Take care of your daughter!
A sequential case study identifying the factors underlying accounting fraud at
subsidiaries
University of Amsterdam Amsterdam Business School Amsterdam, 17 March 2015 L.L.W. Meys (1020692)
Executive Master of Finance & Control Supervisor: Prof. H.P.A.J. Langendijk Thesis
1
Table of Content
Abstract ... 2
Preface ... 3
1. Introduction ... 4
2. Literature review and hypothesis development ... 6
2.1 Types of corporate fraud ... 6
2.2 The control environment of subsidiaries ... 7
2.3 Incentives to engage in accounting fraud ... 8
2.4 The opportunity to engage in accounting fraud ... 10
2.5 The company and subsidiary management attitude towards accounting fraud ... 12
3. Research method, sample selection & hypothesis development ... 14
3.1 Research method ... 14
3.2 Sample selection... 15
3.2.1 Fraudulent companies ... 15
3.2.2 Peer sample selection... 16
3.3 Hypothesis development and parameter selection ... 18
3.3.1 Assessment of the control environment of subsidiaries ... 18
3.3.2 Assessment of the incentive to possess fraudulent behaviour ... 20
3.3.3 Assessment of the opportunity to engage in accounting fraud ... 21
3.3.4 Assessment of the attitude of subsidiary and parental management ... 23
4. Analysis and hypothesis assessment of selected cases... 24
4.1. Assessment of the control environment within subsidiaries ... 24
4.2 Assessment of the subsidiary management incentives to engage in accounting fraud ... 27
4.3 Assessing the opportunity of subsidiary management to engage in accounting fraud ... 30
4.4 Assessment of the management attitude towards accounting fraud ... 32
4.5 Summary of findings ... 34
5. Conclusion, limitations and future research ... 37
5.1. Conclusion ... 37
5.2 Limitations & future research ... 39
Reference list ... 41
2
Abstract
This thesis researches factors underlying accounting fraud at company subsidiaries by means of a sequential case study involving nine companies where accounting fraud at the subsidiary level has been detected and a bribery case. The literature review shows that the control environment of the subsidiary, the incentive of company or subsidiary management to engage in accounting fraud, the opportunity of to commit the fraud as well as the attitude towards fraud of this management can be determining factors. The executed sequential case study compares those companies where accounting fraud was detected on a one-to-one base with peer companies which are selected on factors such as industry and size, indicating differences between the fraudulent group and the peer group. Research on both groups was done by researching the annual reports, newspaper articles as well as internet resources. The case studies show that for companies in which fraud took place by management at the subsidiary level there were control deficiencies reflected in cultural, lingual and physical barriers as well as lack of integration in comparison to the peers. For most of the studied fraudulent subsidiaries, the respective subsidiary management had a clear incentive to engage in accounting fraud. This incentive was reflected in either to receiving a high selling price before acquisition by the parent to achieve personal gains, or subsidiary managements wish to remain relatively independent from the parent. Although the size of these subsidiaries was substantial in many cases, outside stakeholders failed to prevent the fraud, whereas for two peer companies large ownership at management prevented their subsidiaries from engaging in accounting fraud. Last, the aggressive attitude and strategy of subsidiary management made the subsidiary vulnerable to engage in accounting fraud, not prevented by the parent due to weak control. Future research on the area of accounting fraud within subsidiaries is needed to provide empirical evidence. Furthermore, future research may uncover the motivation of the subsidiary to remain independent and the relation between different cultures within one company and accounting fraud.
3
Preface
Upon completion of thesis, I would like to thank prof. Langendijk for his constructive feedback and support on this thesis. Furthermore, I would like to thank Elena for her patience and support . Last, I would like to thank Nuon as well as my colleagues for enabling me to participate in the study and motivating me to bring it to a good end.
Kind regards,
Luc Meys
4
1. Introduction
Corporate scandals have frequently been the center of attention during the last decades. The substantial accounting fraud at Enron, misstatements of oil reserves at Shell or the overstatement of results at the American subsidiary of Ahold indicate that management not always achieves results by merely reporting on what has really happened, but might instead present what they aspired to happen.
Meeting aspiration for growth and results can be stimulated by pressure from shareholders and urge of top management to achieve ambitions and possible bonuses (Jiang & Wang, 2010). Especially for companies with a good track record, management feels superior and might get over confident towards the achievement of these results (Schrand & Zechman, 2012).
Sustainable strategies to meet ambitions or expectations such as organic growth and concentration on core business (Zook, 2010) take time and devotion. A quicker way to achieve growth are acquisitions. In this way, the company can quickly ‘buy’ growth, regardless of the fit of these parties within the sustainable future or business segment or culture. Either via acquisitions or organic growth, an expanding business will eventually become a corporate entity with subsidiaries in different places with their own cultural, geographical and historical background and characteristics, resulting in control challenges.
Next to any ambition of parental management, the respective subsidiary management may also have certain ambitions and incentives. The combination of delegated control, being an acquisition candidate or priorities on local matters in as well as incentive programs by the parent make subsidiaries vulnerable for fraudulent activities. Furthermore, being under pressure, the parent company might set the subsidiary high targets which can result in fraudulent behavior by the subsidiary to either please or cover up unexpected underperformance (Badawi, 2005).
For the reasons stated above, accounting fraud is not merely a matter of the head office management, but may manifest at lower level within management of the company or its subsidiaries.
Although the research concerning accounting fraud within companies is extensive, the fraudulent activities at the level of subsidiary management received low attention in the academic literature. Because of the rather exploratory nature and low amount of publicly available data on subsidiary activities, this study will assess cases in which accounting fraud was revealed in context of subsidiaries by means of a case study of nine companies in which fraud was committed within the
5 subsidiary context and review these cases with insights from existing literature concerning accounting fraud. This review will ultimately answer the following research question:
To provide an answer to this question, the remainder of this thesis is organized as follows. Chapter 2 will elaborate on the academic literature regarding fraud in general, the subsidiary control environment and the respective framework to assess accounting fraud within the context of this thesis. Chapter 3 will explain the research method sample selection and discuss the developed hypothesis and respective parameters which will be assessed. Chapter 4 discusses the analysis of the case studies and provide analysis regarding the hypothesis. Chapter 5 will provide the conclusions of this study as well as limitations and suggestions for future research.
