	0.062	−0.089	(−3.3692) ***
SIZE	17.330	16.321	(−6.9381) ***
SUBS	2.254	1.556	(−2.6828) ***
ROI	1268.633	786.566	(−8.4660) ***
GROWTH	1.636	1.381	(−0.9862)
RECINV	0.229	0.136	(−5.6211) ***
FAM_OWN	0.222	0.226	−0.253
FIN_OWN	0.288	0.159	(−6.3361) ***
BIG4	0.405	0.235	(−4.6235) ***
TENURE	0.533	0.480	(−1.3975)
UNQUALIFIED	0.877	0.811	(−2.5634) ***

*** denotes significance at the 0.01 level.

has a lot of things in it. Using Welch’s approximation, the results of the univariate analysis show a variance in the mean value of variables between two groups: the FV model and the cost model of the firms that use the FVA variable. The Mann-Whitney U-test shows that there is also a significant difference in the means of the variables. If you look at [12,15,50,51,52,53], Table 6 shows that there are significant differences in mean AbFees and all the other control variables (Size, SUBS, ROI, GROWTH, RECINV, FAM OWN, FIN_ OWN, BIG4, TENURE, and OPINION) between the two sub-samples of people. Between 2005 and 2018, the analysis found 172 (77%) FV model businesses and 50 (23%) cost model businesses on the ASE. According to the study, the mean difference in audit fees between normal and abnormal was huge (t-value = −3.3692). Samples of audit fees for both FV and cost are very different. The mean audit fees for FV are much higher than the mean audit costs for cost samples. This is consistent with the assumption of agency theory that when using the FV model, the risk of inherent uncertainties increases because managers are biased. FV audits are more complicated and riskier, which means that audit fees are greater because auditors devote more time and effort to their work [54].

In terms of control variables, [55,56,57,58,59] classify them as follows: customer characteristics (SIZE, SUBS, ROI, GROWTH, and RECINV), auditor attributes (BIG4), and engagement attributes (TENURE and OPINION). The data demonstrate that customers that use an FV approach are more likely to be involved in complicated sectors. These clients have a higher total value of assets and subsidiaries than other auditors’ clients (SIZE and SUBS). Additionally, customers that use the FV concept seem to be profitable and financially secure (ROI, GROWTH, and RECINV). These customers are trying to change their auditor following legal requirements (TENURE) and are far more likely than other clients to get an unqualified audit opinion (OPINION). In addition, clients with sector specializations are more likely to use the Big Four (BIG4) audit firms than local audit firms.

It is noticed that the mean of firms with high financial institutional ownership (FIN_OWN) is higher for firms with FV models than non-FV ones (t-value = −6.3361). This outcome is mainly driven by the superiority of financial corporate ownership samples. Thus, the current finding is mainly based on the fact that the majority of audit clients in Jordan are institutional financial ownership, whereby the agency problem, resulting from separating owners from managers, is minimized. However, in the case of family ownership, the opposite is verified. The mean difference between family-controlled and non-family-controlled enterprises is insignificant, and the mean of firms using a non-FV model is greater than the mean of firms using an FV model. This finding corroborates the notion that family-controlled businesses are unlikely to use the FV approach. In comparison to organizations with a centralized ownership structure, several branches and locations, and varied product lines, these businesses are often classified as small businesses [49,55,56].

Those who are more complicated and financially successful are more inclined to employ FVA in order to use the same financial reporting methodology. This demonstrates to others that you have high-quality financial knowledge, which raises your chances of receiving more money [24].

5.3. Multivariate Analysis

Table 7. Result of fixed effects regression analysis.

DV = AbFees Variables	Baseline Model Coeff. (t)	Model (1) Coeff. (t)	Model (2) Coeff. (t)
Intercept	2.484	−0.312	−0.402
	(15.67) ***	(1.95) **	(2.49) **
FV		0.360
		(4.26) ***
FV1			0.537
			(5.47) **
FV2			0.070
			(0.200)
FV3			1.862
			(1.670) *
SIZE	0.372	0.015	0.021
	(38.28) ***	(−1.570)	(2.10) **
SUBS	0.017	−0.002	−0.003
	(3.70) ***	(−0.520)	(−0.560)
ROI	0.000	0.000	0.000
	(2.66) ***	(−0.040)	(−0.380)
GROWTH	−0.010	0.002	0.002
	(2.17) **	(−0.470)	(−0.480)
RECINV	0.178	0.060	0.067
	(2.57) **	(−0.890)	(−1.000)
FAM_OWN	0.014	0.070	0.069
	(−0.180)	(−0.940)	(−0.930)
FIN_OWN	0.311	0.148	0.161
	(4.91) ***	(2.41) **	(2.63) ***
BIG4	0.417	−0.057	−0.049
	(11.49) ***	(−1.600)	(−1.370)
TENURE	0.153	−0.087	−0.095
	(5.11) ***	(2.73) ***	(2.97) **
OPINION	−0.023	−0.006	−0.006
	(−0.480)	(−0.130)	(−0.130)
Coefficient Comparisons for Model (2)			F-Stat	p-Value
FV1 = FV2= FV3			5.37	(0.001) ***
FV1 = FV2			0.95	(0.387)
FV2 = FV3			7.67	(0.005) ***
Firm clustering	Yes	Yes	Yes
Year FE	Yes	Yes	Yes
N	1470	1470	1470
F-Statistics	(10) ***	(11) ***	(13) ***
R2	73%	78%	81%
Mean VIF	1.28	1.27	1.26

