• No results found

The effect of auditors on goodwill impairments in the European setting

N/A
N/A
Protected

Academic year: 2021

Share "The effect of auditors on goodwill impairments in the European setting"

Copied!
52
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

The effect of auditors on goodwill impairments in the European

setting

Name: Marleen Verburgh Student number: 10986774 Thesis supervisor: Wim Janssen Date: 18-06-2018

Word count: 15,068

MSc Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam

(2)

Statement of Originality

This document is written by student Marleen Verburgh who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

This thesis examines the effect of auditors on goodwill accounting. More specifically, whether big4 accountants add value to implementing the accounting standard IFRS 3 and IAS 36 in the case of goodwill impairments and if this results in better reflecting the underlying economic value of the firm compared to a non-big4 auditor. It includes a period of the European financial crisis because during this period, many companies recognized impairment losses on goodwill. Results indicate that claims made by accounting regulators suggesting that the IFRS regime better reflects the underlying economic value of firms and thus increases the quality of financial reporting, depend on Big4 auditors enforcing and implementing the IFRS standards. However, the predicted stronger effect that the Big4 auditors deliver higher quality and impair goodwill better according to the IFRS intention during the crisis period is not significant. This study contributes to understanding how Big4 audit quality adds value in enhancing the credibility of accounting information.

(4)

Contents

1 Introduction ... 6

Research Question: ... 8

2 Literature ... 10

2.1 Literature Review on Goodwill impairment... 10

2.1.2 Definition of Goodwill, IFRS 3 and IAS 36 ... 10

2.1.3 Accounting for goodwill; introduction of IFRS 3 and IAS 36 ... 10

2.1.4 Goodwill accounting as a tool for earnings management ... 11

2.2 Literature Review on audit quality ... 13

2.2.1 General audit quality ... 13

2.2.2 Big4 audit quality ... 14

2.3 Literature Review on financial crisis ... 15

2.3.1 Financial crisis in Europe and earnings management ... 15

2.3.2 Audit quality in the financial crisis. ... 17

3 Hypothesis development ... 18

3.1.1 Hypothesis development 1 ... 18

3.1.2 Hypothesis development 2 ... 19

4 Methodology ... 20

4.1 Concept of IOS (Investment opportunities) ... 20

4.2 Research Design ... 21

4.3 Variables ... 22

4.3.1 Dependent variable ... 22

4.3.2 Control variables ... 22

4.3.3 Measuring investment opportunities (IOS), Independent variables ... 24

(5)

4.5 Data Sources ... 26

4.6 Sample limitations and assumptions ... 27

5 Results ... 28 5.1 Descriptive statistics ... 28 5.2 Pearson Correlation ... 33 5.3 Tobit regression ... 34 5.4 Robustness test ... 38 6 Conclusions ... 40 References ... 44 Appendix ... 51

(6)

1 Introduction

Accounting for goodwill has historically been the subject of much discussion among scholars and policy makers and has been associated with considerable variation in accounting practice across countries over time (Bloom, 2009). Goodwill impairment is a complex economic construct that has often been put on the International Accounting Standards Board (IASB’s) or (Financial Accounting Standards Board) FASB’s project agenda. Because of this complexity, these boards have revised goodwill accounting and reporting several times over the years in continuing attempts to improve. The current rules set out by the IASB in International financial reporting standards (IFRS) 3 and international accounting standards (IAS) 36 require that goodwill is to be tested for impairment at least annually based on the estimated fair value of the reporting unit (IASB, 2004). The impairment testing for these intangible assets, required by IFRS may result in impairment losses and these impairment losses reduce net income and the carrying value of assets.

According to the IASB the IFRS 3 and IAS 36 standards will improve financial reporting because the financial statements of entities that acquire goodwill and other intangible assets will better reflect the underlying economics of those assets (IASB). Accounting regulators expect that IFRS 3 and IAS 36 improve the financial reporting environment by allowing managers to convey private information about future cash flows. Conversely, the agency theory predicts that managers will use the discretion in the above accounting standards to extract private benefits. Evidence is found in the literature in favor of the agency theory which proves that the practical application of these standards might actually worsen financial reporting. First of all, Li and Sloan (2017) argue that by eliminating the periodic amortization of goodwill, a subjective impairment test becomes the only mechanism through which the expiration of the future benefits represented by goodwill flows through earnings. Moreover, critics of the current goodwill and fair value accounting also say that fair value estimates are subjective and unverifiable. These critics claim that fair value accounting numbers are noisy and subject to bias which reduces their usefulness to investors and other financial statement users (Watts 2003; Ramanna 2008; Ramanna and Watts 2012). Second, given the difficulty in verifying fair value estimates for goodwill, managers may use this discretion to delay impairment (Watts, Ramana 2008 and Watts and Ramana 2012). Managers are presumably not willing to impair goodwill, as any impairment is likely to be interpreted that they overpaid for the associated business acquisition. The subjective nature of goodwill

(7)

impairments also makes it difficult for auditors and regulators to enforce impairments. As a consequence, Li. and Sloan (2017) show that goodwill accounting has led to relatively inflated goodwill balances and untimely impairments. Clearly, goodwill impairment testing has come under increased attention, and is driven by concerns that its complexity and subjectivity may allow companies to significantly underreport the frequency of goodwill impairment. This might result in impairments that do not reflect the underlying economics of the organization.

However, the function of auditing is to enforce the application of appropriate accounting policies including accounting standards (Ball, 1991) such as IFRS, to represent the firms underlying financial position and performance. The role of auditing in this case is crucial, because accountants monitor and enforce the standards to mitigate opportunistic discretion by management. According to Stokes and Webster (2009), accounting outcomes are a joint product of management exercising their discretion in negotiation with their auditors. Therefore, this thesis examines if big4 accountants allow the accounting discretion of managers to reflect the economic value of the underlying economics of a firm. This paper will look at whether auditing adds value to enforcing and implementing IFRS and more specifically, whether the charges of accounting goodwill impairment under IFRS 3 and IAS 36 better reflect the underlying economic value of the goodwill when high quality auditing such as a Big4 auditor is present for the European setting.

Stokes & Webster (2009) performed a similar research in an Australian setting and their results indicate that the claim of the IASB that the current impairment regime reflects the underlying economic attributes of goodwill, depends on the enforcement and implementation of IFRS standards by higher quality assurance providers such as the Big4. According to Moehrle and Wen (2016), providing a similar study for a European setting would be valuable for informing the standard setters and companies about goodwill accounting.

This thesis wants to find support for the proposition of Stokes and Webster by duplicating the research in the European setting. The broad purpose of this study will be to shed light on the reliability of accounting goodwill numbers by examining whether the goodwill impairment test introduced by the IASB reflects the underlying economics of firms in the presence of Big 4 high quality auditing. This leads to the following research question:

(8)

Research Question:

Do goodwill impairment charges under IFRS 3 and IAS 36 better reflect the underlying economic value of the goodwill in the presence of high quality auditing in Europe?