6
2. Literature review and hypothesis development
In parallel with the revelation of corporate accounting fraud, a substantial stream of literature developed concerning this matter, in which lengthy case studies (Jones, 2011), influences of the press (Miller, 2006) or (big four) auditor involvement (Lennox & Pittman, 2010) do not remain untouched. Regarding the involvement of a company subsidiary, the academic literature is scarce. Although the main lever of this study is accounting fraud, the scarcity of cases provided does lead to inclusion of other fraudulent behaviour for one case. For that reason, the first section will elaborate on the different forms of fraud. The second paragraph will discuss the literature on control mechanisms between the parent company management and subsidiary management, drawing the first framework of circumstances allowing accounting fraud. Since many other factors other than control make a subsidiary decide to commit accounting fraud, general literature concerning accounting fraud will be touched upon as well, drawing up the framework by which the cases within this study will be assessed on. This general literature will discuss why management commits accounting fraud (incentive), what their windows of opportunity are and what attitude of management of those firms has towards accounting fraud. These factors will be directly linked to the case of subsidiary accounting fraud.
2.1 Types of corporate fraud
To put the cases into perspective, this section will first discuss the concept of fraud. According to Wells (2011), the three types of corporate fraud are corruption, asset misappropriation and fraudulent accounting. This study mainly covers fraudulent accounting, which is any accounting not being within the regulatory framework (Jones, 2011), where the person committing the fraud knew about the regulatory framework. Fraud concerning corruption mainly covers areas such as bribery, conflicts of interest, illegal gratuities and economic extortion. This type of fraud is slightly related to the research within this thesis as those companies with subsidiaries may encounter practices such as bribery which are, according to the subsidiary culture fully normal, but not accepted within the parental context (Calori et al., 1994). Misappropriation of assets concerns theft of cash or inventory. Since it is unlikely that either the parent or subsidiary allow such behaviour and this fraud is more in a personal instead of corporate collective context, this type of fraud is fully out of scope for this study. For the current study, the main focus on literature and analysis will be on accounting fraud. That is, this type of fraud is relevant in most case studies. For one case in which corruption (by bribery) played a larger role, the general factors concerning control, incentive and attitude of management are assumed to be comparable, providing slight extra evidence to the general fraud findings on the accounting fraud cases.
7
2.2 The control environment of subsidiaries
The relation between the subsidiary and parent company defines parent-subsidiary control. Control of a subsidiary can be defined as “the process by which one entity influences, to varying degrees, the behaviour and output of another entity through the use of power, authority and a wide range of bureaucratic, cultural and informal mechanisms” (Geringer & Hebert, 1989). A parent not able to control the subsidiary will have difficulty understanding the subsidiary activities resulting in a non-transparent situation in which (accounting) fraud by the subsidiary is easier to commit without discovery by parental management due to intransparency and non-standardized procedures.
The formal relationship between subsidiary- and parent management constitute the foundations of control. Concerning ownership, an empirical study by Jausaud and Schaaper (2006) shows that formal mechanisms as well as standardization are the main goals for those companies fully owning their subsidiaries whereas behavioural aspects play a more significant role for company management with subsidiaries in which they have lower stakes. That is, lower the ownership stakes results in increased informal mechanism and a lower extend of formal control. Regardless of ownership structure, high mutual dependence may still induce stringent control. Analysis by Subramaniam and Watson (2006) concludes that with respect to subsidiaries possessing important resources within their own geography, remaining independent is optimal where the parent merely acts as a capital provider. Concerning subsidiaries possessing valuable or complementary resources for the parent, a more integrative approach optimizes results which is accompanied with more stringent control. That is, substantial ownership and mutual dependency create better (respected) formal control compared to a relation with less capital ownership or integration possibilities.
Next to formal relations, mutual understanding is evident for effective control. Alignment and transparency mainly concerns the strategic direction, reflected in goals and incentives improving the relation regarding decision making and control between the subsidiary and parent (Kownatzki et al., 2013). Furthermore, transparency, goals and incentives enhance trust and participation, making sure the subsidiary has less incentive to deviate or oppose to the parent. Ultimately, this trust should also be reflected in aligned culture as show by a qualitative study of Kownatzki et al (2013).
Within the cultural domain, empirical research by Björkman and Piekkari (2009) concludes that more centralized and formal control is in place for foreign subsidiaries in which the subsidiary management does not master the language of the parent sufficiently. Especially in the first years after a foreign acquisition, the management of the acquiring company is likely to execute the control mechanism in line with the national heritage of the parent company, which may clash with the control culture of the acquired company (Calori et al., 1994). This formal control mechanism allows the parent to make
8 the decisions, formally avoiding misinterpretations of any subsidiary decision (Harzing & Feely, 2008). Obviously, divergent interpretations between the subsidiary management and parental management may still occur in this way, neglecting feedback concerning incorrect decisions by the subsidiary as show by cases studies (Ferner, 2000). Furthermore, within the other country or culture, businesses habits where fraudulent behaviour might be ‘normal business’ in the eyes of the subsidiary in case of e.g. bribes or accounting rules (Troy et al., 2009), a matter which is hard to monitor or control in case of centralized control without a strong feedback loop.
To conclude, the extend of control is dependent on multiple factors. First, mutual dependence (Subramaniam & Watson, 2006) and ownership (Jaussaud & Schaaper, 2006) play a significant role. Next, organisation alignment will result in improved relations and control (Kownatzki et al., 2013). Last, for those companies yet acquired or being within a significantly different culture, control is proven to be more difficult to possess (e.g. Ferner (2000)).
Framework for fraud motivation
2.3 Incentives to engage in accounting fraud
Since the opportunity costs of fraud are substantial, committing fraud should reap substantial benefits to overcome these considered costs. This section will discuss the incentives to engage in fraud, where the focus will be largely on accounting fraud, covering most cases within this study. First, generic incentives such as personal gains and company performance will be discussed. Next, the link towards company subsidiaries will be discussed.
The first incentive for engaging in accounting fraud is the personal wealth of the fraudster. As mentioned in early papers of Caughlan an Schmidt (1984) and Jensen and Meckling (1976), pay for performance aligns the incentives between management and shareholders. Although incentives are aligned, in an extensive case study Erickson et al. (2003) find that firms engaging in accounting misstatements have higher proportions of stock based payments. In case the CFO has these substantial equity stakes (Jiang & Wang, 2010) or stock base options are in the money (Efendi et al., 2007), this likelihood will be even higher. That is, although the incentive is in place to align incentives with those of the shareholders, they may not always be aligned with respect to integrity.