Notes: ***, **, and * indicate statistical significance at the 0.01, 0.05, and 0.10 levels, respectively, using a two-tailed test. This table presents the results of fixed-effect regression of anomalous audit fees (AbFees) on the share of assets at fair value (by input level and overall) with firm-aggregated standard errors and fixed-effects for years, following [18,51]. All variables are defined in Table 3.

below presents the fixed effects regression method used to discover the effect of the FV model on atypical audit fees over the period 2005 to 2018. The p-values of the studied models are very important, with the explanatory power of each model ranging from 73% to 81%. Model (2) confirms the existence of a positive significant relationship between the percentage of assets measured at FV and atypical audit fees (coefficient = 0.360, t = 4.26). The study reveals that the use of FVA leads to fee pressure due to the high degree of risk and measurement difficulties introduced by controversial FV metrics. Model (2) examines the relationship between FV hierarchical input levels and anomalous revision fees (coeff. = 0.537, t = 5.47; coeff. = 1.862, t = 1.670). The study also shows that level 1 and level 3 assets have great explanatory power. Although level 2 was not positively significant (coeff. = 0.070, t = 0.200), it was statistically significant (coeff. = 0.070). According to the F-test, the FV hierarchy levels’ coefficients do not seem to be equal (p-value = 0.001). The results show that audit fees for (low and high risk) FV assets have clear effects on audit fees. Although level 1 and level 2 coefficients are indistinguishable, level 3 coefficients are discernible (p-value = 0.005), which corresponds to the initial investigation.

If irregular audit fees and FV measures have a strong link, it may be explained in a number of ways. As a starting point, subjectively, FV assets are more likely to result in audit fees higher than the average amount due to the severe information asymmetry caused by the agency problem [47]. Using FVA has resulted in an audit fee premium to the client because of the additional complexity and risk associated with such a disputed accounting methodology, according to [20]. FV also increases the risk that auditors lose their independence and use discretionary provisions to meet or exceed agreed profit expectations due to conflicts of interest between the manager and the auditor [40,60,61,62]. Another reason auditors deliver unqualified conclusions is that they are paid excessively for their services. This might compromise the auditor’s objectivity and independence, putting the financial statements at risk.

6. Robustness and Additional Analysis

6.1. Using the Presence of the FVA Variable

To demonstrate that the principal regressive findings were resilient to alternative measurements and estimations, the analysis was performed using the presence of an FV model (FVA is coded as 1 if the firm’s assets are reported at fair value and 0 otherwise). Analytical data are tested to see whether they can withstand the presence of an alternative measure of the FV model. It is clear from this study’s findings in

Table 8. Result of fixed effects regression analysis for the FVA variable.

DV = AbFees Variables	Model (4)
Intercept	−0.738
	(4.71) ***
FVA	0.085
	(−1.870) **
SIZE	0.033
	(3.42) ***
SUBS	0.001
	(−0.240)
ROI	0.000
	(2.48) **
GROWTH	0.003
	(−0.660)
RECINV	−0.054
	(−0.790)
FAM_OWN	−0.065
	(−0.900)
FIN_OWN	0.075
	(−1.240)
BIG4	−0.051
	(−1.410)
TENURE	0.015
	(−0.480)
OPINION	0.098
	(2.18) **
Firm clustering	Yes
Year FE	Yes
N	1666
R2	15%
Mean VIF	1.26

Notes: *** and ** indicate statistical significance at the 0.01 and 0.05 levels, respectively, using a two-tailed test. This table presents the results of fixed effects regression of abnormal audit fees (AbFees) on the proportions of fair-valued assets (by input level and in total) with firm-clustered standard errors and fixed effects for years, following [18,51]. All variables are defined in Table 3. Modified model: AbFees = δ0 + δ1FVAti + δ2LnASSETit + δ3SUBSit + δ4ROIit+ δ5LEVit + δ6GROWTHit + δ7RECINVit + δ8FAMILY_OWNit + δ9FIN_OWNit + δ10BIG4it + δ11TENUREit + δ12OPINIONit+ IndFE+ ɛ.

that Jordanian firms are under financial stress due to increased risk and measurement complexity associated with the controversial FV measures. The study showed that the FVA variable is positively significant (coeff. = 0.085, robust t = −1.870).