This study differs of that of Stokes and Webster (2009) in several ways as it identifies several research design choices that contribute to the differences. First, the sample used by Stokes and Webster (2009) is limited to observations only to Australian firms. The sample in this paper looks at the European setting because criticism about the current goodwill accounting method under IFRS in Europe is a popular theme in literature (Devalle and Rizatto, 2012, Li. and Sloan, 2017, Johansson, et al., 2016). Second, it uses a greater sample magnitude, as data is derived from more than 5 years.

Moreover, the European financial crisis initiated a huge public debate in the financial press about the quality of auditors. This lead to greater pressure from regulators for auditors to take a more conservative position on impairment charges. In the final part of this research, a firm’s goodwill impairment behavior is examined over time with a particular interest in the effect of the financial crisis. In most of EU countries the output declined substantially during the crisis years, growth forecasts were marked down, and consequently, valuations in the stock markets plummeted drastically. Thus, one would expect that during this period many companies would have had to reverse projections and business plans underlying acquisitions undertaken in earlier years (Devalle and Rizatto, 2012). One might further expect that these revisions, in turn, would have led to goodwill impairment losses on a rather large scale. In that sense, the crisis serves as a kind of natural experiment for the efficacy of the goodwill impairment approach. During this period, management and auditors were under increasing surveillance to ensure impairments on all assets reflected the underlying economic position of a firm. In the light of conflicting evidence from the literature about the role and quality of auditors during the global financial crisis, this thesis will contribute to the understanding of the quality of auditors during the economic downturn. It aims to provide insight into how firms are applying IFRS 3 and IAS 36 during the financial crisis and concerns whether a decrease in uncertainty about firm value and information asymmetry is associated with the recognition of goodwill impairment. There is no previous study that examines audit quality by investigating goodwill impairments during the global financial crisis. This research attempts to fill this gap in the literature.

(9)

The findings in this thesis are in line with the results from the Australian research carried out by Stokes and Webster (2009): results imply that Big4 auditors better capture the goodwill impairments compared to the non-big4 auditors. In addition, this thesis provides evidence that there is a positive relation between a firm’s IOS and goodwill impairment loss in presence of a Big4 auditor.

However, the predicted stronger effect that the Big4 auditors deliver higher quality and impair goodwill better according to the IFRS intention during the crisis period is not significant. Even after controlling for years within the financial crisis when the crisis was at its greatest, have no effect on significance. This research can therefore not conclude that during the financial crisis, the presence of a Big4 audit firm lead to better application of the goodwill accounting standards.

Overall, the findings contribute to the existing accounting literature in several ways. These findings highlight that, firms use the discretion provided by IFRS 3 and IAS 36 and show commitment to conservative financial reporting strategies in the presence of a Big4 auditor. It highlights the importance in determining how accounting standards with significant discretion are implemented in practice in presence of high quality auditing. With respect to companies, it seems that management has less opportunity to use their discretion under IFRS to manipulate goodwill impairments when they are audited by a Big4 accountant.

Possibly the most important contribution of this study is to provide evidence of the credibility of financial statements as reflections of firms' economic circumstances. This is important to financial statement users, including researchers who use financial statement data to proxy economic characteristics. Investors and users of financial reports benefit when companies are audited by Big4 firms as the Big4 provide greater assurance that the goodwill represented is close to the underlying economic value of the firm. The evidence also contributes to the regulatory debate concerning the treatment of IFRS 3 and IAS 36. Relevant to standard setters, it shows that applying IFRS standards result in improvements in accounting quality when the auditing is performed by a high quality auditor.

(10)

2 Literature

2.1 Literature Review on Goodwill impairment 2.1.2 Definition of Goodwill, IFRS 3 and IAS 36

Goodwill recognized in a business combination represents the payment made by the acquirer in anticipation of future economic benefits from assets that are not capable of being identified individually and recognized separately (IASB 2004a, IFRS 3, para. 52). It is a consequence of the investment opportunity set (IOS) that is not captured by the accounting system, such as expected synergies from the combination of net assets of the acquirer and the acquire and advantages due to market imperfections—the ability to obtain monopoly profits or the existence of barriers to market entry, for example.

This thesis refers to IFRS 3 (IASB 2004a) and IAS 36 (IASB 2004b), which is the accounting regulation that affects EU countries. Despite some differences in the specific impairment rules, the first step, including the estimation of the fair value of the cash-generating unit (CGU) in order to appreciate if an impairment should be recorded, is basically consistent with the related USGAAP—Statement of Financial Accounting Standard (SFAS) 141 (FASB 2001a) and 142 (FASB 2001b). IFRS 3 requires that acquired assets, liabilities, and contingent liabilities are recognized at fair value by the acquirer if they satisfy the recognition criteria, whether or not they have been recognized previously. Any difference between the purchase price and the total fair value of the identifiable net assets should be recognized as goodwill (IASB 2004a, IFRS 3, paragraph. 36).

2.1.3 Accounting for goodwill; introduction of IFRS 3 and IAS 36

The new era of economic development with a growing significance of intangible assets leads to goodwill being a prominent asset and often the largest asset on the balance sheet of companies. According to the growing importance of intangibles there has also been a significant change in standards associated with accounting for goodwill. In 2004 the International Accounting Standard Board (IASB) issued International Financial Reporting Standard (IFRS) 3-Business Combinations and revised International Accounting Standard (IAS) 36- Impairment of Assets and IAS 38- Intangible Assets, which provided a major change in accounting treatment of goodwill after many years. The new accounting standard made a significant change in the accounting rules for business combinations, intangible assets

(11)

and goodwill. The new standard requires that all business combinations which started after March 2004 must use the purchase method and goodwill is no longer amortized but has to be tested for annual impairment (IFRS 3, 2007). The new accounting rule moved ahead American Generally Accepted Accounting Principles (US GAAP), which introduced the approach a few years earlier. The Financial Accounting Standard Board (FASB) issued the Statement of Financial Accounting Standards (SFAS) 141-Business combinations and 142-Goodwill and Other Intangible Assets on July 20, 2001. The Statements changed the unit of account for goodwill and took a different approach on how the goodwill has to be subsequently accounted after its initial recognition (Jerman and Manzin, 2008).

Prior to 2001, goodwill was viewed as a depreciating asset. The value of goodwill in a purchase acquisition was then amortized over a period up to 40 years. To avoid the impact of goodwill amortization expenses on earnings, many firms chose the pooling of interest acquisition method in which purchased goodwill was not recognized and amortized. In June 2001, the FASB issued SFAS 142. It eliminated the pooling of interest acquisition method and required all business acquisitions be accounted for by the purchase acquisition method. In addition, SFAS 142 required sufficient disclosure of the allocation of the purchase price among the assets acquired. SFAS 142 required annual tests for goodwill and other intangible assets. Specifically, it stated that goodwill should be tested for impairment using a two-step process. In the first step, companies compare the carrying value of the reporting unit (including goodwill) to the estimated fair value of the reporting unit. If the carrying value of the reporting unit is less than the estimated fair value of the reporting unit, no impairment in goodwill exists. If the carrying value of the reporting unit exceeds the estimated fair value of the reporting unit, companies perform a second step: to determine and recognize the amount of goodwill impairment loss, which is recorded against earnings. The impairment loss is measured as the difference between the implied value and the carrying value of goodwill. In addition, reversals of goodwill impairment losses are prohibited.