Related to personal benefit is the egocentric mind-set of company management. Especially for companies with a low book to market value, management seems to overestimate its own capabilities in delivering future value over current book value. This overconfidence might lead to overbidding to be acquired companies, where management is confident it will reap more benefits than competition as show in a long horizon event study by Rau and Vermaelen (1998). This type of management
9 possesses aggressive behaviour, where management has the impression to be superior and confident to meet investor forecasts by any means, legal or illegal (Rezaee, 2005). This superior behaviour may as well be reflected in the way management invests. Firms which engage in misstatements often have a history of diverging strategies, where the core business is neglected and the main focus is shifted towards rapid growth by investment in acquisition instead of operations as proven by European and American case studies (Grant & Visconti, 2006). Since non synergetic growth is harder to add value, management seems to compensate in these cases by adjusting results instead of adding real value to the company.
Next to personal benefits or reputation, accounting fraud may as well be exercised to meet debt covenants or shareholder expectations. Empirical findings by Dechow et al. (1994) show that management of companies in financial distress or being subject to aggressive financing cost reductions are more likely to engage in accounting fraud. In case of no distress, companies with higher debt to equity ratios still have incentives to make deliberate accounting information choices to refrain violating debt covenants and stay off radar at financial institutions (Malmquist, 1990); (Jones, 2011).
Parallels to accounting fraud by subsidiary management can be drawn. The underlying power between subsidiary management and headquarter (HQ) management is often based on negotiation and the decision to centralise or decentralise in which factors such as cultural differences, regulations and history take part (Ferner, et al., 2004). In this perspective, the subsidiary can be seen as an alternative to a direct market transaction. In case of low centralisation and integration, the subsidiary management might be tempted to see the HQ management as shareholder and thus have comparable incentives as discussed above where stock ownership and disappointing subsidiary results motivate to smooth adjust earnings. By doing so, no specific attention has to be given to the subsidiary and its management, sustaining the subsidiary autonomy and increasing the chance of collecting a possible managerial bonus related to positive financial results at the subsidiary level (Roth & Nigh, 1992).
The acquisition of a subsidiary may as well create incentives to engage in accounting fraud. Case studies discussed by Jones (2011) show that the bidder has the incentive to increase company value in a cosmetic way in case it pays the acquisition with shares, diluting the amount of shares needed for the acquisition. On the other hand, management of the target company has the incentive to increase its (book) value as to achieve a higher bid. For acquiring firms the evidence on accounting fraud is weak, as for example rejected for cases in Greece (Koukamanos et al., 2005), but confirmed in the UK (Botsari & Meeks, 2008). Synergies displayed during acquisition may be difficult to
10 materialize in reality. Especially for companies with high book to market value and thus perceived potential, Schrand and Zechman (2012) reveal through fifty case studies that the acquired companies of acquiring companies tend to underperform due to a combination of CEO overconfidence and (too) high investor trust. Consequently, eager to show success, it can be argued these overconfident acquiring managers might engage in accounting misstatements to confirm their status of skilled acquirer.
To conclude, the review above shows that main motivations to engage in accounting fraud are personal gain, company performance or market expectation. For the case of subsidiaries, personal gains may trigger accounting fraud in case payment of subsidiary management depends on accounting performance. Next, subsidiary management might wish to remain independent, leading to possible accounting fraud to make sure no alarming results reach parental management. In case of new business or acquisitions, top management may either want to show their acquisition was worth the money or the target company may have intentionally changed its figures to achieve a higher price.
2.4 The opportunity to engage in accounting fraud
The incentive triggers the action. However, in case the opportunity does not exist, accounting fraud or any other (corporate) fraud cannot be committed. According to Jones (2011), the opportunity originates at a weak corporate governance including shareholders, regulators, auditors and banks. This section will discuss the opportunities for company management for committing accounting fraud. Since the section concerning the parent-subsidiary relationship did discuss the control environment within the company, this section will put more focus on the external parties.
Empirical findings by Bell and Carcello (2000) show that a weak control environment is seen as the first and most important opportunity for accounting fraud. With respect to external and supervisory aspects, a weaker control environment is reflected in less independent boards, more unitary structure between chairman and CEO or a not fully independent audit committee (Dechow et al., 1996), (Rezaee, 2005). Although no study touched upon the subsidiary context, a weak parental control environment may also enhance opportunities for the subsidiary to exercise fraudulent behaviour.
In theory, monitoring by external parties should correct any flaws in the internal control of the company, where stakeholders like share- and debt holders as well external parties such as auditors monitor the company in several ways and perspectives, thereby creating an indirect control framework. Rezaee (2005) argues that the role of internal and external auditors possesses the
11 monitoring of an adequate and effective internal control framework for the complete company. Stakeholders such as analysts, banks and majority or institutional shareholders might enhance monitoring and control. Empirical research by O’Brien and Blushan (1990) and Potter (1992), shows that firms with higher institutional ownership, being parties owning more than 5% of the share capital, have more and deeper analysis by analysts. Latham and Jacobs (2000) argue that such an environment with more parties following the company permits less errors since the likelihood that accounting fraud will be discovered increases, which is confirmed in a paired match study. Concerning subsidiaries, this ownership may both be reflected in the parental ownership to the subsidiary, where the parental management has more incentive to follow the subsidiary as well as the attention by external stakeholders in case the subsidiary actions can influence the overall results towards these stakeholders in a substantial way.
Although it is found that auditors seem to assure more effectiveness by withstanding the pressure to waive accounting fraud, knowing their reputation is at stake as well as having more tools to detect these errors (Palmrose, 1988), auditor independence remains a disputable topic. The value added from the auditor is to correct accounting fraud instead of stimulating or not detecting. Where big four auditors seems to be less engaged in accounting fraud compared to smaller firms, detection will remain difficult (Lennox & Pittman, 2010). Cases like Enron show this added value was not delivered, where auditors lost independence due to engagement potential in e.g. consultancy services. Next to the arguments mentioned above concerning independence and size, it might be the question whether the contribution of the subsidiary is material considering the size of the group. A smaller relevance may lead to no assessment of the subsidiary by the (external) auditor, as results or control deficiencies are not material. The same yields for outside investors and banks, who will put the focus on the largest chunks of the company. That is, accounting fraud by the subsidiary management may be more likely to be prevented in case there is more external involvement at parent level as well as the subsidiary, amplified by the relevance of the subsidiary within the parental results.
To conclude, the opportunity for accounting fraud is constitutes in the control environment of the company and is influenced by management as well as the degree of monitoring by shareholders, analysts and auditors. Concerning accounting fraud committed by subsidiary management, low ownership concentration at debt and equity holders as well as less relevance of subsidiary results on the parental level may create the opportunity for subsidiary management to commit accounting fraud without attracting any attention of outside stakeholders or auditors.