6.2. Excluding the GFC Year of 2008

More research is needed as the study period (2005–2018) corresponds to the great financial crisis (GFC) of 2008, which may have affected the main results of the regression. We decided to re-examine the assumptions of earlier studies [8,27,54,55,60,61,62] by excluding firm-year data from the whole sample during the crisis year (2008). For a period of 105 business years, data were collected. The regression results are consistent with the main study’s conclusions. There was endogeneity in relation to the Big 4 audit firms.

6.3. Endogeneity in Relation to Big 4 Audit Firms

The two-stage Heckman test was used to alleviate concerns about the possible endogeneity issue of auditor self-selection. The results obtained remain consistent with the initial analysis (see

Table 9. Results of the Heckman test.

DV = AbFees Variables	Model (1) Probit Regression	Model (2)	Model (3)
Intercept	−7.765	−9.612	−9.365
	(17.54) ***	(6.46) ***	(6.31) ***
FV		0.268
		(3.16) ***
FV1			0.436
			(4.43) ***
FV2			0.133
			(0.390)
FV3			1.297
			(1.170) *
SIZE	0.426	0.370	0.362
	(16.17) **	(6.47) ***	(6.35) ***
SUBS	−0.013	−0.022	−0.022
	−1.190	(4.06) ***	(3.97) ***
ROI	0.000	0.000	0.000
	−0.440	(2.11) **	(−1.670) *
GROWTH	−0.022	−0.013	−0.013
	−1.780	(2.69) ***	(2.58) ***
RECINV	0.544	0.438	0.431
	(2.89) **	(4.88) ***	(4.82) ***
FAM_OWN		0.085	0.083
		(−1.150)	(−1.120)
FIN_OWN		0.148	0.157
		(2.45) **	(2.60) ***
BIG4		−0.063	−0.056
		(−1.800) *	(−1.590)
TENURE		−0.088	−0.095
		(2.80) ***	(3.01) ***
OPINION		−0.020	−0.021
		(−0.440)	(−0.460)
INVMILLS		5.465	5.269
		(6.29) ***	(6.07) ***
ATURN	0.316
	(−1.870) *
CURR	−0.002
	(−0.530)
Log likelihood	−762.25
Firm clustering	Yes	Yes	Yes
Year FE	Yes	Yes	Yes
N	1470	1470	1470
Wald chi2	(7)**	(12)***	(14)***
Pseudo R2	23%	76%	80%
Mean VIF	1.21	1.27	1.26

Notes: ***, **, and * indicate statistical significance at the 0.01, 0.05, and 0.10 levels, respectively, using a two-tailed test. This table presents the results of a regression of the fixed effects of anomalous audit fees (AbFees) on shares of assets at fair value (by entry level and overall) with company-aggregated standard errors and fixed effects over many years, following [18,51]. All variables are defined in Table 3.

).

7. Conclusions

This analysis aims to test the relationship between the FV model and abnormal audit fees in the case of a developing country, namely, Jordan. It seeks to broaden the understanding of abnormal audit fees following the application of FVA, which has not been tested in previous FV audit studies and research to date. Using hand-collected data from 105 Jordanian firms from 2005–2018, the analysis of the fixed effects model confirms the significant impact of FVA on abnormal audit fees. The results are even more pronounced for companies with massive percentages of endogenous (subjective) FVDs (level 3 assets). The positive correlation between FVA and abnormal audit fees in this investigation may be attributed to two factors: either the higher audit risk or poor audit quality. The present findings are driven by the fact that the use of the FV model increases the risk that auditors lose independence and use discretionary accruals to meet or exceed agreed earnings expectations due to conflicts of interest between the manager and the auditor. As a result, auditors spend more time and effort auditing FV figures, ultimately leading to exorbitant audit fees to compensate for the high audit risk, complexity, and extra effort involved in the audit process.

This study’s conclusions provide more insight into the issues that auditors face when auditing FVA in Jordan. This will guide regulatory authorities in monitoring the audit profession and regulating the audit of FVA practices. As a result of this research, non-Western governments and standard setters now have up-to-date evidence regarding the effects of implementing FVA. FVD compliance could be improved by providing more explicit instructions and laws, according to the report’s recommendations.

The present study’s outcomes may be limited by the context, sample, timeline, and the research’s selected method. The nature of the audit market in which financial statements are audited varies between developed and developing countries, and since Jordan is a developing country, the results relevant to it may not be suitable for firms operating in other advanced economies. In comparison with the previous literature, the results of the current study can be generalized and benefit a large number of developing countries, especially those in the Middle East, where the same standards, regulations, and reporting framework systems are applied. The results provide valuable insights for policymakers into the challenges of auditing FVD. These insights present valuable input for the development of FVD policies and practices, as well as providing guidance for updating auditor prices. Additionally, the results provide a foundation for policymakers and regulators to introduce and update FV auditing practices.

Future research would extend this analysis for other FV model agents on audit fees, such as intangible assets and investment properties and accounts. It is also preferable to extend the current period of this analysis to examine the impact of economic fluctuations during the devastating COVID-19 crisis.