2.1.4 Goodwill accounting as a tool for earnings management

The debate about accounting for goodwill in my thesis centers on the issue of how goodwill accounting can be used as a tool for earnings management. The potential correlation between goodwill impairment and earnings management practices is justified by the fact that as demonstrated by the application of IAS 36, there is considerable scope for discretion. Due to absence of reliable information and in the process of valuing assets that by nature are not

(12)

homogeneous, comparison is difficult. In addition, some of the other factors that increase the possibility of earnings management are the strong information asymmetry between management and stakeholders and the fact that great discretion is granted to management, as goodwill cannot be independently valued using the fair value or fair value in use. For the above reasons, according to Caruso (2016), goodwill impairment is a candidate to be an important tool for managers to affect accounting year-end valuations. Moreover, Watts (2003) argues that reliance on unverifiable fair value estimates such as those required by IFRS 3 and IAS 36, can increase managerial manipulation potential.

Literature discusses that the impairment-only approach is frequently associated with earnings management. Healy and Wahlen (1999) provide the definition of this practice:

“Earnings management occurs when managers use judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcomes that depend on accounting numbers” (Healy and Wahlen 1999, p. 368).

Nevertheless, and despite the prior definition, Healy and Wahlen (1999, p. 369) admit that ‘‘managers can also use accounting judgment to make financial reports more informative for users.’’ The asset impairment request has been embedded in the historic accounting model, and although it is not an easy task to find out whether firms act in an opportunistic way or reflect the underlying economics of the business when recording impairment losses, some authors have tried to answer the question. Reporting incentives, such as debt contracting, ‘‘big bath’’ and income smoothing seem to be the main reason for asset impairment. (Francis et al. 1996; Riedl 2004).

Jarva (2009) finds that goodwill write-offs under SFAS No. 142 predict one- and two-year-ahead cash flows, suggesting that they do convey information. However, he reports that agency-based incentives reduce impairments’ predictive ability. Filip et al. (2015) extend this prior work by providing evidence that managers manipulate current period cash flows to avoid recording goodwill impairment losses. Their study also finds that real activities manipulation to avoid impairment losses is detrimental to firms’ future performance.

Li and Sloan (2017) report that the goodwill impairment does not reflect the economics of the business, as the IAS 36 standard has led to relatively inflated goodwill balances and untimely impairments. According to Li and Sloan; ‘the fair value model envisioned in the new model

(13)

is no doubt well-intentioned but is subject to managerial opportunism. Managers are presumably reluctant to impair goodwill, as any impairment is likely to be interpreted as an admission that they overpaid for the associated business combination’.

Ramanna and Watts (2012) analyze a sample of US firms with market indications of goodwill impairment. Their results are consistent with managers avoiding timely goodwill write-offs in circumstances in which they have agency-based motives to do so; particularly CEO compensation and reputation and debt-covenant violation concerns.

The study of Godfrey and Koh (2009) examines whether goodwill impairment write-offs reflect firms’ investment opportunities (IOS) during the first years of the US goodwill impairment accounting regime. They find that impairment write-offs are negatively associated with firms’ underlying investment opportunities. Consistent with their hypothesis that managers exercise accounting discretion to reflect underlying economic attributes, they find that as firms’ IOS increase, managers write off less goodwill impairment losses.

Chalmers et al. (2011) consider the relationship between impairment and decline in IOS in Australia. They compare the pre-IFRS period with the post-IFRS period, and concur with Godfrey and Koh (2009) that the impairment charges are a better reflection of the underlying economic attributes of goodwill than amortization charges are. Stokes and Webster (2009) extend the Chalmers et al. (2011) analysis by examining the influence of audit quality, confirming the positive role of the Big 4 accounting firms.

2.2 Literature Review on audit quality 2.2.1 General audit quality

As audit quality is a critical issue for the auditing profession, many studies on this topic have been performed. Auditors increase the trust between a firm and its current and prospective investors by giving an independent opinion as to whether the financial statements give a true and fair view of the financial position of the firm. As the auditors’ role is important in helping investors in making informed decisions and enhancing the integrity of financial markets, regulators around the world always aim to improve the quality of audits performed by the auditors.

According to Becker et al., (1998), auditing reduces information asymmetries that exist between managers and firm stakeholders by allowing outsiders to verify the validity of

(14)

financial statements. The effectiveness of auditing, and its ability to constrain the management of earnings, is expected to vary with the quality of the auditor. In comparison to low-quality auditors, higher quality auditors are expected to be less willing to accept questionable accounting methods and are more likely to detect and reports errors and irregularities. It is a valuable form of monitoring used by firms to reduce agency costs between debt holders and stockholders (Jensen and Meckling, 1976, and Watts and Zimmerman, 1983). De Angelo (1981) defines audit quality as the joint probability of detecting and reporting financial statement errors.

2.2.2 Big4 audit quality

A common proxy for audit quality is a dummy variable for BigN/non-BigN membership, and several studies have found support for this (De Angelo, 1981; Francis and Wang, 2008; Chung, 2005). De Angelo (1981) finds that audit quality is directly linked to firm size. Besides, much literature documents that the Big4 auditors provide higher quality audits and offer greater credibility to clients’ financial statements than the non-Big4 auditors.

The standard explanation is that Big4 auditors impose a high level of earnings quality in order to protect their brand name reputation from legal exposure and reputation risk, which can arise from misleading financial reports by clients (Francis and Wang, 2008). St. Pierre and Anderson (1984) report that while they find that auditors are frequently sued for allowing income overstatements, they find no cases of auditors being sued for allowing income understatements. The reputation risk and legal exposure argument is also highlighted by Evans (2015) who assumes that the litigation and reputation exposure to Big4 firms encourage them to exert stronger monitoring and oversight of management’s reporting behavior. In addition, Chung et al. (2005) studies the effect of Big6 audit firms and institutional investors with substantial shareholders in mitigating the low or negative earnings. They assume that Big6 auditors have strong incentives to provide a high audit quality in order to protect their reputation. Their results show that big6 audit firms are effective in deterring managers’ opportunistic earnings management. Therefore, the risk of a Big4 auditor damaging their brand-name reputation is greater compared to a non Big4 auditor.

Another reason investors have greater confidence in the reported earnings of Big 4 clients is that Big 4 auditors are more likely to issue going-concern warnings than non–Big 4 auditors

(15)

for the same set of client circumstances (Francis and Krishnan 1999, 2002). Moreover, Lennox (1999) suggests that the Big N auditors give more accurate signals of financial distress in their audit opinions.

According to Nichols and Smith (1983), the stock market reacts more favorably when a client switches to a Big N auditor than when it switches to a non-Big N auditor. There is evidence by Teoh and Wong (1993) that the financial information audited by BigN firms are being perceived as more credible by the users.