12
2.5 The company and subsidiary management attitude towards accounting fraud
Next to the motivation and opportunity of subsidiary management, accounting fraud is stimulated by the attitude of parental and subsidiary management towards accounting fraud. In this context, the perception about own behaviour concerning the accounting fraud as well as the perception on what is a desired situation plays a role.
Attitude starts with influence. Abernethy et al. (2014) prove empirically that employees within organisations with fewer shared goals are becoming more egoistic and tend to engage more often in accounting fraud. Even in case targets are shared and more persons support one another, Carpenter and Reimers (2005) argue that in case people close to management support the violation, managers are more likely to violate GAAP to meet earnings predictions. Furthermore, those managers with a positive attitude concerning violating GAAP will have more intentions to violate GAAP to meet earnings targets.
Next, according to Schrand and Zechman (2012), overconfident management tends to engage in accounting fraud in case of decreasing performance, confident of their personal skills improve underlying performance before accounting fraud is revealed. Analyses of CEO facial features by Jia et al. (2014) shows that CEO’s with more than average testosterone are more likely to engage in accounting fraud. Within the context of the parent subsidiary relationship, those parental boards having an aggressive attitude might not limit the subsidiary in engaging in accounting fraud. Furthermore, being eager to achieve their goals, they might even be less focussed on accounting fraud investigation than more integer and less aggressive managers (Dikolli et al. ,2013).
The prevention of accounting fraud is influenced by the integrity and attitude of the CFO. Stakeholders expect this the CFO position to guarantee an organisation which is in control and produces true and fair reports. As mentioned by Feng et al. (2011), the CFO has higher opportunity costs compared to the CEO, being end responsible for the financial figures and less dependent on pay for performance. Empirical research by Hennes et al. (2008) shows that 64% of the CFO’s being caught on accounting fraud leave the respective company voluntary or are being laid off. Besides share based compensation (Jiang & Wang, 2010), the attitude of the CEO as well as the company seems to pursue the CFO in making the decision to commit accounting fraud. Within this climate the CEO will link the future opportunities of the CFO to the desired actions to take, exerting the CFO to follow the CEO in the decision to engage in accounting fraud irrespective of the possible personal costs incurred later (Feng et al., 2011). This pressure may either exist within the outer limits of high growth or financial distress situations in which the attitude of the parent company and its board are less integer.
13 Summarizing the analysis above, less shared goals, more aggressive strategy and testosterone as well as overconfidence by company management supported by low integrity by a CFO may foster a positive attitude towards accounting fraud. Within the context of this research, less shared goals to achieve between the subsidiary and parent may foster fraud, where egocentric management at the subsidiary may get the opportunity to strive its own goals. Next, a masculine and testosterone driven management reflected in an aggressive strategy may create the attitude more fraud reluctance at the parent and subsidiary, decreasing integrity and thereby increasing the likelihood of accounting fraud.
14
3. Research method, sample selection & hypothesis development
This section will discuss the research method, sample selection and hypothesis development applied to investigate the research question. First the research method will be explained. Next, the sample selection of the selected fraudulent companies and peer companies will be discussed. Last, hypothesis resulting from the literature review as well as their underlying parameters to be assessed will be addressed.
3.1 Research method
The research method chosen for the current research is a sequential)case study. As mentioned by Yin (2014), in general case studies enable the research of topics by analysis of observations in a descriptive and explanatory way. That is, unlike empirical studies, a case study may provide answers towards which factors were underlying the observations in both a qualitative and quantitative way whereas empirical studies may only detect the relations between several factors. This study, researching the qualitative factors underlying accounting fraud at subsidiaries, is for that reason best supported by a sequential case study.
Next, the amount of observations leads to favouring a sequential case study as well. Although it is likely that a large amount of companies do cope with these issues, relatively few were reported in the media. Within these cases, a relative small set has financial, analytical and historical data available to assess the factors causing or underlying the accounting fraud, making it difficult to design an empirical setting.
To provide evidence from multiple sources and provide context with companies in which no accounting fraud was detected, a sequential case study is applied within this thesis. Sequential case studies are often used within epidemical research or medical research (e.g. Llewellyn et al., 1998). Sequential case studies assess cases which have a common shared factor, irrespective of the cause to be researched. With respect to this thesis, the common factor is fraud at corporations involving the subsidiary, either before the subsidiary was acquired or during governance under parental regime. Next, a peer group of companies resembling the initial sample are selected to end up in ‘pairs’ of comparable companies. For these companies and the subsidiary of the fraudulent company, deviating factors between the fraudulent en non fraudulent companies are identified. Furthermore, factors in which the fraudulent companies deviate from each other will be identified
Since the sample of nine companies is small, results are not generalizable. However, due to in depth research of the respective cases and peer companies, outcomes may be comparable between the selected cases taken (Yin, 2014). The outcome of the research will only be case related resulting in
15 exploratory findings which cannot be generalized. Furthermore, although peer companies were researched on the commitment of fraud, it cannot be guaranteed no (non-revealed) fraud took place at the peer group during the investigated time frame.
3.2 Sample selection
This section will explain the sample selection. First, criteria for fraudulent companies to be included into the research are elaborated on. Next, the selection of peer companies will be explained.
3.2.1 Fraudulent companies
Eight companies were identified in which accounting fraud was committed concerning subsidiaries for which appropriate data was available. To enlarge the sample, one case (Philips) was added in which corruption played a role, although the underlying motivation of the Polish management, opportunity and attitude as well as control environment may make the case comparable to the other accounting fraud cases. The first aspect concerning the relevance a case is that the respective company was involved in fraudulent accounting activities. As explained by Jones (2011), fraud or accounting fraud is any activity or accounting interpretation not being within the regulatory framework. The regulatory framework applied is that of the parent. That is, even if the (acquired) subsidiary committed accounting fraud or bribery which is allowed within its own jurisdiction, it may be marked as fraudulent in case the parents’ jurisdiction does not allow the specific accounting treatment or action. Next, the accounting fraud has to have a relationship with the subsidiary management of the respective company. Sample selection took place by researching the internet on keywords such as ‘fraud subsidiary’, ‘accounting misstatement subsidiary’ or any related term. After initial selection, the fraud of the companies involved have been researched by searching for annual reports, restatements and newspaper or internet articles describing the respective cases. For those companies selected, fraud was detected and the involved personnel or legal entity has been sanctioned. Furthermore, for six out of nine, parent financial statements had to be revised. For the other three, fraud took place before the parent integrated the annual figures (Catterpillar, HP) or bribes were detected resulting in lawsuits (Philips).
16 The selected cases are as follows:
A quick analysis shows that for three cases the accounting fraud took place at the intersection of the acquisition of the respective subsidiary (Catterpillar, HP and Unit4), possibly revealing a first incentive for fraud. The case of Ahold has been scoped down to the US Foodservice case which covers a substantial part of the overall scandal. Furthermore, the fraud was committed by the subsidiary, making the case more relevant for this study as compared to the other Ahold cases which cover consolidation subjects on holding level.
3.2.2 Peer sample selection
The companies stated above have a large variety of characteristics, making comparability difficult. In correspondence with the research method of the sequential case, peers of the respective fraudulent companies have been selected to place all cases into their specific context.
The matching process of peer companies started with research on websites of financial newswires such as Morningstart, Yahoo Finance and Financial Times for peer companies. Next, the dominant (SIC coded) industry is compared between the fraud case company and the peer company. In case more than one company match after this step, potential peers are assessed consecutively on the following criteria:
1. Revenue per year in currency of the fraudulent company
2. Number of full time equivalent (FTE) employed
3. Continent of headquarters
4. Continents on which the company has business activities
Applying these criteria results in the following results for the selected companies (see next page):
Case Year of fraud revealed Kind of fraud Before/after acquisition
Ahold 2002 Purchase bonusses Before & After
Caterpillar 2013 Revenue and cost accounting Before
DE Masterblenders 2012 Discounts , invoice shifting, inventory valuation After
Getronics 2006 Cost accounting After
HP 2011 Revenue recognition, growth and gross margin Before
Philips 2011 Bribes n.r.
Unit 4 2011 Incomplete transaction history, revenue recognition Before
Imtech 2012 Revenu recognition and cross financing n.r.
Symmetry Medical 2007 Receivables, understating expenses, overstating assets After Table 1: selection of fraudulent cases . For data sources, refer to the appendix
17 N a m e # E m p lo y e e s R e v e n u e i n m ln F X H Q E u ro p e N o rt h A m e ri c a S o u t A m e ri c a A si a A fr ic a O ce a n aD o m in a n t se ct o r S IC Y e a r d a ta A h o ld 2 3 1 .0 0 0 5 2 .0 0 0 € N e th e rl a n d s x x x G ro ce ry s to re s 5 4 1 1 2 0 0 4 T e sc o 2 2 0 .0 0 0 4 8 .0 0 0 € U K x x G ro ce ry s to re s 5 4 1 1 2 0 0 4 C a te rp il la r 1 1 8 .5 0 0 5 5 .5 6 5 $ U S A x x x x x x C o n st ru ct io n m a ch in e ry a n d e q u ip m e n t 3 5 3 1 2 0 1 3 K o m a ts u 4 7 .2 0 8 1 8 .8 4 0 $ Ja p a n x x x x x x M in in g M a ch in e ry a n d E q u ip m e n t 3 5 3 2 2 0 1 3 D E M a st e rb le n d e rs 2 .2 0 0 2 6 1 € N e th e rl a n d s x x x M is ce ll a n e o u s F o o d P re p a ra ti o n s & K in d re d P r S e ct 2 0 8 4 2 0 1 3 M ik o N V 7 5 8 1 4 4 € B e lg iu m x x R o a st e d c o ff e e 2 0 9 5 2 0 1 3 G e tr o n ic s * 1 1 9 .8 3 3 2 .5 0 0 € N e th e rl a n d s x x E le ct ro n ic c o m p o n e n ts , n o t e ls e w h e re s p e ci fi e d 3 6 7 9 2 0 0 6 Lo g ic a C M G * 2 3 8 .7 4 0 3 .0 7 3 € U K x x x x x x M a n a g e m e n t co n su lt in g s e rv ic e s 8 7 4 2 2 0 0 6 H P 3 4 9 .6 0 0 1 2 7 .2 4 5 $ U S A x x x x x x C o m p u te r p e ri p h e ra l e q u ip m e n t, n o t e ls e w h e re s p e ci fi e d 3 5 7 7 2 0 1 1 IB M 4 3 3 .3 6 3 1 0 6 .9 1 6 $ U S A x x x x x x C o m p u te r in te g ra te d s y st e m s d e si g n 7 3 7 3 2 0 1 1 P h il ip s 1 2 5 .0 0 0 2 2 .5 7 9 € N e th e rl a n d s x x x x x x E le ct ro n ic s (d o m in a n t) 3 x x 2 0 1 1 T o sh ib a 2 0 0 .0 0 0 5 8 .0 1 5 € Ja p a n x x x x x E le ct ro n ic s (d o m in a n t) 3 6 x x 2 0 1 1 U n it 4 4 .0 4 8 4 5 5 € N e th e rl a n d s x x x C o m p u te r p ro g ra m m in g s e rv ic e s 7 3 7 1 2 0 1 1 E x a ct 1 .6 7 9 2 1 3 € N e th e rl a n d s x x x P re p a ck a g e d s o ft w a re 7 3 7 2 2 0 1 1 Im te ch 2 9 .4 7 3 3 .5 6 3 € N e th e rl a n d s x E n g in e e ri n g s e rv ic e s 8 7 1 1 2 0 1 1 A rc a d is 1 9 .5 0 0 1 .8 7 8 € N e th e rl a n d s x x x x x E n g in e e ri n g s e rv ic e s 8 7 1 1 2 0 1 2 S y m m e tr y M e d ic a l 2 .9 7 9 2 9 1 $ U S A x x x S u rg ic a l a n d M e d ic a l In st ru m e n ts a n d A p p a ra tu s 3 8 4 1 2 0 0 7 IC U m e d ic a l 2 .2 3 9 3 1 6 $ U S A x x x x x S u rg ic a l a n d M e d ic a l In st ru m e n ts a n d A p p a ra tu s 3 8 4 1 2 0 1 2 * 2 a cq u ir e d b y C G I in 2 0 1 2 * 1 a cq u ir e d b y K P N i n 2 0 0 8 R e g io n s a ct iv e
18 For all companies, the industry was comparable. However, for Miko, Toshiba and IBM, the scope of business was partly larger or different from the fraudulent peer, although main business activity was shared. Revenue differs significantly between fraudulent companies and their peers, though being maximal three times that of the fraudulent company in case of Catterpillar. Last, for the matched peers Philips-Toshiba and Catterpillar-Komatsu, the head offices are on different continents. However, since their respective business is of global nature the companies are comparable. Furthermore, the governance of Japanese versus American/European governance may be valuable for the full case study.