While many researches provide evidence that support the assumption of a positive association between audit firm size and audit quality, Krishnan and Schauer (2000) make a countervailing argument: they state that some scholars argue that audit quality is independent of accounting firm size. Caneghem (2004) studied based on a sample of listed UK companies, the association between Big 5 audit firms and the employment of earning rounding-up behaviors as an indication of earning management. His findings are inconsistent with the hypothesis that Big 5 audit firms are more likely to provide higher audit quality by the constringent of earnings management practices. Khurana and Raman (2004) researched the difference in audit quality between Big4 and nonBig4 auditors in ASEAN countries and compared those results with auditors in the US. They found evidence that the relation between auditor size and audit quality, contrary to US auditors, was not significantly different for Big4 in comparison with non-Big4 auditors in ASEAN countries (Khurana & Raman, 2004).

2.3 Literature Review on financial crisis

2.3.1 Financial crisis in Europe and earnings management

During the financial crisis (FC) in 2008, many firms were confronted by unprecedented market volatility, substantial declines in profitability and high falls in stock prices, needed to recognize asset impairments in accordance with financial reporting standards. These include goodwill impairments. It is known that firms are able to apply alternative accounting methods due to various motivations in order to influence the output of the accounting system in a particular way. It is possible that under the pressure of the economic environment during a financial crisis, firms will be also motivated to make accounting choices that reduce costs and strengthen the picture of their financial position. Impairments can be considered as ‘movable’ expenses used to achieve earnings targets (Watts, 2003). This view predicts managers use

(16)

their discretion over asset impairments to report opportunistically. Opportunistic impairments can manifest in two ways: (1) delaying the recording of impairment losses in the hope that conditions will improve in time to avoid taking the write-off, and (2) recording a larger impairment than is economically justified in order to increase reported income in future periods (i.e., a “big bath”). Since the extent and severity of the crisis was unknown, managers may prefer to delay the recording of bad news in the hope that conditions will improve. Empirical research provides evidence consistent with recent increases in both delayed reporting of impairment losses and big bath behavior (Beatty & Weber, 2006; Ramanna & Watts, 2012; Riedl, 2004). However, according to Devalle and Rizatto (2012) the international financial crisis had lead many companies to recognize impairment losses on goodwill. In this period, management and auditors were under increasing surveillance to ensure impairments on all assets reflected the underlying economic position of a firm. Cimini (2014) investigated whether and how in the European Union (EU), the burst of the 2008 financial crisis affected misrepresentation of financial information due to earnings management. The study found evidence of earnings management decreases after 2008. Explanation for these findings include the fact that companies have common incentives during a crisis to attract potential investors by using high quality financial reporting.

Investigating the European context, Iatridis and Dimitras (2013) explore the change of value relevance and earnings management over the period 2005–2008 and 2009– 2011 in a sample of 66 Portuguese, 48 Irish, 273 Italian, 245 Greek and 157 Spanish non-financial listed entities audited by the Big 4. Their findings suggest that Portugal, Italy and Greece display a stronger tendency towards earnings management, Ireland exhibits less evidence of earnings manipulation, while the findings for Spain are somewhat conflicting (Iatridis and Dimitras, 2013, p. 160).

Kousenidis et al. (2013) address the same issues but achieve homogeneous results investigating whether and how the financial crisis affects earning quality across the EU. In their work, these authors analyzed a sample of 552 non-financial entities listed in Greece, Ireland, Italy, Portugal and Spain over the period 2008–2011. The countries analyzed are those with weak fiscal sustainability and that are under the supervision of the European authorities. As to the evidence about earnings management, they found a reduction of manipulations after the financial crisis, due to a greater interest of entities in disclosing less smoothed and less managed earnings, because firms that rely on external financing and

(17)

struggle with liquidity problems have very strong incentives for increasing their financial reporting quality in order to attract prospective investors (Kousenidis et al., 2013, p. 351).

2.3.2 Audit quality in the financial crisis.

The recent global financial crisis led regulators and the financial media to question whether (Big 4) auditors carried out their auditing duties properly. For instance, the Public Company Accounting Oversight Board (PCAOB) accused auditors of not having applied the PCAOB auditing standards in connection with areas (income statement and balance sheet accounts) that were affected by the economic crisis (Shazad et al. 2017). There are arguments both for and against the possibility that audit quality might have decreased during global financial crisis.

Shazad (2017) suggests that audit quality decreases during downward market conditions because auditors’ decisions are influenced by downward market situation and that auditors ‘may soften their usual skepticism throughout the market madness’. Furthermore, according to Knechel and Willekens (2006), the role of auditing in influencing earnings management during the global financial crisis of 2008 is lacking.

However, the counter stream of academic literature finds that there is no evidence of a decline in quality of auditors instead these studies show that the quality of auditors increased during the global crisis because auditors were under increasing surveillance by regulators to ensure the financial statements reflected the underlying economic position of a firm. (Geiger et al., 2014; Xu et al., 2013).

The findings of Iatris and Dimitras (2013) suggest that firms that are audited by a Big4 auditor are likely to exhibit higher audit quality despite the crisis period. The same results are provided by Chia et al. (2007) who showed that Big4 auditors are able to significantly constrain the earnings management by firms.

(18)

3 Hypothesis development

3.1.1 Hypothesis development 1

Taken together, the literature review on high auditing quality suggests that accountants better reflect the underlying economics of a company in presenting ‘a true and fair view of the financial position of the company’ whereas the IFRS 3 and IAS 36 standards create much discretion for accounting numbers manipulation. Big4 auditors could have a positive effect on the implementation and enforcement of the accounting standards because Big4 auditors impose a high level of earnings quality in order to protect their brand name reputation from legal exposure and reputation risk.

The first motive suggests that Big4 auditors are legally liable for audit failures so they have the incentive to deliver high-quality audit to avoid the costs of litigation. Economically larger clients (often listed companies) often carry greater weight in a Big4 auditor’s portfolio and thus higher litigation-costs are involved. Therefore, a Big 4 auditor may have a higher incentive to report higher audit quality, make sure numbers are transparent, and impair goodwill according to the underlying value of their client-firms.

The second motive applies when Big4 auditors have reputational incentives to avoid audit failures because audit quality is valuable to clients and so priced in the market for audit services. Under this view, clients change to other auditors when the audit firm’s reputation for quality deteriorates. Auditors jeopardize their reputations if their clients are found to have overstated earnings. Often, auditors and clients tend to disagree about accounting choices that are income-increasing rather than income-decreasing and auditors generally require their clients to adjust earnings downwards. The reasons above would imply that there is a greater likelihood Big 4 auditors are more sensitive than non-Big 4 auditors to manage misreporting and it is more likely that a Big4 auditors better comply with IFRS and in this case for goodwill impairment standards in particular. In contrast, the non-Big 4 auditors have less reputation capital at risk and are less likely to challenge client firm defections; so there is a greater likelihood of client firm misreporting with these audits ((Francis et. al., 2004 and Francis and Wang, 2008).

This paper predicts that Big4 audits, relative to that of non-Big 4 audits, will improve the relationship between goodwill impairment charges and their client firms underlying economic circumstances. According to this and in line with the argumentation of Stokes & Webster

(19)

(2009) the following hypothesis is investigated.