3.3 Hypothesis development and parameter selection
Within the literature review, the control environment of subsidiaries as well the incentive, opportunity and attitude of subsidiary and parental management towards accounting fraud have been discussed. This section will develop the hypotheses assessing the literature within the context of the fraud committed for the respective case studies.
Since the hypotheses cannot be assessed directly, a framework including indirect indicators supporting the hypothesis is created. Subsidiaries, parent companies and peer companies will be assessed on this framework. This section will discuss the underlying selected indicators supporting the hypotheses concerning the control of subsidiaries, the incentive and opportunity of subsidiary management to engage in accounting fraud as well as the attitude of the respective management of the subsidiary and the parent.
3.3.1 Assessment of the control environment of subsidiaries
The first factor on which the selected cases will be assessed is their control environment. As discussed, this control environment can be split up in the formal relationship between parent and subsidiary, dependency between both of them, the alignment of the organisational setup and the influences of language and culture.
In line with research by Jausaud and Schaaper (2006) concerning ownership and findings by Subramaniam and Watson (2006) concerning mutual dependency, the following hypothesis is stated to assess the relation between control and mutual dependency in ownership as well as other recourse dependency:
H1: it is more difficult to control companies with lower dependency in equity stakes or shared
19 This hypothesis is analysed by researching the following parameters:
1. Ownership stakes
2. Ownership during accounting fraud
3. Main business activity of the parent and subsidiary
Selected companies are assessed on the ownership stakes of the parent in the subsidiary since less ownership can indicate control to be shared by other parties resulting in lower control. Higher stakes indicate more dependency and thus more likelihood of control (Subramaniam & Watson, 2006). The second indicator to research the hypothesis is the assessment whether the subsidiary was owned by the parent at the time the fraud was committed. That is, in case the subsidiary was an independent company at time of fraud commitment, any parental control restrictions are ineffective. Last, control by shared business is assessed by looking at the main business of both parent ad subsidiary.
Following the literature review on the aspect of strategy and mutual understanding, where transparency and aligned strategy enhance transparency, trust and less incentive to deviate from the parental direction(Kownatzki et al., 2013), the following hypothesis can be stated within the context of this study:
H2: there is a positive relation between an aligned organisational setup and control, resulting in a
lower likelihood for accounting fraud
Assessing the second hypothesis is conducted by analysing the setup, control mechanism and existence of an internal audit department. To analyse the organisational setup, the structure of the complete company as well as the role of the subsidiary within this structure as stated in the annual report of the parent and peer company is reviewed. In case of regional steering, a less aligned strategy is assumed in case no functional link between subsidiary and parent can be made allowing the subsidiary to make decisions independent within its geographical boundaries.
Next, the mechanisms of parental management to control the subsidiary are assessed. Since scorecards are not publicly available, the differences between subsidiaries not yet acquired during the accounting fraud and national versus functional steered locations are researched.
The last parameter to research the second hypothesis is the existence of an internal audit department. Existence of such a department may indicate priority by parent manager to understand and control the full company where the internal audit function acts as a watchdog for accounting fraud or irregularities within the company and its subsidiaries and does prevent material misstatements (Lin et al., 2011).
20 The last factor described in the literature review concerning the control environment of the subsidiary is the cultural and lingual barrier which may exist for foreign entities in multinational companies (Ferner, 2000). The assessment of such difficulties will be done by reviewing the selected cases on the following hypothesis:
H3: foreign subsidiaries are more difficult to control than domestic subsidiaries, resulting in a higher
likelihood for accounting fraud
H3 will be researched by analysing the location of the head office of the parent company compared to the location of the subsidiary and likelihood of any existing linguistic or distance barriers. These factors are derived from the research of Björkman and Peikkari (2009) arguing weaker and more formal control is in place in case there is a significant discrepancy between the parental and subsidiary culture and language. Since cultural aspects are difficult to derive from available data, the country culture, distance between headoffice and subsidiary (intercontinental) and linguistically discrepancies are used as proxies for cultural aspects.
3.3.2 Assessment of the incentive to possess fraudulent behaviour
Concerning the incentive of management to engage in accounting fraud, literature by amongst others Jiang and Wang (2010), Malmquist (2010) and Jones (2011) discussed personal gain, the performance of the company, market expectations and the wish or the subsidiary to remain independent may be an additional incentives. Next, those companies recently acquired may have overpromised financial performance to end up with a better acquisition price (Botsari & Meeks, 2008).
Concerning personal gain, the stock based incentives are an incentive for accounting fraud, where manipulation may result in better pay off as shown by Erickson et al. (2003). Assuming subsidiary management is vulnerable to the same treats and parental management is eager to see good results to gain their stock based bonus, the following hypothesis is stated:
H4: There is a positive relation between stock based payment of subsidiary and parental
management and the likelihood of accounting fraud at the subsidiary.
Stock based payment for the parental and subsidiary board is assed. Especially for those companies recently acquired, ownership by subsidiary executive management is assessed, where higher stakes before sale might uncover a bright incentive for management to engage in accounting fraud for own sake
21 The wish to be independent may also create an incentive to engage in accounting fraud. To remain independent, it is likely the subsidiary wants to attract as less attention as possible, showing constant good performance. This results in the following hypothesis:
H5: There is a positive relation between the past good performance of the subsidiary and the
likelihood subsidiary management engages in accounting fraud .
This hypothesis mainly researches the involvement in accounting fraud for a subsidiary to remain independent. For that reason, only for those subsidiaries being part of the parent company at time of fraud commitment will be assessed. For these companies, the performance ratio of the EBITDA over assets and EBITDA over sales are used as a performance benchmark in case information is available at this level.
The last factor to assess the subsidiary management incentive to engage in accounting fraud is the recent acquisition of the respective subsidiary. Research by Jones (2011) as well as Schrand and Zechman (2012) shows that either due to overpromising of the target or overconfidence of the bidder, performance issues can take place. The management of the selling party being eager to gain a substantial takeover price with a bidder being blind due to overconfidence creates a perfect incentive, which results in the following hypothesis:
H6: There is a positive relation between subsidiary accounting fraud and the recent acquisition of this
subsidiary
Assessing this hypothesis involves the assessment whether the respective subsidiary has been acquired as well as the relation between the acquisition and the accounting fraud. That is, in case the subsidiary was recently acquired, subsidiary management might engage in accounting fraud after acquisition to show its perceived value or to remain independent (Schrand & Zechman, 2012). On the other hand, a to be acquired party might be eager to increase its perceived value, thereby increasing the possible sales price (Botsari & Meeks, 2008). For peer companies, this variable cannot be assessed. For that reason, the most recent acquisition activity of peer companies will be researched to provide comparability into the acquisition behavior peers compared to the selected fraudulent companies.
3.3.3 Assessment of the opportunity to engage in accounting fraud
The third aspect discussed in the literature review is the opportunity to engage in accounting fraud. As discussed, factors creating the opportunity are the control environment, lack of integration by the subsidiary and lower external involvement by either auditors or other stakeholders. Since the control
22 environment of the parent company and subsidiary are discussed above, the hypotheses below will cover the areas concerning external involvement.
The first hypothesis will discuss the concentration of the ownership by both the parent and subsidiary. That is, in case a (to be acquired) subsidiary has a high concentration of ownership by external holders, analyst will follow the parent or to be acquired subsidiary, reducing the opportunity to engage in accounting fraud without being caught (Lattham & Jacobs, 2000). This results in the following hypothesis:
H7: There is a negative relation between ownership concentration by external parties and the opportunity to engage in accounting fraud
To test this hypothesis, the ownership structure of the parent, the ownership of the subsidiary at time of fraud conduction as well as the size of the subsidiary relative to the parent are assessed. Since all parent companies within the sample are publicly listed, ownership will be relatively dispersed. However, institutional a major shareholding owners may still exist resulting in significant external stakes. For that reason, large holdings (>5% of total share value) reported in the annual reports are researched. Since most countries oblige such disclosure by law (e.g. form 13F enforced by US SEC (SEC, 2015) or Dutch AFM (AFM, 2015)), this source can be seen as relatively objective. Those companies not stating substantial ownership are researched via Yahoo Finance or other financial websites. Next to institutional and majority ownership, a larger concentration of debt holders may indicate more stringent external control. For that reason, the long term debt to equity ratio is assessed for the parent, subsidiaries and peers, where it is assumed that in case the long term debt ratio is high, debt holders have incentive to control the company by either disclosure or covenants and are committed to the company for a longer period in time. Concerning the ownership of the subsidiary, debt holders as well as major shareholdings are assessed as well where information was available.
In case the subsidiary was already under parental regime at time of the initiation the accounting fraud, external parties such as auditors or institutional shareholders may only put their attention towards the subsidiary in case its performance is material. Non-material subsidiaries may not reach the span of attention of these parties, opening up opportunities for accounting fraud. This results in the following hypothesis:
H8: There is a positive relation between the relevance of its subsidiary for the parent and the likelihood that the accounting fraud will be detected by an external party
23 The relative size of the subsidiary compared to the parent is indicated in terms of revenue and number of FTE. This metric indicates in case the subsidiary is of substantial size compared to the parent, it will attract more external attention increasing the likelihood of the fraud being detected by external parties.
3.3.4 Assessment of the attitude of subsidiary and parental management
The last factor assessed in the literature review is the attitude of management or any other stakeholders towards accounting fraud or fraud in general. As found in existing literature, aggressive strategies, testosterone, overconfidence and low integrity may foster a positive attitude towards accounting fraud in a wish to see better financial results or perceive personal status or benefits. Concerning the aggressive attitude of management, the first hypothesis was as follows:
H9: there is a positive relationship between an aggressive parental strategy and the likelihood to
engage in accounting fraud
This hypothesis tries to assess factors such as overconfidence and testosterone of parental management, accepting accounting fraud or bribery to achieve their goals. Since testosterone, assessed in the paper of Jia et al. (2014) by looking at facial characteristics, is difficult to assess, this research will use a proxy being aggressive goals of the parent company, to be assessed by researching newspaper articles and annual report. As stated in the literature review both the parent companies and their peers are assessed on whether the strategy involves (aggressive) growth as well as a possible going concern issue, triggering a panicking attitude.
The last hypothesis assesses the attitude of the subsidiary management. That is, irrespective of the parental management, the (to be acquired) subsidiaries may have divergent goals which make subsidiary management reluctant to fraud prevention by means to achieve these goals. This results in the following hypothesis:
H10: : there is a positive relationship between aggressive goals for subsidiaries and the likelihood to
engage in accounting fraud
To assess this hypothesis, internet resources, annual reports and previous ownership data are assessed to see whether the subsidiary management only has to commit to internal goals or may have a ‘hidden agenda’ such as the sale of the company or independency.
24
4. Analysis and hypothesis assessment of selected cases
This section discusses the analysis of the hypotheses, providing an answer to the research question investigating what factors cause a company to be subject to accounting fraud in a subsidiary. As stated in the research method, data of the parental company, subsidiary and peer for all nine cases will be involved. In line with the previously determined framework, first the control concerning subsidiaries will be analyzed. Next, the subsidiary managements’ incentive to engage in (accounting) fraud will be reviewed followed by the opportunity by subsidiary management to commit accounting fraud. Last, the attitude of management within the case subset will be analyzed. Finally, a summary of the findings will be provided.
4.1. Assessment of the control environment within subsidiaries
This first section will analyse the selected pairs on the control of the subsidiaries. This assessment will be done by researching hypothesis 1 to 3 focussing on dependency and shared business, aligned strategy and control and the location of and language spoken within the subsidiary as compared to the parental company.
Concerning the first hypothesis, the following data has been retrieved:
Analysing the results of the data retrieval shows that the vast majority of the fraudulent subsidiaries were eventually 100% owned by the parent. However, at the time of the fraud conduction, three subsidiaries were still independent companies, thus not being in control of any parent. Analysing the main business, it can be seen even those companies which were acquired at a later stage were involved in business activities comparable to the parent. The only exception is Ahold, where US Foodervice is a business to business focussed market whereas Ahold’s main business is on the business to customer segment. Summarizing, this hypothesis cannot be accepted since the equity stakes of most parent in the subsidiaries were high and the main business of most cases was aligned, not confirming results by Jaussaud & Schaaper concerning ownership. Nevertheless, at time of fraud commitment, three cases were yet to be acquired, which influences any control over this company
Table 3: Data concerning the ownership and strategy of the company and subsidiary. For data sources, refer to the appendix Owned
during fraud
Company Parent Peer Parent Parent Subsidiary
Ahold 100% Retail 100%, property 50% Y Grocery stores Food service
Caterpillar 100% Majority 100% N Construction machinery and equipment Construction machinery and equipment DE Masterblenders 100% New subsidiaries <100% Y Wines, brandy and brandy spirits Coffee
Getronics 100% Only 100% holdings Y Computer Integrated Systems Design Computer Integrated Systems Design HP 87% Majority 100% N Software company Computer consultancy activities Philips 100% Large part join venture Y Electronics Electronics
Unit 4 100% Majority 100% N Business software Business software Imtech 100% Majority 100% Y Engineering services Engineering services
Symmetry Medical 100% Majority 100% Y Surgical and Medical Instruments Surgical and Medical Instruments
25 by any parent policies or indicates that the due diligence by the parent did not turn out to be firm enough.