H1) Firms that are audited by a big4 auditor are likely to exhibit a stronger positive association between their goodwill impairment and the underlying economic goodwill compared to a non-big4.

3.1.2 Hypothesis development 2

As discussed earlier, the global financial crisis invited huge criticism on the role of auditors and much of that criticism was dedicated to the big4 auditors (Sikka, 2009). The academic research, however, does not provide much evidence of sloppiness in the Big4 auditor’s role.

There are several reasons to believe that in the EU the burst of the financial crisis and presence of Big4 auditors lead to better application of IFRS standards 3 and IAS 36.

The recent financial crisis increases the demand for high financial reporting quality. Especially in the EU, companies had to comply with a high-quality set of accounting standards after the burst of the financial crisis and big4 auditors possess the expertise to address the complex rules set out by the regulators concerning goodwill accounting. Besides, Xu et al. (2011) report that Big 4 auditors responded to the crisis earlier than non-Big 4 auditors which causes Big4 audit quality to increase and application of accounting standards in crisis years to be better.

A second reason is that the European financial crisis initiated a huge public debate in the financial press about the quality of auditors. This lead to increasing surveillance by oversight bodies and greater pressure from regulators for (Big4) auditors to take a more conservative position on impairment charges.

Finally, when conditions began to deteriorate at the onset of the financial crisis, many firms had no other option instead to impair their assets. In this setting, the relationship between the IOS of a firm and the goodwill impairment loss is much stronger and this could potentially result in a stronger effect of Big4 auditors applying the IFRS standards accordingly. In other words, when no impairments are done, Big4 firms cannot make a difference to the effect. According to the arguments mentioned above, this paper predicts that Big4 auditors deliver high quality and impair goodwill according to the IFRS intention during the crisis period. These arguments lead to formulate the second hypothesis:

(20)

4 Methodology

4.1 Concept of IOS (Investment opportunities) IOS and Goodwill

According to Chalmers (2011), the IOS is a concept that is used in the accounting and finance literature to capture a firm’s portfolio of growth opportunities. Initially, it was Myers (1977) who introduced the notion of IOS. He argued that firm value is represented by a combination of assets-in-place and future investment opportunities (growth options) and that investment opportunities can be seen as a call option on real assets with the exercise price being the cost of future investment needed to acquire these assets. Hence, as the IOS captures projects that foster firm growth, the IOS can be seen as a measure that encompasses firm growth opportunity and is an important component of market value. The investment opportunity set of a firm also influences the way a firm is viewed by managers, owners, investors and creditors.

This paper examines the association between IOS variables and accounting policy relating to the goodwill impairment. Goodwill impairments reflect diminution of purchased economic goodwill. Managers aiming to reflect their firms' higher economic goodwill (i.e. growth options) would choose lower accounting impairments than managers of firms with low economic goodwill. In addition, Godfrey and Koh (2009) find that accounting discretion is exercised in a manner that reflects firms’ underlying IOS. They investigate whether managers of US firms use their goodwill impairment write-off discretion to reflect firms’ underling investment opportunities (IOS). They find a negative association between goodwill impairment losses and firms’ IOS during 2002-2004. In particular, firms with greater (lower) investment opportunities maintain higher (lower) amounts of goodwill in their balance sheet, reflecting higher (lower) levels of economic goodwill. This suggests that in the early years of the standards impairment regime, firms’ goodwill accounting approaches reflect the underlying economic attributes of their unidentifiable assets. This thesis will study whether the findings of Godfrey & Koh (2009) generalize to the European impairment regime and draws upon the argument that manager’s use their discretion afforded to them to reflect firms’ underlying economic attributes when they account for goodwill.

(21)

4.2 Research Design

The hypothesis in this paper implies a test for the link between a firms goodwill impairment charges against income and the firms underlying economic goodwill for separate samples of client firms of Big 4 and those of non-Big 4 auditors. In line with the research from Godfrey & Koh (2009) the Tobit regression is used. The Tobit model, (sometimes called a censored regression model) is designed to estimate linear relationships between variables when there is either left- or right-censoring in the dependent variable (also known as censoring from below and above, respectively). Censoring from above takes place when cases with a value at or above some threshold, all take on the value of that threshold, so that the true value might be equal to the threshold, but it might also be higher. In the case of censoring from below, values those that fall at or below some threshold are censored. It is appropriate to use the Tobit regression in this research, given that the dependent variable, GWIL, is left censored at zero. This means that the observations for those firms impairing zero goodwill are also included. This model relates the goodwill impairment charge to the underlying economic value of the goodwill tied up in a firms’ investment opportunity (IOS) is used including a number of control variables (see table 1):

GWILt = a + β1IOSt + β2SIZEt + β3LEVt + β4ROAt + β5RETt + εt …

This paper predicts a negative IOS as this suggests that lower goodwill impairments are associated with higher underlying economic goodwill or investment opportunities of firms. When better investment opportunities reflect a more healthy company, it is expected that less goodwill write-downs are necessary.

Table 1

Variables of the model

where:

GWILt Goodwill impairment measured as goodwill impairment losses(time)/total assets(time).

IOSt Investment Opportunity Set (IOS) measured as the IOS factor derived from six IOS measures

described below.

SIZEt Natural logarithm of total assets(time)

LEVt Total debt(time)/total assets(time)

ROAt Net income(time)/total assets(time)

RETt Stock return for the 12 month period t-1 to t measured as (Price(time) – Price

(22)

4.3 Variables

4.3.1 Dependent variable

Goodwill impairment loss (GWIL) is measured as the firm’s reported goodwill impairment loss before tax, divided by total assets. The model tests whether investment opportunities (IOS) of firms in the sample are associated with goodwill impairment losses and so a negative coefficient on IOS (β1) would imply that lower goodwill impairments are associated with higher underlying economic goodwill in the firms IOS. In other words, the higher (lower) their IOS, the less (more) the firms write off as goodwill impairment. The financial crisis and big4 auditor are moderator variables on the model.

4.3.2 Control variables

While the primary concern is the relation between the firm’s IOS and its goodwill impairment charge, a number of control variables are included in the model to control for firm specific factors that could affect an audit client firm’s goodwill impairment charge. Opportunistic accounting discretion in the presence of debt contracts will increase assets (for example to reduce the debt to asset ratio) and/or increase income. As goodwill impairment charges are often not added back to income in debt covenant ratios (Whittred and Zimmer, 1986) there is potentially an opportunistic motive to play with goodwill impairment charges. This also applies for management compensation contracts that are based on unadjusted earnings. Because of this, the model must include a variable related to debt and management compensation contracts. To proxy for the likelihood of debt contracting incentives to manage the amount of goodwill impairment loss reported in the given year, leverage (LEV) is one of the control variables. Leverage is measured as total debt/total assets – goodwill. Some researchers argue that LEV is negatively associated with goodwill impairment losses since firms may have incentives to delay or avoid goodwill impairments if they are constrained by debt covenants, if they operate in industries where the debt to equity ratio is important variable, or if they generally want to protect their credit ratings. However, the debt hypothesis in this paper predicts a positive association between LEV and the goodwill impairment charge, as leverage reflects financial risk and is likely to be higher for firms that engage in acquisition activity.

Given that the size of a firm could influence its reporting practices, firm size (SIZE) is an important factor to consider when examining goodwill impairments. Compared to larger

(23)

companies, smaller firm might for example have limited resources to complete the complex impairment testing process (Chalmers et al, 2011). Larger companies, on the other hand, are assumed to have experienced more business combinations and acquisitions and are more complex in their structure. Larger companies might also be under closer external monitoring, which means that firm size might not only proxy for the ability to comply with accounting standards, but also for the political pressure of doing so (Watts and Zimmerman, 1990). SIZE is measured as the log of net assets. It is expected that larger companies have a positive correlation with the goodwill impairment losses compared to smaller companies.

Stock return (RET) is used in the model, measured over the current fiscal year of the firm, and predict a negative association between RET and goodwill impairment losses if firms are opportunistic in writing off goodwill slower when stock returns are poor. A firm’s stock return is a market-based, external measure of economic performance. As share prices generally reflect in a timely manner information about a company’s ability to generate cash flows, a negative stock market performance is an indication that the company assets have lost their ability to generate future cash flows and therefore need to be impaired. An argument against this however, is that managers of firms with poor returns are likely to attempt to manage earnings upwards to hide their poor performance. If this argument holds, the association would be positive. Return on assets (ROA) is included as second performance measure and is an internal measure of a company’s financial performance that also potentially reflects management’s expectation of future performance. Existing studies have found companies with poor past performance to report greater impairment losses (Chalmers et al., 2011) and suggested that firms with superior earnings are less likely to experience events that initiate goodwill impairments. This means that higher earnings in previous periods appear to uphold return expectations and thereby the value of goodwill. As with RET, a negative correlation between ROA and goodwill impairment is expected because a firm’s accounting procedure choice is related to how well or badly a firm performs. A firm who performs poorly is more likely to select income increasing accounting procedures. It is important to control for accounting performance because accounting performance might be correlated to a firms IOS: a firm whose assets in place comprise a large fraction of value may have performed better than other firms. ROA is calculated as net income divided by total assets.

(24)

4.3.3 Measuring investment opportunities (IOS), Independent variables

Investment intensity (INVINT), growth in market value of assets (MVAGR), market-to-book value of assets (MBVA), R&D expenditure to assets (R&D), market-to-book value of equity (MKBVE) and earnings-to-price ratio (EP) all have been used to proxy for IOS. These factors are measured taken from the research of Godfrey and Koh (2009), see table 1. The research hypotheses are tested using a factor analysis. The factor analysis process of analyzing the six measures of firms’ IOS results in a one factor per year observation1. The six individual

investment opportunities measures are factor analyzed to obtain a single composite measure of firms’ investment opportunities. The factor loadings are shown in the appendix table A. This is consistent with previous studies which indicate that market-to-book value measures load strongly on the composite IOS measure. However, while a factor score extracts commonalties among a group of variables and reduces multi-collinearity, it also eliminates the ability of each variable to make a unique contribution to the construct of IOS.

After running the regression model for the overall sample, the sample is divided into two sub-samples and a dummy variable is created for public companies that are audited by a Big4 audit firm (Big4= 1) and public companies that are audited by a non-Big4 firm (Non-Big4 = 0). In this second part of the research, I test whether the association between goodwill impairment charges against income and a firm’s IOS is stronger in the presence of a Big4 auditor than a non-Big 4 auditor. Two variables to the first model are added:

Big 4t: dummy variable; 1 if the client firms is audited by a Big 4 auditor defined to be EY, PwC, KPMG & Deloitte; 0 otherwise.

Big 4*IOS: captures the contribution of the association between a firms IOS and higher audit quality under IFRS. The regression will be:

GWIL = a + β1IOSt + β2Big 4t + β3Big 4t*IOSt + β4SIZEt + β5LEVt + β6ROAt + β7RETt + εt

The final part will examine whether the association between goodwill impairment charges against income and a firm’s IOS is stronger in the presence of a Big4 auditor than a non-Big 4 auditor during the global financial crisis. The regression will capture two additional variables; a dummy variable 1 if the firm is audited by a Big4 during the global financial

1 These are 0.08504 (INVINT), 0.13992 (MVACR), 0.42823 (MKBV), 0.02901 (R&D), 0.38470 (MKBVE) and 0.01689 (EP).

(25)

crisis in the period 2008-2013 and 0 otherwise. The Big 4*IOS*financial crisis variable captures the contribution of the association between a firms IOS and higher audit quality under IFRS during the financial crisis:

GWIL = a + β1IOSt + β2Big 4t + β3Big 4t*IOSt + β4FCt + β5Big 4t*IOSt*FC + β6FC*Big4t + β6SIZEt + β7LEVt + β8ROAt + β9RETt + εt

Table 2

The investment opportunity set proxies used to construct the IOS

Variable Description Calculation

INVINT Investment intensity ∑(Capital expenditure + R&D expenditure + Acquisitions)i / (Depreciation)i

where i= t-2 to t

MVAGR Growth in market value of assets

(MVAt / MVAt-n)^1/n

where n= max (1,2,3) for which data are available where MVA = market value of assets = total assets – common equity + market value of equity – goodwill

MBVA market-to-book value of assets

Market value of assets (MVA) / book value of total assets – goodwill

R&D R&D expenditure to assets

(R&D Expenditure)t/ book value of assetst (BVA)

MKBVE market-to-book value of equity

Market value of equity / book value of equity

EP earnings-to-price ratio (Earnings per share + goodwill impairment loss per share)t / (price per share)t

4.4 Sample description

The European countries that are analyzed in this paper are Greece, Ireland, Portugal, Italy and Spain. Those are countries with weak fiscal sustainability during the financial crisis and which are under the supervision of the European authorities. Greece has often been in the spotlight for inadequate quality of financial reporting (Tsipouridou and Spathis, 2012). Before the implementation of IFRS, the quality of Greek accounting standards and disclosure

(26)

practices had been criticized in the European financial press and investors’ community. Some of the complaints were that Greek accounting standards allowed firms too much discretion, lacked disclosures and permitted reporting that was too heavily influenced by tax avoidance strategies. When Greece was affected by the global recession, the crisis deepened and the financial problems of Greece had spread to other weak economies such as Ireland, Portugal and Italy. According to the research of Iatrids and Dimitras (2013) Portugal and Italy show higher level of earnings management practice. Further, the inclusion of Spain is justified because Leuz et al. (2003) find that Spain has developed a code-law institutional framework which has been associated with higher accounting manipulation. Concretely, they find that earnings management is more pervasive in countries where legal protection of a firm’s outsiders is weak (such as in Spain) because insiders have stronger incentives to buzzle firm performance. This paper focuses on these countries that are more affected by the economic downturn which might increase the effect Big4 auditors have on financial reporting quality. The sample comprises all of the non-finance firms listed on the European stock exchanges that report goodwill amounts and goodwill impairments in the their financial statements. By including only those firms that report goodwill amounts and goodwill impairment charges, firms for whom the goodwill accounting treatment is material are captured. The analysis will exclude banks, insurance, pension and brokerage firms, as their accounting measures are not always comparable with those of industrial firms.

4.5 Data Sources

Accounting and financial data are collected from the COMPUSTAT Global database. Other variables like goodwill impairment, total debt and share price are obtained from DATASTREAM. Finally, the auditor variable is found in the Bureau van Dijk/AMADEUS database. Some observations were excluded from the sample for the following reasons: missing independent variables, missing share price data, missing goodwill and impairment data. In order to test the hypotheses, only on those companies applying the IFRS impairment regime were included. The empirical analyses examines 20 Portuguese (219 observations), 19 Irish (190 observations), 89 Italian (911 observations), 33 Greek (315 observations) and 25 Spanish (247 observations) non-financial listed sample companies. The total observations is 1,883. Of these observations, 1,248 have all the data needed to run the statistical analysis.

(27)

From this sample, 137 (out of 186) are audited by a Big4 auditor. Some observations and outliers in data have been deleted (see table B in the appendix for further details).

4.6 Sample limitations and assumptions

To complete the dataset, some assumptions are involved. As the variable auditor (AUDITOR) is a static variable in most databases, this variable needed to be retrieved by hand collecting the data. However, as most companies do not publish all of their annual or audit reports publicly for the sample period 2005-2015, the assumption was made that companies did not switch auditor (from Big4 to non-Big 4 or vice versa) during the sample years. Chang Cheng and Reichtelt (2010) studied the frequency and trend of switching from big4 to small accounting firms in Europe based on data from Audit Analytics. They found that auditor switching is relatively low. According to them, during the period of 2000-2006 6.3% of firms audited by a Big4 firm switched to a smaller firm and 2.1% of the companies initially audited by a small firm changed to a big4 auditor. This indicates that making the assumption mentioned above has only limited effect on the results of this paper. Another reason supporting the assumption that the companies in this sample have not changed their auditor is based on the research of Corbella et al. (2015) who performed a study in the Italian setting where they show that legislation in Italy proves that an audit firm is appointed for approximately 9 years with a 3 years cooling off period. This is in line with the suggestion above that it can be assumed the auditor variable is fixed and doesn’t change over the years. Nevertheless, the final sample excludes certain variables due to missing data from the databases (see table B in the Appendix). A second assumption made was that for those companies reporting zero goodwill impairments during a year, the DATASTREAM database presents N/A. For all those years presenting with N/A the assumption was made that the N/A equaled zero goodwill impairments for that year. A distinction has been made between those years with N/A and the companies who did not provide any goodwill impairment related information which resulted in an ERROR in DATASTREAM. About 30 reports were hand collected to make sure the N/A equaled zero. For all of the N/A observations in the database, the goodwill impairment in the annual reports was set to zero supporting the assumption above.

(28)

5 Results

5.1 Descriptive statistics

Panel A of table 3a reports the descriptive statistics for the variables included in the model estimations for the Big4 sample of firms and the non-Big4 sample. Panel B and C reports them separately for each category. Table 3b reports the variables of crisis and non-crisis years together with a pooled sample. The mean carrying value of goodwill impairment by Big4 audited client firms is 7.5 million which is greater than the amount of goodwill impairment of 5.1 million for non-Big4 audited firms. As the IOS is a factor that composites multiple variables and is not measured on an interval or ratio scale, no conclusions can be drawn based upon the relative magnitudes or signs of the statistics (Chalmers et. al. 2009). However, larger IOS measures indicate higher levels of growth opportunities than lower measures; it is the relativity only that cannot be assessed. The IOS descriptive reported in table 2 indicates a higher IOS (1.117) for the Big4 sample compared to the non-Big 4 sample (0.996). This could mean that Big4 auditors are better able to choose clients that are companies with higher growth factors.

The descriptive statistics further indicate that the Big4 sample firms are larger than the non-Big4 sample firms, with the mean natural logarithms of total assets (SIZE) of 7.00 and 5.90 respectively. The mean leverage (LEV) is 65.6 percent in the Big4 sample which is a fraction smaller than the 68.4 percent of the non-Big 4 sample. In other words, no more than 65.6 percent of the total assets are financed by debt for those companies audited by a Big4 compared to the 68.4 percent for the non-Big4 client firms. Borrowing may be a positive sign, for example firms could be taking out a loan to expect future sales or cash flows that make up the borrowing costs. However, large debt makes business vulnerable. The descriptive results indicate that the Big4 auditor takes a more conservative position about financing with debt compared to the non-Big 4 auditors. ROA explicitly takes into account the assets used to support business activities. It determines whether the company is able to generate an adequate return on these asset. The Big4 sample firms report a ROA of 7 percent whereas to the ROA of non-Big 4 firms which is about 5 percent. Thus, the companies being audited by a Big4 earn about a 2 percent higher return on their assets. Stock returns (RET) descriptive statistics reveal that the mean RET of 9.3 percent in the Big4 sample is significantly greater compared to the mean RET of the Non-big4 sample of -2.1 percent. This

(29)

is in line with the argumentation of Nichols and Smith (1983) who explain that the stock market reacts positively when the client is being audited by a Big4.

Table 3a Descriptive Statistics

Panel A: Pooled Sample (n=1205)

VARIABLES MEAN ST. DEV. MINIMUM MAXIMUM

GWIL 0.0068 0.047 0 1.071 IOS 1.086 0.868 -2.825 10.663 SIZE 6.739 1.890 0.767 12.053 LEV 0.663 0.236 0.062 2.574 ROA 0.066 0.066 -0.629 0.290 RET 0.064 1.483 -0.952 41.462

Panel B: Big 4 Sample (n=905)

VARIABLES MEAN ST. DEV. MINIMUM MAXIMUM

GWIL 0.0075 0.053 0 1.072 IOS 1.116 0.843 -2.825 9.673 SIZE 7.009 1.895 0.767 12.053 LEV 0.656 0.215 0.062 2.575 ROA 0.072 0.063 -0.629 0.263 RET 0.093 1.664 -0.952 41.463

Panel C: non-Big4 Sample (n=300)

VARIABLES MEAN ST. DEV. MINIMUM MAXIMUM

GWIL 0.005 0.021 0 0.207 IOS 0.996 0.935 -1.420 10.663 SIZE 5.926 1.625 2.572 10.373 LEV 0.684 0.289 0.142 2.312 ROA 0.049 0.072 -0.472 0.291 RET -0.021 0.696 -0.921 5.951

(30)

Table 3b Descriptive Statistics

Panel A: Crisis years 2008-2013 sample (n=840)

Variables Mean St. Dev. Minimum Maximum

GWIL 0.008 0.054 0 1.071 IOS 1.044 0.953 -2.825 16.289 SIZE 6.697 1.876 1.416 12.053 LEV 0.678 0.551 0.115 15.346 ROA 0.066 0.070 -0.629 0.291 RET -0.020 2.331 -1 43

Panel B: Non crisis years sample (n=436)

Variables Mean St. Dev. Minimum Maximum

GWIL 0.003 0.023 0 0.321 IOS 1.222 1.028 -1.128 9.673 SIZE 6.644 1.965 0.767 12.023 LEV 0.669 0.282 0.062 2.574 ROA 0.064 0.060 -0.258 0.227 RET 0.235 0.680 -0.766 6.640

Key: GWIL = goodwill impairment (goodwill impairment losses(time)/total assets(time)) IOS = Investment opportunities set (includes INVEST, EP, MBVA, R&D, MKBVE, ) SIZE = natural logarithm of assets of the firm (log. of total assets) RET = stock return: firm's stock return measured over the firm's fiscal year ((pricet-price-t)/(price-t)), LEV= total debt/total assets. ROA = return on assets measured as net income/total assets.

According to Devalle and Rizatto (2012) the crisis created a large amount of uncertainty about the value of firms’ assets and required managers to re-examine expectations about future cash flows. As a result, which is confirmed by table 3b, this time period experienced a higher number of asset impairments. During the financial crisis, the average goodwill impairment loss was significantly higher compared to the years with no crisis; with a mean of 8 million during the crisis years and 3.5 million during the non-crisis years. It suggests that due to unprecedented market volatility, substantial declines in profitability and big falls in stock prices as a result of the crisis, firms needed to recognize asset impairments in accordance with IFRS 3 and IAS 36.

(31)

The descriptive statistics further indicate that both SIZE and the debt ratio (LEV) of firms do not differ during the years of economic downturn. Not surprisingly, table 4 presents a lower IOS for the crisis year sample (1.044) than the non-crisis years sample (1.22) as a larger IOS measure imply higher levels of growth opportunities than lower measures, which means that the growth opportunities for firms have declined during the years of the crisis. Return on Assets (ROA) does not differ much across both periods, with a ROA of 6.6 percent during the crisis and 6.5 percent during the other years. RET however is significantly lower during the years of crisis, showing a negative average return of -2% compared to the 23% of return during the years of no crisis. A possible interpretation of this data is that this results in a stronger tendency towards earnings management for firms, as their stock returns and profitability dropped during the crisis years.

Table 4: T-test

Big 4 sample versus non-Big4 sample

Variables Mean differences T-stat P-value

GWIL -0.002 -0.776 0.450 IOS -0.121 -2.09 0.037** SIZE -1.083 -8.871 0.000*** LEV 0.030 1.776 0.080* ROA -0.022 -5.006 0.000*** RET -0.115 -1.163 0.245

Crisis versus non-crisis

Variables Mean differences T-stat P-value

GWIL -0.005 -1.841 0.066* IOS 0.211 4.083 0.000*** SIZE -0.149 -1.316 0.189 LEV 0.009 0.648 0.517 ROA -0.002 -0.620 -0.536 RET 0.266 3.001 0.003**

Key:Result is significant at p=0.10 **Result is significant at p=0.05 ***Result is significant at p=0.00 See table 1 for abbreviations.

(32)

Table 4 presents the mean differences of the variables. The tests of differences between the Big 4 and non-Big 4 samples indicate that the Big 4 sample firms have a significant different IOS, different SIZE and different ROA compared to the companies audited by a non-Big4. The IOS and stock returns (RET) are significantly different among the crisis years and non-crisis years. The amount of goodwill impairment losses also differ for the non-crisis years, at a significance level of 0.10.

The graph below provides the results of the descriptive data for firms audited by a Big4 compared to the average of the total sample. Generally, it shows that the IOS (underlying value of firms) and the goodwill impairments move in opposite directions illustrating a certain correlation between these variables. For those firms whose IOS decreases, the graph shows an increase in goodwill write-offs and vice versa. However, the peak in goodwill impairments seem to take place one year before the peak in IOS. This might suggest that the auditor anticipates the decreasing IOS and writes off goodwill accordingly. Further, the graph shows that the Big4 auditor carries out slightly higher impairments compared to average impairments over the years.

(33)

5.2 Pearson Correlation

Table 5 reports the Pearson correlation coefficients of the full sample. Even though graph 1 displays a possible relation between the investment opportunities and goodwill impairment loss, the Pearson correlation table shows that these are not significant (-0.029).

Table 5 Pearson Correlation Table 5: Pearson Correlations

GWIL IOS SIZE LEV ROA RET Big4 Crisis Years GWIL 1.000 IOS -0.029 1.000 SIZE -0.022 -0.024 1.000 LEV 0.207* 0.057* -0.030 1.000 ROA -0.187* 0.172* 0.159* -0.324* 1.000 RET Big4 Crisis Years -0.031 0.022 0.053* 0.128* 0.060* -0.0117 -0.019 0.127* 0.005 -0.027 -0.051* -0.019 -0.017 0.143* 0.018 1.000 0.034 -0.086* 1.000 -0.006 1.000

Key:* significant for p=0,05. See key table 1 for abbreviations.

The highest correlation is between two control variables leverage and return on assets. This negative correlation of -0.324 is not surprising, as a higher net income leads to lower debt. The correlation between size and leverage is -0.030 which is consistent with the prediction that firm with more assets in place have greater capacity to support debt. Crisis years is positive significant correlated with GWIL, suggesting that some years might have greater effect on the goodwill impairment compared to others. Contrary to the expectations, there is a negative correlation between SIZE and GWIL (r= -0.022) which is not at the significant level. ROA is negatively correlated with goodwill impairments. Stock returns (RET) and Big4 are also not significantly correlated with goodwill impairment losses. Big 4 is significantly correlated to IOS, SIZE, LEV and ROA. YEAR is correlated to stock returns, possibly attributable to the lower returns during the crisis.

Referenties

GERELATEERDE DOCUMENTEN

Thus, it is considered an office space, yet, with public interventions - such as the Ubuntu Stadstuin community garden, which will be discussed at length in the following section -

Vincent Boschma stelt verder dat hij weinig maatschappijkritische kunst in de commerciële kunstsector tegenkomt: “Ik moet je eerlijk bekennen dat ik in Amsterdam, ook niet, ik ga

To test the value relevance of the impairments and to determine whether managers use the accounting discretion opportunistically or to convey private information about future

‘Auditors’ zijn niet geheel zeker of zij wel ge- noeg inzicht hebben in de bedrijfssituatie; managers zijn bezorgd dat er onvoldoende bedrijfsspecifieke in- formatie in de

Thus, there are significant differences in the value relevance of the accounting metrics across the samples, especially for the goodwill impairment and the earnings surprise as

In de sectoren transport en staal productie hebben alle ondernemingen een goodwill impairment verantwoord.. boekwaarde van goodwill het laagste. Daarnaast is

(1998) and Bugeja and Gallery (2006) it is expect that recent goodwill creates significantly more negative abnormal returns upon accounting impairment announcement

In the positive earnings sample earnings are insignificant when book value of equity is included, but when earnings are the only independent variable the coefficient is positive