Analysing hypothesis 2 concerning organisational setup as reflected in table 4, the cases of Caterpillar, HP and Unit4 show that, irrespective of any organisational setup at the parent, no alignment could have taken place due to the independence of the subsidiary at time of fraud commitment. The case of Getronics indicates accounting fraud partly due to a divergent organisational setup with different regions in which management had significant communication difficulty (Shouten, 2007), which is similar to the case of Ahold (Govind & Dutta, 2006). This lack of communication due to different regions and scattered control is also reflected within the case of DE Masterblenders and Imtech where local management was involved in several fraudulent activities encompassing several segments and no triggered action by either lower or top management (VEB, 2012); (Imtech, 2013). Concerning Symmetry Medical and Philips, annual reports show that these companies did have a segmented setup, though fraud still occurred. Analysing the respective peer companies, it can be concluded that except for Getronics and Imtech, comparable structures are in place. That is, in case the internal control system and steering is comparable and the peers are free of substantial fraud, the pure setup may not decrease the likelihood of accounting fraud. The difference lies in the three fraudulent cases involving acquisitions where the parent company steering would not yet be effective. The setup issues are also reflected within the steering of the companies. For the non-acquired part, most steering took place on a national level either within the functional domain or the local results.
Concerning the internal audit departments, it can be concluded that for most cases the mere existence of an internal audit department did not decrease the likelihood of the accounting fraud being discovered. Looking into the exceptions shows that Unit4, not having a group internal audit, relies on local audits and procedures according to its 2012 annual report, whereas Imtech states that
Control mechanism subsidiary
Company Parent Peer Parent Parent Peer
Ahold Regional Regional Board of directors, no direct control Y Y
Caterpillar Segments, different companies Segments None, different companies Y Y
DE Masterblenders Regional Regional n.n. Y N
Getronics Regional Segments Regional setup results Y Y
HP Segments, different companies Segments None, different companies Y Y
Philips Segments, sales results per region Segments National sales results Y Y
Unit 4 Regional, different companies Regional None, different companies N Y
Imtech Regional Segments Regional results N Y
Symmetry Medical Segments Segments Product segment steering Y Y, since 2010
Aligned strategy Internal Audit department
26 ‘there is no need for such a department for the time being, because of adequate controls and management systems and in view of the wide scope of KPMG’s audit’ (Imtech Annual report 2011). In case of Symmetry Medical, the effectiveness of the internal audit function was questioned by the Securities and Exchange Commission (SEC), where the fraud included a 30% overestimation of receivables at its UK subsidiary in 2006 (SEC., 2012). Concerning the peer companies, an internal audit department seems to be common business, not highlighting substantial differences with their fraudulent counterparts though questioning the effectiveness of the cases where accounting fraud took place.
Based on the case descriptions above, H2 may be rejected, where many fraudulent companies where either set up in an indirect or regional matter or could not possess any structure or strategy due as the subsidiary was still to be acquired. However, for cases such as Ahold, Getronics, Imtech and DE Masterblenders, the research by Subramaniam and Watson concerning control deficiencies in case of non-aligned strategies are supported, since the management of subsidiary and parent was not aligned. The inclusion of an internal audit department did not prevent the fraud from occurring in most cases, although for one third the function was not in place or assessed to be highly ineffective. Concerning hypothesis 3 stating foreign subsidiaries are more difficult to control, the following data has been retrieved:
The table above shows that all subsidiaries were located within a different country than the parental head office. For the researched peer companies, neither of them had all business activities and subsidiaries within one country, indicating that the location should not matter per se. Zooming in into the mother tongue spoken at head offices and subsidiaries, especially the cases of Caterpillar, Getronics and Unit4 are interesting. These entities involve acquired parties in different countries where severe language barriers can be assumed (Germanic versus respectively Mandarin, Latin or Slavic languages). Legacy of old business culture as well as language barriers can play a role in understanding the parental control or subsidiary business, especially in case this business has not been set up by the parent. Furthermore, the language barrier may refrain lower personnel from
27 signalling fraud towards the parental head office. Concerning DE Masterblenders, a physical and language barrier might explain why the fraud went unnoticed for five years (VEB, 2012). Although the same might yield for Philips, it is unlikely that with a long polish history stemming from 1921 (Philips, 2014), the language might have been an opportunity to engage in fraud. Concerning the other cases of Ahold, HP, Imtech and Symmetry Medical, either the mother tongue was English or the parties are not assumed to have any language communication difficulties, although for Ahold and Symmetry Medical there is physical distance, though comparable Anglo-Saxon business attitudes. Taking into consideration the peer companies, most do have a set up comparable to their fraudulent counterparts which makes the findings above not generalizable. Concluding, in line with Harzing and Feely (2008), the subsidiary management was given some more freedom in several cases. The location, distance, culture and language spoken may have an effect on the ease of control over the respective subsidiary, especially in case of non-integrated subsidiaries (Björkman & Piekkari, 2009), although the peer company review shows that even with comparable structures, accounting fraud may not occur. This leads to weak acceptance of hypothesis three within the case of the recently acquired parties.
To conclude on the control of the subsidiary, the analysis above showed that ownership and business segments of the parent company and its subsidiaries did not seem to make the difference, nor did they differentiate from their peers in this respect. However, the fact that the subsidiary was recently acquired will increase the chance of accounting fraud being committed. Next, less integration of either the (recently) acquired entity may affect the effectiveness of any internal audit to reveal the accounting fraud, though not per se triggering it. Last, for the selected cases, in some instances clear language or physical barriers existed, decreasing the control environment effectiveness.
4.2 Assessment of the subsidiary management incentives to engage in accounting fraud
This section will analyse the data concerning the incentives of subsidiary management to engage in accounting fraud. The hypotheses regarding stock/incentive based payments, past performance of the subsidiary and a recent acquisition will be assessed.
First, assessing the hypothesis concerning stock or incentive based payments, the following data has been gathered from the annual reports of the respective companies, their (independent) subsidiaries and their peers:
